While during the first assessment, we could take an opinion of what we perceive from the organization with the chance of not being part of certain characteristics, in this case it’s going to be mainly about how well IT has been doing their due diligence to provide effectiveness in the execution. There’s no success in service delivery if you don’t understand your users.

Stop all this weeping, swallow your pride
You will not die, it’s not poison
Bob Dylan – Tombstone Blues (1965)

Let’s take a closer look about this IT self-assessment around users understanding with some simple questions.

2. Users Understanding Self-Assessment

2.1 Does your company have definitions around Personas and User Journeys?

• No one in the company can define what’s the purpose around Personas and/or user journeys.
• There are user profile definitions in some of our tools but mainly to define their company role and responsibilities.
• The user profiles definitions are inconsistent among different departments and tools available.

• There are personas definitions in technology but only to define what type of device the user receives.
• HR conducts user surveys to understand more about user preferences, behaviors, etc. but no clear action items come out of it.
• There was a personas definition completed in the past, but there were no implementations, processes in place or updates about these definitions.

• There is a consistent personas definition and user journeys across different areas, including IT, HR, security, etc. which serve to provide personalized experiences.
• The personas and user journeys definitions consider technology components, HR and overall user sentiment / user experience.
• There are tools in place to automatically measure the end user experience and personas adoption, which can be used to update definitions.

I’m going to dedicate an entire article to talk around personas and user journeys, so we’ll go over these themes in high level for now. Personas is a topic that is well-known in several organizations, but not very consistently. In some cases, they use that term just to define a VDI profile or an attribute to what type of device you’ll get. User journeys are not that common just yet, since it has been associated with an HR function for climate surveys in a company or how developers perceive the user interaction with their UX.

Personas is a critical aspect in any organization, a solid persona definition should include the technical side (computing and digital requirements; alignment with IT services; collaboration and productivity tools required; etc.) as well as the human side (user sentiment and user experience, which can include current issues, challenges, etc.). With the human side feeding input into the technology definitions, critical aspect to gain true technology enablement.

User/Employees journeys represents a timeline showing the details of how the user feels, managing different interactions and performing their normal activities. Understanding this allows you to perceive several important factors that shape the overall impression and perspective employees have. Valuing their motivations and challenges will help you to create an engaging and better user experience.

Example of a user journey definition in a sales rep persona

If your company is looking to improve end-user experience, provide truly personalized experiences in the workplace and enrich the overall IT perception, personas and user journeys are decisive topics. If all the action items to keep employees happy have been just buying new laptops and smartphones without these definitions, you should stop right now and re-assess your strategy.

2.2 Do people collaborate easily across the organization? Do they feel encouraged to share talent and ideas?

• Organization is divided by clear siloes, collaboration is minimum.
• Allocating time and effort to collaborate with other areas is very challenging since the company does not encourage this behavior.
• Little is known between different departments activities and planning to understand when collaboration will be valuable.

• Collaboration is based on situation basis, only when is required.
• Collaboration is accomplished mainly because of personal relationships.
• Managers might encourage collaboration if other departments have required skills.

• Corporate goals and objectives determine the collaboration channels.
• Cross-functional teams are rewarded for driving business value.
• Projects require documenting delivery methods for future teams to reference.

It is very challenging to measure collaboration and therefore also encouraging it if it’s not developed organically, this usually relates to more into a cultural aspect in organizations. You cannot force collaboration, that’s just a fact. We, as technology enablers, need to ensure that the proper tools and technology are in place for a natural integration.

There are some important aspects an organization needs to have for collaboration to occur naturally, one being as mentioned in my previous article about inclusion and diversity. You will have a hard time having people working and expanding in a healthy manner if the company is having challenges on these areas.

Before setting this topic aside and blame it on cultural aspects of the company, it is our responsibility in IT to understand users’ challenges, limitations and needs that go beyond what an organizational role is telling us. Enter personas and user journeys. Do the research and the assessments so your IT area can act as a true facilitator and enabler.

There are several platforms available in the enterprise that are indeed technology enablers for collaboration, having these in place will make easier to solve some of the cultural problems. The two that seem are going to be the leaders enterprise-wide are Slack and Microsoft Teams. They have similar components and functionalities (Teams being newer in the market), same as several other tools available. No matter the platform you decide on, the main modules present that you should look for are:

• Channels: Single place for teams to share messages, tools and files. One of the main aspects is that are organized by threads, to maintain an organized view about these conversations. Usually structured by teams or projects within an organization. Channels can be public or private.
• Messaging and conferencing: Features to allow 1:1 or 1:N instant messaging, audio and/or video conferencing within the app.
• Content management and collaboration: Centralized repository to share files and content between users. Besides sharing them within channels and messages, versioning is key feature required for content in these platforms.
• 3^rd party integration: Having a solution that is locked to only use features offered by the tool itself should be the first sign of a platform you shouldn’t be using. Most of the solid platforms today have thousands of integrations possible with 3^rd parties to encourage collaboration, for example: Dropbox, Box, Office 365, G Suite, Azure, AWS, Zoom, WebEx, etc.
• Mobile availability: Do we even need to say this is a requirement? If the platform is no mobile friendly, it should be discarded right away.
• Other features to consider (some of these are available with 3^rd party integrations): Wikis; polls creation and management; high quality video generation and streaming; creating and using bots (AI and automation key aspects); project management features; etc.

2.3 Are there clear content management policies, processes and tools in place?

• There are no formal policies, processes or tools. Each person / group handles it differently.
• Because of lack of policies, shadow IT is common. Areas or projects use preferred platforms to manage content and/or sharing.
• The content management is highly restricted and very limited possibilities available.

• There are guidelines for content management with some tools in place.
• Content sharing is possible inside and outside the organization with some limitations provided within the tools.
• Guidelines in place, but the governance process is inexistent.

• Clear processes and tools defined for content management and collaboration. Users can access these policies easily as well as getting the tools they need.
• Flexible model in place to support content and collaboration with other companies and different platforms if needed.
• There are AI and automation processes that provide analytics around content identification to optimize collaboration and end-user experience.

Content management tends to be chaotic in most organizations. You need to be flexible by providing different tools and ways for people to manage content and collaborate; and at the same time you need a balance to maintain simplicity, governance and a compliant environment.

A content services strategy will be required, assessing your current environment and the content use cases associated to business requirements. Do your research and follow the successes obtained in the past or current about tools and/or processes around content to build your strategy, you would want to build something that the users can feel relatable.

For example, if a considerable group (doesn’t have to be big, but representative and/or able to influence on others) has a good experience with certain platform for content (i.e. OneDrive, Box, SharePoint, etc.) and the platform suits the requirements you have based on the research, start enabling these functionalities in a planned manner.

Consider about the company’s culture to make the adoption simpler, aligning as well with IT and the business strategy. Market the tools in a proper way, generate buzz, build campaigns, interact with other departments to make sure you can be successful in this deployment. After all, we must remember in IT you are always building and maintaining a brand.  

If you are offering new technology and better ways to collaborate, don’t be afraid to challenge some content management practices.

2.4 Do employees feel encouraged to provide feedback and suggestions?

• Organization culture is adverse on providing feedback.
• Employees only provide feedback about improvements to their managers in annual reviews, and most of the suggestions never go anywhere else.
• People take credit for another people’s work.

• The company has a process and website for submitting ideas
• C-level executives are talking about driving enterprise-wide behaviors.
• New ideas are mainly around implementing new technologies.

• Structures, channels, tools and processes are in place to support feedback and suggestions.
• Processes to provide and obtain feedback are done timely and in an effective manner.
• Leadership recognizes that technology is a key innovation enabler and empowers IT to pursue initiatives to that end.

Providing and receiving meaningful feedback and suggestions is a powerful asset in most organizations. “Meaningful” being the keyword, which it’s not only tied to the employee providing constructive suggestions, but also in the way a leader or areas are requesting this feedback.

We need to avoid consider these as bureaucratic processes. If you are providing annual climate surveys with endless and non-engaging questions, and you are complaining that employees are not using properly that chance to be constructive, then you probably need to revisit your strategy.

Obtaining suggestions and feedback should also not be an exclusive HR activity, within IT there are several ways we are engaging people, we should be able to get their opinion in an effective manner to improve our work and services.

This is also another great chance to shape your culture in a positive way, having these channels in place are going to encourage people to communicate and work together. What’s the collaboration you can expect if the company prefers only for employees to provide feedback to their manager once a year?

Some of the aspects that you should take into consideration when you are designing the feedback and suggestions processes:

• Timing is everything. Obtaining feedback at the right time for IT is crucial, getting it closer to a significant event (for example, new collaboration tool deployed) will make the responders have the feedback fresh in their minds.
  There are several platforms built for this purpose, like Nexthink, which among other things can bring a non-intrusive feedback integration that can trigger simple questions after an event.

New web portal deployed, question asked, and the popup disappears after that

• Easy access. Providing feedback should be simple and fast, like the platform mentioned above. If the user needs to struggle to find a link somewhere, no one will feel encouraged to provide feedback proactively. The same happens if your feedback and suggestions process is just a mailbox waiting for emails, don’t expect any eloquent interactions.
• Encourage direct feedback but respect anonymous preferences. Direct feedback requires an evolution of culture for most companies, not as simple to implement and have good results. Starting this habit in people, developing a positive process for providing direct feedback will provide transparency and encourage constructive opinions.
  If people feel that anonymous feedback is the only possibility, while you need to respect that, it could be a sign that there are some issues in the company’s culture.
• Good feedback and suggestions practices start from the top. Managers and leaders need to have a good habit of providing feedback, which in this case needs to be direct.
• Link suggestions to action items, results and compensation. Any culture change you may try to introduce can be disregarded easily if the people can’t perceive that the feedback they are providing cannot be linked to direct results.

2.5 Does IT attract and retain good talent when needed?

• There have been 3 or more CIOs in the past 5 years.
• Constant organizational changes in IT, considerable IT attrition.
• The company is not considered an attractive employer.

• IT and business planning have revealed the need for new skills.
• Budget available for increasing IT staff when necessary.

• Digital initiatives required to hire and promote technology delivery roles.
• Company rewards new ideas and promotes people who constantly deliver.
• Company receives applications for young and a recently graduated workforce, as well as consolidated talent in the industry.

Retention is a very consistent challenge most companies have, and in IT that is highly noticeable. A recent study by LinkedIn indicated that technology companies have the highest rate of turnover: 13.2%.

There are several aspects to name regarding this attrition trend in IT, an interesting analysis can be found in this LinkedIn report about the reasons people leaving their jobs. The #1 motive being lack of advancement opportunities, followed by: unsatisfied of senior management; unsatisfied with company culture; seeking more challenging work; compensation; and lack of recognition.

The strategy recommendations to reduce turnover are many, but I’m going to focus briefly on attracting talent and how technology is a key aspect in this pursue. The reason for that is simple: Gen Z employees (born between 1995 and 2015) represents 74 million people in the US right now. Of course, not all are old enough to be employees in a company, but they are going to be ones shaping business requirements, culture and user needs for the years to come.

Here are some important facts about the Gen Z to consider and how it relates to technology:

• 80% want to work with cutting edge technology and more than one-third are interested in IT careers.
• 91% say that technology offered by an employer would be a factor in choosing among similar job offers.
• 80% believe technology and automation will create a more equitable work environment.

If you were wondering about Millennials / Gen Y (born between 1980 and 1994), one key statistic to consider is that 44% of this population think their workspace isn’t smart enough and 42% state that would quit a job with poor technology.

Technology industry trends generate an impact on people, same as in IT. Outdated technologies do not appeal to young talent, this is one of the biggest factors these workers would define and share with peers that a company is not attractive.

2.6 What is employees’ perception regarding the workplace environment?

• The company maintains an old design and layout for the offices, without considering any budget on improving the workplace
• Layout is focusing on cubicles and closed offices, interaction is minimum.

Did you know that the cubicle was designed in a way to mitigate the efficiencies inserted within the Taylorism?

Frederick Winslow Taylor designed (from a scientific paper in 1917) a series of desks rows to optimize space and workflow. In 1930s workers complained about this layout reduced them into numbers and the workplace seemed more suited for machinery than people. The labor laws those days were not very beneficial to the normal employee.

Workers went to strike asking for pensions, increased salaries and a better treatment in general by companies. Demands to companies arise to treat workers with care about their psychology and not as machines.

Some of the changes introduced in the following years were new designs in the office layouts, which included the cubicle in the 60s as way to squeeze more people into less space for as little money as possible by giving them a “personal space”, which at the end of the day encouraged isolation and little communication. Eventually being contradictory for the original request of a more humane treatment.

• Company is investing money on decorating interiors and exteriors
• Cubicle layout is being reduced for more open spaces

• Company has a defined and modern aesthetic in the workplace
• The workplace is designed with the work habits in mind. Open spaces approach and plenty of meeting rooms
• Company offers recreation and fitness facilities

There are different aspects to consider about a good workplace environment regarding facilities. One of the big problems for how most companies manages this aspect: The total cost of occupancy is typically the second largest expense, with the caveat of the actual utilization rates being 50% or less.

With the cubicle style there’s also the emotional aspect of employees that don’t perceive collaboration is encouraged, therefore they also don’t feel valued. People connections are low, same as pride and desirability to work in the physical workspace.

The open space approach is the common layout for companies looking to transform. Which should not be considered as an “open office” itself, since worker segmentation is required and a variety of different types of spaces. Technology, again, needs to be an enabler for these types of facilities.

Within IT we must not be outsiders in this conversation. We are enablers and facilitators, we -should- understand user needs, motivations, challenges and collaboration requirements. If your IT area has no saying on the facilities discussions, speak up; demonstrate the valuable information from personas and user journeys and how the workplace should adapt.

Regarding facilities and common best practices, are aligned with WELL initiatives and certification. Here are some of the key characteristics needed for workplace from this perspective:

• Town hall: Public common space. Food and large screen projection.
• Neighborhood: Group areas for similar work. Monitors, docking stations and mobile devices available.
• Workbench: Similar than meeting rooms, temporary workspaces for collaborative work. Whiteboards included.
• Learning lab: Areas for education and knowledge sharing. Usually with workstations.
• Library: Communal spaces for ad-hoc work requiring solitude. Quite area.
• Alcove: Individual spaces for work or relaxation.
• Wellness room: Multipurpose area for recreation and/or fitness.

The third article for the self-assessment guide will also be related about IT’s involvement on understanding the business and other areas in the company. Integrating with the business and leadership are key aspects for every IT leader, this is also something that several of us struggle with.

We’ll discuss about operational planning and how reactive IT has truly become, acting as a detriment on making IT a business key stakeholder.

I’ve covered in the first article the “busy dying” portions that, of course, represent the habits, misconceptions and some of the existing problems IT departments have. Discovering and accepting those will be a key aspect to improve and getting into the busy being born.

If your time to you is worth savin’
You better start swimming, or you’ll sink like a stone
For the time’s they are a-changing
Bob Dylan – The Times They Are a-Changin’ (1965)

One of the key aspects most IT organizations should start would be on being capable to do an honest self-assessment. Which should apply for the IT department, the organization they are part of, the understanding they have around their environment and the business.

Having a consistent self-assessment will set the foundation for a path moving forward. This will acknowledge the areas you need to overcome, where there’s some room of improvement and areas where you’re delivering what is required. This self-assessment can be even shared with your leadership and other areas, setting precedent about a healthy practice to implement and maintain over time.

The self-assessment set of articles will be divided in different sections:

• Organization-wide assessment: Perspective regarding the overall organization you are part of.
• Users understanding: Your take on how well you understand the users you are supporting and their singularities.
• IT’s integration with other areas and leadership: Understanding how well the IT integration with other departments and the leadership connection.
• IT’s integration with the business: Perspective on the involvement and incorporation of the business goals and strategies within IT.
• IT and the technology industry trends: Self-assessing the technology and tools offered from IT and if those are following the latest trends.

This set of articles can serve as a clear self-assessment guide for IT organizations. Each of the questions of will have different answers with a certain weight, rated in stars, it should be self explanatory. To close each question, I’ll provide my thoughts on how to overcome if you are getting a lower rating and the aspects you should be paying attention the most. 

1. Organization-Wide Assessment

1.1 Is the current organizational structure reflecting company’s strategy and priorities?

• Organizational charters don’t exist.
• Departments don’t have formal links in the org chart connecting them.
• There are always conversations and rumors about organizational changes, but no clear strategy revealed justifying it.

• Organizational structure is clear to all employees.
• Department charters are published and job roles are clear.
• Departments still work mainly separate with their own strategies.

• Organizational structure will change when necessary to accommodate business goals.
• There are organizational success metrics defined.
• There are clear rules for operation within the organization to avoid responsibilities overlap or conflicts.

There are certain organizations where the first thing to do is to define the chain of command, level of centralization (functional, divisional, etc.), and who are the people that are going to be leaders. And after that, adapt your structure based on those requirements. In those scenarios, innovation activities will mainly depend on the leader attached to that area.

Speaking in a very high level, to understand if the organizational structure can be successful for the business goals (once those are defined), there are a few components that need to be present:

• Governance plan needed to make decisions. Here a steering committee can be defined initially to develop the business plan and formalize metrics. With the leaders’ definitions, the hierarchy and organizational chart can be demarcated.
• Rules for operation. This is where you state how formal and informal groups will operate. It is important to have some of the critical definitions to avoid confusions or conflicts.
• Work distribution. Clear demarcation of activities and roles, answering to the organization goals.
• Adaptability to change. The business plan will change; therefore, your organization structure should change as well. Divisions and departments should adapt as you are evolving and the company grows.
• Simplified communications among divisions. Structures and processes that could transform the divisions into silos should be avoided. In these cases, areas not able (or willing) to communicate with other areas generate dysfunctional organizations. Formal and informal channels of communications are needed in every organization.

1.2 Can the company leaders connect strategy with the work being done across business units?

• Our leadership doesn’t discuss corporate direction, or if they do, it is very vague and inconsistent.
• There are no publications or briefs around company’s strategy.

• Each department has its own plan, which informs the projects IT takes on.
• One or more key leaders think strategically and formalize a vision or guidelines for the future.
• Company’s has promoted new executive and innovative positions (like Chief of Employee Experience Officer) that can feed into a strategy.

• Strategy definition is an enterprise wide practice and can be revised recurrently if needed.
• IT leaders are asked to actively participate in strategy meetings and ensure that IT projects are aligned with business goals.

As I mentioned in my previous article, strategy is a never-ending journey. As business and technologies are always changing, our strategy and hence our activities should adapt as well.

For giving you a concise example, in the agile framework the product backlog defines the functionalities needed for the solution that is being asked, usually translates specific use cases required by the customer into line items that need to be delivered. A sprint backlog defines activities for the team to execute within a sprint (one week, two weeks or another period defined).

A very basic principle of the methodology is that every item in the sprint backlog needs ALWAYS to be associated with an item in the product backlog. If you have items that don’t match, you are doing something that is not required to be delivered.

Talking about organization wide, the principle should stay the same. Each department and roles need to have clear activities and responsibilities, and must match with business goals, including when those change.

There’s also a common misconception about IT people that assume they were hired because they have a specific skillset/experience to execute certain type of activities (i.e. implement and operate a specific backup solution). You are being hired to support a business requirement, which can require adapting your activities and responsibilities when those business requirements and strategy change (i.e. manage a different backup solution or change how you operate the solution).

1.3 Is your company supportive of changes when they are necessary?

• It is safer to reject than accept an idea when someone suggests a change.
• People can be penalized or accused of “not being compliant” if they intervene or try to change existing processes in a positive way.

• The company has some leaders looking to implement changes, only small changes are being applied.
• Recently leadership introduced a change that represents a big shift from the business as usual mindset.
• A center of excellence area/team has been defined to deliver new capabilities in the organization.

• Employees are results oriented and they are rewarded for being part of new outcomes.
• Change and innovation are encouraged by providing resources and tools (i.e. innovation labs).
• There was a recent disruptive change that produced good results, people associated with that initiative were recognized.

Inertia is Newton’s first law of motion, a fundamental principle of physics. He defined it as “the innate force of matter, is a power of resisting by which every body, as much as in it lies, endeavors to preserve its present state, whether it be of rest or of moving uniformly forward in a straight line”.

This is a very powerful force within most organizations, based on the concept of cognitive inertia. This is the tendency to stick to existing ideas, beliefs, and habits even when they no longer serve us well. It’s much easier to keep the same train of thought/opinion/course of action than to reflect on the chance that we might be wrong and update our views. It takes a lot of work to stop that inertia from happening commonly in organizations.

Inertia as the basic principle to explain objects in state of motion or stay at rest

“Business as usual” you may hear from certain leaders when there’s a significant change in the market and not encouraging employees to adapt accordingly.

“It’s OK. We’ll be fine” example provided in my first article regarding the Blackberry leadership when the iPhone was released. We know the end of that story.

Not all changes or alterations would require drastic strategy modifications, but you are required to adapt. Make sure you are measuring the right metrics and rewarding results that offer innovation.

1.4 How is your company staffing projects and initiatives?

• Company does not qualify who works on what project, it is mainly decided on who’s available when the work is needed.
• People with good personal relationships with leadership tend to get assigned to the most interesting work.
• Employees stay within the area and project they get hired, moving to different areas it’s not encouraged.

• There is certain formal delineation regarding existing talent and skills that are used to match initiatives.
• Team collaboration occurs when there’s a specific need, no cross-team proactive assignments when the initiatives are being planned.
• Managers with most successful projects tend to keep their staff.

• Business strategy defines projects and suited team members to participate.
• Employees are being measured by company and team goals.
• Leaders right balance the assignments to make sure even people without the required qualifications can participate in new initiatives to keep them developing.

Assigning people to projects and initiatives can be very much a cultural aspect in each company, based on certain “comfort zones” and sometimes not taking risks to make changes. Looking to maintain the inertia mentioned earlier.

Matching all initiatives with only people with the required skillsets won’t be always a good idea. As an IT leader you need to make sure you are providing the right level of “discomfort” to your employees. Keep them motivated with meaningful work, which in certain cases are not going to be the activities they know how to do it.

If you have motivated and proactive employees and you want to keep them that way, why give them something to do that they know how to do it by heart?

1.5 What are, in your perception, the company’s levels of inclusion and diversity?

• Company leaders don’t talk much about diversity or inclusion activities.
• The company tends to hire and/or promote people with the same educational background and personality types.
• It is a known secret that there are pay gaps among employees with different genders, ethnicity and/or educational backgrounds.

• Executives have begun emphasizing the importance of diversity and inclusion.
• The existent pay gap have been detected as an issue and the leaders are acting against it.
• HR training is being developed and occasionally provided to employees around diversity and inclusion.

• Diversity and inclusion are visible at all levels within the company.
• Leadership sponsors and communicates around programs for diversity, female leadership and other inclusion initiatives.
• Diversity and inclusion are mandatory trainings for new employees. Corporate direction enforces the message of the importance and penalties that could apply if they are not met.

Diversity and inclusion are always sensitive subjects that are difficult to discuss in a company, even with peers. There’s a very important cultural factor on these topics and, if there’s an existing problem, needs to be tackled right away.

The behavior needs to be consistent across the organization. If you are leading an IT department, make sure you are doing your part. For example:

• Rethink and redesign hiring and promotion practices.
• Make sure team members can speak up, propose new ideas and implement suggestions.
• Request feedback, ask your employees what motivates them, makes them happy and valued. And, of course, make use of that information.

A recent Gartner study, considering IT and technology organizations, showed that teams that are inclusive and diverse outperform the teams that are not:

1.6 Is there a clear process to determine prioritization in the organization?

• The person that screams the loudest (i.e. escalates rapidly) takes priority.
• Personal relationships and politics are main factors on defining priority.
• Disagreements about the importance of business initiatives are common.

• Business units have roadmaps that provide guidelines with priorities.
• Prioritization disagreements happen from time to time, but there are clear escalation paths to solve the conflict.
• People are encouraged to accept any new requests to avoid conflict with other areas.

• Governance informs delivery priorities and there’s a steering committee in place.
• The prioritization process is common across business units and are tied to business priorities.

If you are an IT leader, you know there’s a recurring problem about saying yes to any new request that comes in, with the urge that some people feel the need to satisfy the person making the ask. Similar than a ticketing system, trying to solve them as they show up in the queue. There are some cases when leaders believe that saying yes to everything makes you a “business driven” department, it does not.

Organizations suffer, based on a common human behavior, that we tend to prioritize short-term tasks with faster results, than executing activities that will give us long-term outcomes.

Having an efficient and formalized governance process, integrated with the business goals, is going to be critical to solve any prioritization conflicts. Effective being one of the keywords associated to governance processes, taking weeks for a committee to define a project approval might upset stakeholders quite easily.

1.7 Is your company reaching out to partners to enrich its products and services?

• The company prefers not to partner with the fear of losing control by having other companies involved in the initiatives.
• Leadership believes that all the work should be done internally.
• Acquiring services from partner companies is usually very challenging to be able to justify.

• Partnership opportunities are mainly opportunistic based on a specific need.
• Partners are usually former employees that have good personal relationship with leadership.
• There are partners in place but no centralized ownership or negotiation.

• Company relies on partners to empower products and services.
• Partnership network is built based on strategic definition and gaps in company’s capabilities.
• Tiered partners are defined, including preferred partners.

A study from the CMO Council says that 85% of business owners believe that partnerships are essential for business success. But still, forming partners and alliances represents a challenging task for most organizations.

There’s an interesting aspect about partners, usually the strong and fruitful relationships transform not only on successful initiatives but also on opening new markets, attracting more clients and so on. In those cases, partners can be considered indeed as an important customer, yet they are not treated like that.

Dr. Evil informing about a new formed alliance that will benefit both companies strategically 

The main issue on having meaningful alliances resides on the lack of governance to formalize a strategy for partnerships. We must be clear that an alliance will require short-term mutual interest and a long-term compatibility. You don’t have to agree on everything; but the values, vision, financial resources and goals need to be compatible to make the relationship successful.

When you’re formalizing the strategy, consider objectives, goals, milestones, how to measure results and exit strategy for partnerships. Having those formalized will set expectations properly with your potential partners.

1.8 What is company’s position regarding Research & Development (R&D) and skills development for employees?

• Even though the company says they are, there is little to no investment on R&D.
• Everyone’s replaceable, HR only needs to look for candidates with the skillsets required.
• Performance reviews only depend on managers, there’s no formalization.

• Company allocates time for R&D, but not consistently.
• Job descriptions are comprehensive with clear details around skills requirements.
• Performance reviews process formalized and consistent.

• Company encourages the use of R&D, innovation labs and other activities to enhance employees’ skillsets.
• Interview and review processes encourage hard and soft skills.
• Changes are routinely made to job descriptions and measurements criteria.

This is something I’ve mentioned in my previous post. There’s seem to be a lack of sincerity in a lot of companies regarding how much time and budget is actually included in research and development.

If you are in an organization that does not include IT as part of the business and strategy planning, most likely your focus will stay around keeping the lights on, which will tend to define your activities as mainly operational and not prioritizing on innovation.

As mentioned before, IT must do their part, you need to deliver on running day-to-day activities to get leadership sponsorship to start innovating. It’s going to be challenging to obtain sponsorship if IT provides a lot of headaches around system availability or solving critical problems.

Of course, the business side must do their part as well, they cannot expect for IT to deliver excellence or brand-new technology if executive leadership is not providing the budget to upgrade old platforms that takes most of IT team to support it.

The second post of this self-assessment guide for IT organizations will focus around how well do you understand your end-users. This is probably where a lot of IT departments may be struggling the most since it represents one of the topics they feel there are no understanding gaps. And it’s hard to solve a problem you can’t fully accept. 

We are going to talk around personas, user journeys, IT capabilities for attracting talent and reducing unnecessary attrition, and the overall perception of the workplace. 

I decided to come back to my writing process with a musical approach and related to a topic we all see quite often, even if we are not paying attention or just deciding to look away. Let’s get right into it.

What’s the Bob Dylan Reference?

I’m not a Bob Dylan expert, but you don’t have to be quite knowledgeable to respect a Noble prize winning artist with almost 60 years of writing songs (started in 1959). The song that I’m referring in this case is “It’s Alright, Ma (I’m Only Bleeding)” (from album “Bring It All Back Home”, 1965) and the specific quote:

“Proves to warn that he not busy being born, is busy dying”

I’m not going to go over the message around the entire song, there are already entire articles and papers published about that. I’m going to focus in the particular quote mentioned and how this relates to the IT culture.

Bob Dylan’s message in this quote is fairly simple: If you stop reinventing yourself, you’ve started dying. IT is no different, ever changing technologies and solutions as we have today require redefining yourself, your goals and paths moving forward.

In IT, as in many other industries, the challenge is to live and grow with a constant presence of uncertainty and change. We feel obligated to be aware that the notion of failing to continue growing, is the same as killing our careers.

How many of us have seen IT people, roles, departments or even entire organizations with a steady and unchanged definition and position around technology in general? Even though we can see that as a very common factor in the industry, how many of us are actually convinced or willing to make the change for a transformative approach?

Article Series Approach

The idea in this series is to go over the common scenarios I’ve discovered and discussed with companies around the true IT Transformation present in most of these organizations.

Being “true IT Transformation” the key term. As you probably know, some organizations or leaders refer to IT or Digital Transformation as simply a way to represent budget reduction or asking to implement that fancy new definition in today’s industry (some of them call it “cloud”, “big data”, “blockchain”, you name it) but without any true strategy behind it.

The first section I’ll cover will be the “Busy Dying” portion, which is the summary of some of the common bad habits present around IT and companies. Some of the behaviors that define an IT organization that can’t or won’t innovate itself and the business they are supporting.

The second section will be the series of recommendations and best practices I’ve seen IT departments take in order to truly transform themselves, the “Busy Being Born”.

Ironically (or not), we’ll start with the part that relates to “dying”, this is needed in order to comprehend and assimilate the areas where we may be having problems or at least there’s room of improvement. If we don’t acknowledge this first, will most likely mean that any innovative approach, “being born”, will be a new thing where old habits will prevail and nothing changes.

There must be a transparent and honest assessment of the “as-is” environment in order to be ready to define the “to-be” state. I’ve learned that this is something any organization should do, it doesn’t matter if you just have a small shop or you are a Fortune 50.

Busy Dying: Business Disconnect

This is by far the most common discussion topic that I’m sure most of us already heard enough about it. After all, IT can’t thrive if it’s not collaborating with the business to succeed.

We will also notice that some of these areas about to cover in this issue are related to other topics described later in this article.

I would like to list some of the most common scenarios and talking points I’ve found, including the different perspectives between IT and the business itself. Hopefully, if you are starting the path of defining your current state, some of these should sound familiar:

The IT Perspective about the Business Disconnect

• “How can we add value within IT if we don’t have any insight into business planning?”
• “Our IT management does not have a seat at the table when important business decisions are being made”
• “We don’t have the authority and/or funding to do what needs to get done”
• “We keep getting requests to reduce our IT budget in order to support other business units”
• “IT is not invited to participate on key business initiatives unless is absolutely necessary”
• “How can we look for innovation in a corporate culture that’s deadline driven?”
• “We have to provide 24x7x365 support and almost all of our man hours to maintain old and complex applications”
• “Business is asking us to implement nice and shiny dashboards and get a brand new look for their day to day apps, but still our systems are supporting legacy applications the business is not willing to change”.

The Business Perspective about the IT Disconnect

• “IT is telling us that we have to provide funding. How much of their development should come out of our budget?”
• “Funding should support the business needs and not IT’s”
• “How do we get CIO/IT Director commitment around fulfilling corporate objectives?”
• “IT seems more interested on resume building instead of driving business value”
• “We don’t understand when IT is talking to us”
• “We don’t have any option to obtain support if IT says they’re too busy”
• “How can we trust in IT when the organization keeps changing their technical personnel constantly?”

As we can see, some of these scenarios could be related to communication issues and not necessarily to unwillingness to support from one to the other. I’ve received some of these similar contradictory comments within the same meeting between IT and business leadership.

We need to understand within IT that if we are looking to lead with innovation but the business people are not actively engaged, as much as you can promise that innovation is coming, it probably won’t mean much to them.

The other key takeaway regarding this is that, as long as business stakeholders stay alienated from IT, they will make assumptions about what IT can and can’t deliver. Which is a quite dangerous to say the least.

Another big misconception with IT departments today is that they confuse saying yes to everything with being a business driven department. Saying yes to any ask doesn’t make you automatically understand and prioritize properly what the business really needs, on the contrary. Most likely in order to say yes to something you had to say no to different ask.

Busy Dying: Lack of Training and Effective R&D

When a company says they invest 20% of their time/budget around research and development, in my mind I’m leaning more to a 2% being the real percentage.

As you can imagine, this does not relies only on the business side “allowing” IT to obtain the proper training and knowledge, it also relies on IT being able to prioritize accordingly and having a good sense of strategy around the research and training we are looking for.

There’s some clear reasoning around the business comment mentioned earlier about IT being more interested on resume building and getting that really nice certification that few people have. Therefore, do we understand what are the key business asks that we need to support and therefore commit with R&D according a business prioritization? Have we formalized that plan and share that with the business to obtain feedback?

There’s seem to be a misconception of IT people that believe they are being hired solely because of their specific skillsets/experience; the truth is you are being hired to support a business requirement.

And of course on the other side, organizations and business leadership must have the willingness to support IT if innovation is required. Making assumptions that IT by default will always have free time and budget available to support their own R&D is an erroneous belief.

There’s another aspect that IT suffers regarding R&D based on the amount of new solutions and technologies being released on daily basis, it could generate confusion and “paralysis” from IT leadership on deciding what is the right approach and solutions to start investing, and the “easy button” in most cases tend to not to invest on anything instead of making a mistake. This, again, has to do primarily with the strategy behind and having a business integration.

Busy Dying: Fear of Strategy Definition

So, what are we talking about when we say “strategy”? Basically strategy is the craft we follow on figuring out which purposes are both worth pursuing and capable of being accomplished, with the main objective of gaining competitive advantage.

The key aspect to understand about it is that building your strategy is a never ending journey, especially around technology. With solutions and requirements that constantly change, strategy is just something we should define and revise periodically.

The problem with strategy definition and IT is that seems outside of everyone’s comfort zone; there are certain IT leaders that feel their main focus should be around defining the right approach for platforms operations. In addition, that approach is mostly concentrated around resource allocation, budget and IT projects, but without any innovative or strategic quality.

What some IT leaders call IT strategy definition ends up being operational planning to keep the lights on.

On the the continuous quest of strategy definition and adaptability is fundamental to be inherited from an organizational culture. It’s quite challenging for an innovative IT leadership to succeed if the organizational leadership is saying “Don’t worry about it, we’ll be just fine with the solutions we have now”.

After all, “culture eats strategy for breakfast”.

“It’s OK. We’ll be fine”

Those are the words that first came out from BlackBerry CEO, Jim Balsillie, to Mike Lazaridis, the founder and vice chairman of BlackBerry at the time, when the iPhone was first launched in 2007.

Jim Balsillie and Mike Lazaridis, the good old days for Blackberry

Lazaridis was so amazed with Apple’s product that called his CEO, Balsillie, to review the launch video and the features of the smartphone that changed the industry entirely.

Blackberry’s CEO was confident that, because of iPhone’s poor battery, security concerns and other aspects, the BlackBerry market control in the Enterprise space was not even close to feel threatened. “It’s OK. We’ll be fine” said Balsillie to Lazaridis. The company’s leadership shared the same confident with a “not to worry” message to their entire organization.

We all know what happened. BlackBerry (ironically enough originally called RIM – “Research in Motion”) came from a +$1.9B profit on yearly basis to -$5.8B deficit in less than 5 years.

Unfortunately there are CEOs that, when they connect strategy and IT, it’s usually in the context of saving costs or outsourcing systems and skills.

Busy Dying: IT with No Connection with the Rest of the Organization

This is related to the business disconnect mentioned earlier, but an IT organization with a clear idea of what the business is requiring can still be failing on connecting and collaborating with other areas within the organization.

Being “too busy” to talk with HR around their onboarding process, or saying “not our priority” to support the new campaign for the Marketing department, it is also another way of failing.

Have you seen an IT department that is defined as successful in an organization but has little to no communication with other departments?

Busy Dying: Prioritizing Formal Processes and SLAs instead of Trust and Integration

Every IT organization that has been around for some time will have formal processes and governance models that are required to be consistent with the services you are providing. But there are some drawbacks that some of these organizations fail to see as areas of improvement.

A good and simple example are the IT committees designed to revise and approve requirements. Sometimes these committees will review every request made to IT, even the simplest ones. This, if not designed properly, can slow things down tremendously and generate a lot of frustrated stakeholders. Some of the issues residing here could be around having a lot of members in these committees, making the decision process longer since generating consensus in a large group is always harder.

Service Level Agreements (SLAs) can also represent a trap for IT. There are IT organizations that treat SLAs as their main contract vehicle and try to shield them behind it at all cost. In those cases, IT just stops worrying about the end result or business value, as long as IT is respecting the contract, they feel like their job is done. Sometimes these agreement contracts are outdated or inadequate, considering goals that are no longer the same since the business goals might have changed as well.

Busy Dying: Lack of an Offering and Failure to Provide New Services Periodically

Probably if you are not offering a new service every month, you may be starting to lose credibility and trust within your organization.

If the problem resides on the fact that IT is not able to offer new and enhanced services because they are still supporting old legacy applications, which may have been with the company for +20 years, they are slow and outdated, we must remember that the organization’s perception is that IT is the one slow and outdated. We, as IT, are tied to the perception generated by the systems we support.

These new services IT should offer periodically are not usually required to be complex solutions that radically change the organization, it could be as simple as new set of applications, improving recurrent communications, etc. These interactions and new offerings will facilitate the legitimization process for the IT organization. Even if you are limited, if you are showing a clear intention to add value, your actions will do the talking.

Busy Dying: IT Only Being Measured to Keep Lights On

There’s a common behavior in most organizations around their CEO or leadership missing the opportunity on providing clear and valuable performance metrics. It becomes more relevant and impactful for IT when there’s a CEO not very comfortable around technology in general.

Unfortunately there are plenty of CEOs and CFOs continue to measure IT on the trendiness of their smart devices, email uptime, and office applications. In addition to that, this leadership is constantly revising IT costs, system availability and help desk response time as key metrics. This represents a huge challenge for IT to maintain a good reputation and relationship while you need to innovate.

There’s also another recurrent discussion from leadership about Total Cost of Ownership (TCO), usually oriented on reducing these costs as mentioned earlier. TCO discussions don’t care about benefits, even more, they tend to discourage benefits and value. The explanation for this is simple: The easiest way to cut costs is to support technology that is less functional and capable; even though some would not refer to this concept literally as described, and probably use terms like “optimization” or “efficiency”. If you reduce the utilization of a service, you are reducing costs.

There’s a reality behind all this, IT needs to excel at running the day-to-day platforms so you can discuss going beyond the activities around keeping the lights on. Most of the times when we hear about an IT leader that was removed from his/her position is due an operational failure; it is really uncommon to hear about a person being removed because they were succeeding in operating IT but failing to incorporate innovation.

Busy Dying: Lack of Understanding between Important and Urgent

Managing your time comes down to one important thing: understanding the difference between importance and urgency. A big part of the challenge of time management emerges when you start to lose the sense of importance.

Email is probably the worst vehicle for conveying urgent, time-sensitive information. Soon as you sit in your computer, emails start coming up and you feel the need to start answering them, if you are really considerate you’ll start replying the older ones first (typically received the night before or really early in the morning) since those people have been waiting longer than others. It feels urgent, but it isn’t urgent.

Differentiating important with urgent tends to be more of an intuitive decision. The propensity for confusion here is related about IT, in general, prioritizes on short-term concerns with faster results than long-term initiatives. The intention is to go deeper around this topic in the next article, but the best reference around understanding the difference relates to the Eisenhower Decision Matrix.

Eisenhower Matrix summary from https://appfluence.com/

Related to some of the discussions earlier, IT feels far more at ease solving technical challenges around the platforms they are supporting. Technically inclined staff are more comfortable playing in the technology sandbox than solving business problems. So, IT might “feel the urgency” to solve a challenge that does not have much criticality instead of supporting a business request that does not involve a technical challenge.

Busy Dying: High Dependency on Manual Provisioning and Discarding Self-Service

A recent report from McKinsey suggests that half of the job functions performed today by people could be automated using existing technology. This does not represent less jobs for IT, it means there should be more freedom to be more strategic.

If you are able to demonstrate real value, no one will eliminate your job. And we have to embrace that a lot of times in order to show value, you need to think strategically. Even more, these McKinsey reports state that jobs may grow in the upcoming years since productivity increases and, with the progression in technology, more and new type of roles will be required in IT.

We’ve all seen our fair share of scenarios where shadow IT takes preference among power users/organizations (i.e. development teams). Shadow IT happens, BYOD has become BYOIT for some companies. If they can obtain the full environment they need within minutes by doing it on their own, or alternatively they can ask IT and get it after several days or weeks, the selection process it’s not that hard. And this problem isn’t an intrinsic limitation from IT, it’s a choice.

Today, lots of IT is happening outside the IT department, mostly in the form of SaaS/PaaS brought in by business managers as islands of automation. We must be ready for that.

With millennials being the workforce of the future (estimated 75% by 2025) it is nearly impossible to ignore self-service capabilities. Millennials do not require much training around applications, that’s a known fact, most of today’s productivity (like most smartphone apps) is represented by intuitive processes. Our IT services need to be flexible, accessible, engaging and predictable in order to keep the final users invested.

As we already know, Azure Stack is intended to be the Azure instance in your datacenter, but even though they are using the same code, there are still some disclaimers to consider regarding Azure features available in Azure Stack. Also the intention in this article is to provide easy and quick comparison charts between Azure Stack vs Azure components.

Not all the components described in this article will be IaaS per-se, I’m separating the services in different articles for practical purposes. What I’ll cover in this article is:

• Azure Stack Virtual Machines: Understanding disclaimers, expand the concepts about VMs Extensions and the comparison with public Azure.
• Azure Stack Networking: Reviewing Virtual Networks, Load Balancer and VPN Gateways features; including their considerations and the comparison with the public Azure versions.
• Azure Stack Storage: Understanding what are the components within Azure Stack storage, considerations and a table comparing with Azure.
• Azure Stack Marketplace: Reviewing the Marketplace available, integration capabilities with Azure and the existing considerations.

For deeper dives into detailed features available in the GA version of Azure Stack, please review my previous article: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More”.

Also more details about Azure Stack pricing available here: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed”.

Azure Stack Virtual Machines

Offering Virtual Machines is, as we all know, the main concept for “Infrastructure as a Service” (IaaS). The public Azure version has over 20 (to this date) different VM types, which translates in several more different sizes within VM types; Azure Stack has a reduced version of those sizes.

Here are some important disclaimers about Azure Stack VMs to consider:

• Virtual Machine sizes in Azure Stack are consistent with those found in Azure and custom sizes are not supported. Tables below show details about sizes and specs.
• As reviewed, not all sizes are present but more will be added in future updates. It is also expected that Azure Stack will not have certain sizes available (very small or very large VMs).
• Azure Stack will not ship with base OS images in the gallery, those must be imported or obtained through Marketplace.
• In single-node POC (included in Azure Stack Development Kit), a single fault domain and upgrade domain are provided.
• Some current and legacy distributions of Linux which ship with the Azure Agent codebase may have an incompatible version of the Azure Agent.

In the Dell EMC blog, there’s an interesting article written by Paul Galjan, regarding IaaS and Azure Stack. The interesting discussion relies on a comment made about a Microsoft Program Manager: “Azure Stack is not a virtual machine dispenser”.

Before generating a confusion with that statement, the intent of that comment is related about if you just need to provide virtual machines and nothing else, there are probably more efficient ways to deliver that instead of Azure Stack. The solution is intended to provide much more than a “virtual machine dispenser”.

Azure Stack Virtual Machine Extensions

Virtual machine extensions is a component highly used in Azure, basically the extensions are small applications that provide post-deployment configuration and automation tasks on Azure virtual machines. There are common extensions available to allow simple tasks as resetting administrator’s password or deploy an anti-virus; or more complex ones like implementing a Docker configuration.

The possibilities for tasks to execute are pretty much infinite since the extension for “custom script” is available, where you can integrate it with ARM templates, Azure CLI, PowerShell or the REST API.

In Azure Stack, here’s the list of current VM extensions available (more extensions expected to be available in future updates):

• AzureLogCollector: Used to collect logs from one or more VMs.
• BGInfo: SysInternals tool that writes useful server information to the desktop as a wallpaper.
• CustomScriptExtension: Executes custom scripts as part of the deployment.
• JsonADDomainExtension: Extension to join a VM to a specific domain.
• VMAccessAgent: Several activities can be executed with this extension, like resetting passwords or adding a new user.
• CustomScriptForLinux: Executes custom scripts in a Linux VM.
• OSPatchingForLinux: Extension to provide OS patching within Linux VMs.
• VMAccessForLinux: Provides several activities like resetting passwords or adding new user in Linux VMs.
• DockerExtension: Allows installs and configures the Docker daemon, Docker client, and Docker Compose.
• DSC: Applies Desired State Configurations to a VM.
• IaaSAntimalware: Enables Microsoft Antimalware for Azure feature to provide real-time protection.
• IaaSDiagnostics: Enables monitoring and diagnostics capabilities.
• MicrosoftMonitoringAgent: Configures VM monitoring.
• SqlIaaSAgent: Automates SQL activities.

Azure Stack VMs Extensions available:

Azure Stack Virtual Machines vs Public Azure Virtual Machines

As a quick overview let’s take a look for a comparison between Azure Stack VMs and Public Azure VMs:

In case you were wondering the specifics on the resources available on each of the VMs for Azure Stack, here it is:

Azure Stack Networking

In the Azure Stack Networking world, we’ll cover here the Virtual Networks, Load Balancers and VPN gateways.

Azure Stack Virtual Networks

Virtual Networks in Azure Stack is the simple representation of physical network and a logical unit of isolation.

Here are some disclaimers about Azure Stack Virtual Networks to consider:

• Virtual Machine network speeds are not configurable within Azure Stack.
• Standard public-facing DNS URIs for standard services must be handled manually.
• For Network Security Groups (NSGs, network rules to allow or deny traffic) and User Defined Routes (UDRs, manual routes created), some keywords which are specific to Azure may not be available.
• The ability to add User Defined Routes (UDRs) to Gateway subnet is not part of the initial release.
• Network Virtual Appliances (NVAs, used to control the flow of network traffic) with modified versions of the Azure Agent may not function properly in Azure Stack.
• IPv4 is only supported at this time (IPv6 support will be supported in a future release).
• Features such as Traffic Manager, App Gateway, and ExpressRoute will not be available in the current release of Azure Stack.

Azure Stack Load Balancer

The Azure Load Balancer represents the service to provide high availability and network performance to applications integrated with this component. An Azure Load Balancer is a Layer 4 (TCP, UDP) load balancer that distributes incoming traffic among healthy instances of services defined in a load-balanced set.

We can use this component to load balance Internet traffic to a VM; traffic between VMs; between VMs and cloud services; between on-premises devices and cloud VMs; or forward external traffic to a specific VM.

There are two considerations around Azure Stack Load Balancers:

• This feature is expected to generally work the same as Azure and Windows Server 2016 today
• The Load Balancer feature uses the Windows Server 2016 Network Controller

Azure Stack VPN Gateways

A Virtual Network gateway is used to send network traffic between Azure Virtual Networks and on-premises locations and also between virtual networks within Azure Stack (VNet-to-VNet). Each virtual network can have only one Virtual Network gateway.

Here are some considerations regarding VPN Gateways in Azure Stack:

• VPN from Azure Stack-to-Azure Stack and Azure–to-Azure Stack is supported with public IPs.
• Azure VPNs are not officially supported through NAT.
• ExpressRoute is not supported however Site-to-Site VPN is available.
• The VPN Gateway will not support L3 and GRE gateway in this Azure Stack release.
• Most recent Azure features released will not be available in the current release of Azure Stack (for example, network peering and network diagnostics).

Azure Stack Networks vs Public Azure Networks

Here’s a comparison chart for the networks between Azure Stack and Public Azure:

Introducing iDNS in Azure Stack

Since Azure Stack could be represented as a segment network separated from the public cloud and/or the on-premises environment, Microsoft included the Internal DNS feature (iDNS) feature. This concept originally appeared in Windows Server 2016 in the Software Defined Network (SDN) definition.

In Azure Stack, iDNS allows you to resolve external DNS names and internal virtual network names. Customers can still bring their own DNS servers into the environment, but if the requirement is just to resolve Internet DNS names and connect to other virtual machines in the network, iDNS can do the trick.

It is a cost-effective solution since you don’t need to deploy your DNS servers and additionally you don’t have to expose these servers to malicious attacks.

Azure Stack Storage

Azure Stack storage was built with the same principles than Windows Server 2016 Software-Defined-Storage (SDS). The storage components we can find in Azure Stack are: Blobs (what we usually use as the OS or data disks in VMs), Tables (NoSQL key-attribute data store for fast access to large quantities of data), and Queues (letting cloud software communicate via messages).

Here are some considerations about Storage in Azure Stack:

• A single region is supported.
• Locally Redundant Storage (LRS) is the only supported availability model.
• Premium Storage is supported at the API level (including template support for Premium Storage).
• The transaction data in storage metrics does not differentiate internal or external network bandwidth.
• The transaction data in storage metrics does not include virtual machine access to the mounted disks.

Initially it was expected that Azure Stack in GA was not going to include Storage metrics features available, but eventually it ended up included. Here’s the reference about the cmdlets available in Azure Stack related to storage and includes metrics: “AzureRM.AzureStackStorage”.

Azure Stack Storage vs Public Azure Storage

Here’s a comparison chart for the storage between Azure Stack and Public Azure:

Azure Stack Marketplace

The Marketplace is virtual store and portal containing certified, open source, and community software applications, developer services, and data which are pre-configured to run in Microsoft Azure Stack.

The ideal scenario is to integrate the Azure Stack Marketplace with Azure’s Marketplace, so available and compatible items for Azure Stack are available and can be implemented easily.

There’s a list of Marketplace items available for Azure Stack that is updated constantly: “Azure marketplace items available for Azure Stack”.

The Azure Stack Marketplace also has some considerations:

• It should not be expected that all Azure Marketplace Images and VMExtensions will be present in Azure Stack.
• Azure Stack will have a syndication service which will allow for specified images and VMExtensions to be downloaded and included in Azure Stack deployments.
• Each ISV needs to mark images as “Azure Stack Capable“. If you wish to deploy a solution with dependencies, it should be developed and deployed as a “self-contained” image.

In the following article, I’ll make a review about the PaaS services available in Azure Stack, their considerations and the comparison between public Azure.

So, the long-awaited Azure Stack finally has been released to the Integrated Systems, meaning the shipments will start in September. And what to do now?

Do I know if Azure Stack will be a good investment? What are the things I can and I cannot do with Azure Stack? Should I connect my platform with Azure or maintain a disconnected mode?

I decided to create a few posts with some clarity about what we can achieve with Azure Stack, different licensing models, features that will be available, how the IT model could change with this concept, and some other topics.

Before getting into specific functionalities, a good start is clarifying regarding the options available to connect Azure Stack to your Azure instance and the alternative on having it completely disconnected.

Azure Stack Connected to Azure – Consumption Model

Like we mentioned, Azure Stack main purpose is to act as an extension of Azure in your datacenter, therefore this “pay-as-you-use” model should be the familiar approach for all customers already using Azure. In the Consumption Model, Azure Stack licensing is just like public Azure.

Here are the important takeaways for the consumption model:

• Fee for consumption: only pay for services running on Azure Stack.
• No upfront licensing fees: don’t pay until you use the service. The only upfront cost will be the hardware from the Integrated System.
• Unified billing, one invoice with public Azure.
• Available in Enterprise Agreement (EA) and Cloud Solution Provider (CSP) channels.
• Enterprise Agreement (EA) deployments of Azure Stack consume same monetary commit as public Azure.

“Monetary commitment” is a concept associated with Enterprise Agreements (EA), when there’s a certain amount of money allocated in the contract to use in Azure. That money can be used in different type of services as credits, and consumed with the “as-you-use” concept.

Azure Stack Resource Providers (RP)

Before talking about SQL and MySQL functions in Azure Stack, it is important to understand what a Resource Provider (RP) represents in Azure. This concept encapsulates an Azure service that supplies the resources you can deploy and manage through Azure Resource Manager (ARM).

Basically a Resource Provider is responsible for creating resources of a specific type, for example the RP “Microsoft.Compute” supplies the virtual machine resource; the RP “Microsoft.Storage” supplies the storage account resource. Each resource provider offers operations for working with the resources that are deployed.

This is a simple diagram to understand where the Azure Resource Providers interact:

Azure Stack SQL and MySQL Resource Providers

Azure Stack can provide SQL and MySQL in the Platform as a Service (PaaS) mode by using their Resource Providers. Here are some important notes to consider regarding using SQL and MySQL as a PaaS service:

• With these platforms customers must provide the licenses (for MySQL licenses only apply with certain specific products).
• Each Resource Provider require two virtual machines: One VM dedicated for RP services and the second VM with SQL / MySQL platform installed (Hosting Service).
• No cost is associated for the virtual machines running the RP.
• The costs that apply are for the data storage (blob) and the SQL / MySQL VM.
• These Resource Providers are not available by default, you need to deploy and enable them.

For more information about the RP and how to install them in Azure Stack, review the following links:

“Use SQL databases on Microsoft Azure Stack”

“Use MySQL databases on Microsoft Azure Stack”

As you can see, these PaaS services are presented by virtual machines, therefore you should actually consider these Resource Providers as Infrastructure as a Service (IaaS) platform.

SQL Resource Provider in Azure Stack

Azure Stack Consumption Costs

The Azure Stack pricing was released some time back and I reviewed earlier in my blog, here’s a quick table to put some numbers:

Regarding Windows Server virtual machines, summarizing the scenario for bring your own license in Azure Stack:

• In case you are using an existing license, you will only pay for the “Base Virtual Machine” fee ($0.008/vCPU/hour).
• If you don’t have (or don’t want to use) existing licenses, you will pay for the “Windows Server virtual machine” fee ($0.046/vCPU/hour).
• SQL Server virtual machines (working in PaaS mode as a RP or not), always require to have an existing license.

Azure Stack Disconnected – Capacity Model

Some months prior of the announcement I reflected some rumors that were pointing to an Azure Stack offering with a fixed-fee approach, disconnected from public Azure. And eventually came true, Microsoft confirmed that the “Capacity Model” will not have consumption based billing, but a fixed cost.

Here are some important notes about the capacity or disconnected model for Azure Stack:

• No usage metering or connection to public Azure.
• Only accessible through Enterprise Agreements (EA).
• Fixed fee, annual subscription: based on number of physical cores with unlimited IaaS for licensed cores.
• It will not provide perpetual rights, only annual agreements based on EA.
• Separate transaction from Azure: cannot use monetary commit, it will have different billing.
• You can only use existing Windows and SQL licenses in this scenario.
• There will be two packages available to acquire using the capacity model: App Service package and IaaS package.

Microsoft prefers and recommends to avoid the capacity model and use Azure Stack on consumption basis. No surprise there.

App Service Package in Azure Stack Capacity Model

This option in the capacity model has the following features:

• Includes App Service (Web Apps, Mobile Apps, API Apps, and Functions), Base VM and Azure Storage.
• All physical cores must be licensed in the appliance.
• Windows Server and SQL requires existing licenses (bring your own license).
• The cost is $400/core/year.

IaaS Package in Azure Stack Capacity Model

This is a reduced version of the App Service package, has the following features:

• Includes Base VM and Azure Storage.
• All physical cores must be licensed in the appliance.
• Windows Server and SQL requires existing licenses (bring your own license).
• The cost is $144/year.

Comparison Table as a Summary

To close this article, here’s a table summarizing the features and characteristics for the Azure Stack consumption and capacity models:

To review a more detailed view about services and features for the Azure Stack Consumption Model, check my previous article: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More”.

As we reviewed earlier, Azure Stack will be available only via Enterprise Agreement (EA) and from the Cloud Service Providers (CSP) program. Microsoft will incorporate the pay-as-you-use model with Azure Stack, including capabilities to bring-your-own-license for Windows Server and SQL and is intended to provide unified billing from customers using Azure.

For more information regarding pricing, licenses use and other topics in Azure Stack, review the following article: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed”.

I also have to thank, again, Mary Jo Foley from ZDNET for referencing my blog in her article. To check Mary Jo’s great editorial about this Azure Stack release, please check this link: “Microsoft Azure Stack is ready to order from Dell EMC, HPE, and Lenovo”.

Azure Stack Updated Features and Roadmap

In the General Availability (GA) release, which is now called “Initial Availability” according to Microsoft, there are some changes and updates regarding previous announcements about features available (for more information: “Azure Stack: What should we expect for General Availability (GA)”).

Here are the updated tables of features and functionalities available for Azure Stack in this release (highlighting the updates from previous table):

It is expected that the Azure Stack Market Place will start increasing the products available within the next couple of months, therefore more and more functionalities will be available for Azure Stack.

Also, according to Microsoft, the future releases this year and the 2018 will have some updates as well (highlighting the updates from previous table):

Azure Stack Service Bus functionality is not yet available and has not been confirmed when it will be released. In the public version of Azure the Service Bus is a frequently required feature for developers since it allows a simpler way for two or more applications interact with each other using namespaces.

Here’s a graphical representation of Azure Stack scale:

For more details on Azure Stack instances and scales, check an earlier article of mine: “Yada Yada Cloud: Azure Stack, what’s your story?”

Azure Stack Availability

Azure Stack will be available in 46 countries in total from its release. The initial announcement I made earlier about the country availability stays pretty much the same (Vietnam is added)

• Americas: Argentina, Brazil, Canada, Chile, Colombia, Mexico, Peru, United States, Uruguay
• EMEA: Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Israel, Italy, Luxembourg, Monaco, Morocco, Netherlands, Norway, Portugal, Russia, Saudi Arabia, South Africa, Spain, Sweden, Switzerland, Turkey, United Arab Emirates, United Kingdom
• APAC: Australia, Hong Kong, India, Indonesia, Japan, Korea, Malaysia, New Zealand, Philippines, Singapore, Taiwan, Thailand, Vietnam

Bolded countries are the ones in which systems will be available from all three partners, Dell EMC, HPE, and Lenovo. Other countries have availability from at least one partner.

Other integrated systems: Azure Stack from Cisco will be available some time during 2017 and Huawei will release their version in Q1 of 2018.

Azure Stack Development Kit (Single-Node)

The Azure Stack Development Kit (previously named Proof-of-Concept (POC) mode) represents the single-node deployment model for this appliance. Intended only for non-production environments (it will not be supported by Microsoft) provides the alternative to deploy Azure Stack on existing hardware in a single-server mode.

It is a great addition to any customer looking to explore into Azure Stack, as a much needed test environment for those organizations already owning Azure Stack that need to validate an application or deployment before moving into production. The best part of the story, as I reviewed a while back, the Azure Stack Development Kit is completely free and ready for download.

Azure Stack Development Kit logical architecture diagram

More information: Microsoft Azure Stack Development Kit architecture

The Azure Stack Development Kit in this release has also some updates and modifications. Some of the most important ones are the graphical interface for the installation process instead of a sometimes tedious PowerShell screen; and they also added the support of Microsoft Accounts for the deployment.

Here’s what’s new in Azure Stack Development Kit:

• You can now use CLI 2.0 to manage Azure Stack resources from a command line on popular operating systems.
• DSV2 virtual machine sizes enable template portability between Azure and Azure Stack.
• Cloud operators can preview the capacity management experience within the capacity management blade.
• You can now use the Azure Diagnostics extension to gather diagnostic data from your virtual machines. Capturing this data is useful when analyzing workload performance and for investigating issues.
• A new deployment experience replaces previous scripted steps for deployment. The new deployment experience provides a common graphical interface through the entire deployment lifecycle.
• Microsoft Accounts (MSA) are now supported during deployment.
• Multi-Factor Authentication (MFA) is now supported during deployment. Previously, MFA must be disabled during deployment.

What else is New for Azure Stack?

The Azure Stack Market Place library will start increasing for sure, so we should expect updates on functionalities and capabilities at any moment. Going in that direction, Microsoft also released/updated two important factors:

1. Azure Stack documentation: Everything that is needed to start can be found here. Pre-requisites and installation process for Azure Stack Development Kit; how to enable features in Azure Stack; etc etc.

No product is finalized until the vendor released the documentation to support it, right?

2. System Center Management Pack for Microsoft Azure Stack: Even though with the REST APIs provided by Azure Stack you can monitor the platform using tools like Nagios, System Center Operations Manager (SCOM) still represents one of the more robust and reliable ways to monitor Microsoft products. This MP will adapt easily to the Azure Stack environment you deploy.

The System Center Management Pack for Microsoft Azure Stack enables you to monitor the availability of the Azure Stack Infrastructure. The management pack runs on a specified resource pool and then uses various Azure Stack APIs to remotely discover and collect instrumentation information, such as a Deployments, Regions and Alerts. The Management Pack for Microsoft Azure Stack provides no functionality on import. For each Microsoft Azure Stack deployment you want to monitor, you must configure discovery and monitoring by first using the Add Deployment wizard in the administration section of the OM Console.

3. Azure Stack Tools on GitHub: This community repository is known for the contributions made for other Azure Stack users around the world with sample scripts and processes.

Already has some contributions available after just a few hours from the Azure Stack release.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Azure Stack Final Pricing

So, the numbers are here in the pay-as-you-go model (or pay-as-you-use as MS is referring in Azure Stack). And as we all thought, Azure Stack consumption model is going to be cheaper for obvious reasons than Azure. Here are the details:

It is important to remember that after GA there will be available more features than the ones listed in the table above. For more information, check the following article: “Azure Stack: What should we expect for General Availability (GA)”.

Post-GA features:

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Azure Stack and Licensing

As we reviewed earlier, you will be able to use existing licenses in your organization to apply in Azure Stack. For example, if you already own Windows Server or SQL licenses from a Windows Enterprise Agreement (EA), can be applied to the Azure Stack environment.

Here’s a quick reference about using a Windows Server virtual machine in Azure Stack:

• In case you are using an existing license, you will only pay for the “Base Virtual Machine” fee ($0.008/vCPU/hour).
• If you don’t have (or don’t want to use) existing licenses, you will pay for the “Windows Server virtual machine” fee ($0.046/vCPU/hour).

For SQL Server, the “bring your own license” will be the only option available.

Azure Stack Disconnected and Fixed Fee Model

One of the rumors I shared a couple of months ago was the “disconnected” and fixed fee option for Azure Stack, to include those scenarios with no public internet connection to be integrated with the Azure Stack platform. And the answer is yes, the disconnected and fixed fee option will be available.

Licensing for this model will be based on physical cores available in Azure Stack and with a fixed fee per year basis.

Here are some important takeaways from this model:

• You can only use existing Windows and SQL licenses in this scenario.
• It is only available for purchase with an Enterprise Agreement (EA).

Azure Stack Support Model

It was also confirmed a previous announcement I made regarding the support model available in Azure Stack considering the Integrated Systems. There will be two support contracts: One for Microsoft and one for the Integrated System providing the hardware.

• When support assistance is needed, customers can call the Integrated System or Microsoft. A consistent support experience no matter who you contact for support will be delivered.
• Each Integrated System will have their own support contract, different from Microsoft’s support.
• The Integrated Systems will have different levels of support, as they have right now for their hardware solutions.
• Coordinated escalation and resolution process.
• Hardware solutions support will be provided by the Integrated System and the cloud services support delivered by Microsoft.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Azure Stack confirmed release date has not yet been confirmed, the mid-2017 has arrived but there are strong rumors indicating that the definitive release date for Azure Stack will be August 2017. While we get closer to the GA deadline, Microsoft confirmed two new hardware OEM Vendors: Avanade and Huawei.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

We already know that Azure Stack after GA will be released with specific hardware vendors, the Integrated Systems. Azure Stack will only be able to obtain from these Integrated Systems: Dell EMC, HPE, Lenovo, Cisco (available at some point after GA); and now adding Avanade and Huawei.

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

For Avanade, there’s no specific date for when the Azure Stack solution will be available. Huawei commented that they expect to have Azure Stack commercially available in Q1 of 2018.

Avanade, the Partner Within

There’s no big surprise on Avanade being one of the hardware OEM vendors with an Azure Stack solution, this company based in Seattle started in 2000 as a joint-venture between Microsoft and Accenture.

Avanade, besides the fact that is co-owned by Microsoft, is one of the strongest Azure partners. Providing Azure migration and managed services, but also an Azure Stack lab in their datacenters where customers can access the platform.

Recently announced: “Avanade is excited to reveal the upcoming availability of a complete Microsoft Azure Stack solution from a single provider. This fully managed solution will offer an Azure-consistent experience at a competitive price point, available in a consumption-based or pre-purchased pricing model.”

The hardware to be provided from Avanade is not disclosed if it will come from a specific manufacturer, but they added: “We have partnered with a leading datacenter hardware solution provider and a leading network equipment company to manufacture the hardware”.

Additionally the hardware will include “a space-constrained turnkey Azure Stack with four nodes in 6U (including switching and management host), the second is a density/capacity optimized azure stack with four nodes in in 8U (including switching and management host). Both are expandable to the full 12 node maximum.”

As we reviewed earlier in “Azure Stack: What should we expect for General Availability (GA)”, 12 nodes will be the maximum per Azure Stack scale. The scale can expand in what’s called Azure Stack Region.

Azure Stack Scale = Scale Unit x Scale Units per Region x Number of Regions

For more information about Azure Stack scales and regions, review my article: “Yada Yada Cloud: Azure Stack, what’s your story?”

Remembering Fujitsu in Azure and Why Huawei Won’t Be the Same

Back in 2010 in Microsoft’s World Partner Conference (WPC) the first concept of Azure Stack was introduced, it was called “Azure Appliance”, and it was going to have the same basic concepts including the Integrated Systems. At that time the OEM vendors selected were: Dell (before the EMC acquisition of course), HP and surprisingly Fujitsu.

This Azure Appliance had in eBay as one of the early adopters, one of their public websites was being hosted in this platform. It was also mentioned there were around 20 companies beta testing this appliance.

The Azure Appliance continued the development stage but no one knew exactly what happened to Fujitsu when Azure Stack was finally announced. Several assumed that Fujitsu’s participation in the server market was not substantial and the investment they needed to perform was not going to provide much return.

This Fujitsu scenario won’t be the case for Huawei, this Chinese company is starting to get a strong presence in the servers’ market share. Gartner informed that in Q4 of 2016, Huawei was #3 in total servers’ shipments worldwide, beating Lenovo.

Originally a networking company, focused in providing enterprise-scale but cost-effective solutions competing side by side with Cisco, now Huawei stated that 10% of their revenue comes from innovative technologies.

In this path of innovative technologies, Huawei recently announced their participation in the Azure Stack market. The Chinese company shared the solution components: “The jointly-engineered solution – based on Huawei’s latest-generation FusionServer and CloudEngine switches – uniquely offers built-in integration between Huawei’s eSight management software and Azure Stack, enabling customers to implement 360-degree management on hardware devices”.

FusionServer and CloudEngine technologies represent one of the main assets for Huawei to take market share for the datacenters in this new hybrid world. Regarding the Azure Stack solution, they also added: “Huawei’s hybrid cloud infrastructure features FusionServer, which delivers integrated high-density architecture, large storage capacity, and highly reliable management. It is an ideal choice for cloud computing, big data, and HPC applications. Huawei CloudEngine switch provides high-density 10G access and up to 100G interconnect, enabling networks of high performance and reliability, large cache, and low latency.”

For those thinking that there was nothing else coming up for Azure Stack TP3 and we only needed to wait for GA, Microsoft released an update for the preview which enables Azure PaaS services. These new services enabled are: Azure App Service and Azure Functions.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Additionally, SQL and MySQL platforms receive updated versions, there were some minor but important changes in the deploy process (see below for details) and some other stability improvements.

Here’s a summary about Azure Stack TP3 refresh (for what was new with the initial TP3 release, please review the following: “Azure Stack TP3 is Here! And with Pricing”):

• Azure App Service enabled. This service allows a true DevOps environment to build and manage hosted applications in the Azure environment. This update includes the possibility to include web apps, API apps and mobile apps.

• Azure Functions. This is the serverless scenario for your apps and services, where you don’t need to worry to deploy virtual machines, network or any other component, the Azure platform provides what’s necessary for organizations to run the code for your application. These functions have several benefits, like running a specific code when an event is triggered or in a specific time frame.

• Multi-tenancy support. You can use ADFS to trust other domains and provide service to those domains.

• Updated versions for SQL and MySQL database services.

• Reduced RAM consumption for IaaS. The VMs used in an IaaS scenario will consume less AM in the platform.

• For SQL VMs in IaaS, additional management functions and SQL metrics.

• Updated deployment process for Azure Stack TP3. More details below.

Updated Deployment Process and Some Gotchas

The overall process to deploy Azure Stack stays similar as the initial Technical Previews (more info here: “Yada Yada Cloud: Azure Stack, what’s your story?”); but there are a few additional steps to consider and avoid deployment issues.

Here’s an updated step-by-step process to deploy Azure Stack TP3 with the new update:

1. Confirm that you meet the requirements. You can also use the Deployment Checker for Azure Stack to confirm your requirements.

2. Download Azure Stack Technical Preview, accessing this link.

3. If you already have Azure Stack implemented, remove all partitions of the four drives and start with a clean installation.

4. Extract Azure Stack files and copy the CloudBuilder.vhdx file into the C:\ drive.

5. Download the Azure Stack TP3 support files using PowerShell script.

# Variables
$Uri = ‘https://raw.githubusercontent.com/Azure/AzureStack-Tools/master/Deployment/’
$LocalPath = ‘c:\AzureStack_SupportFiles’
# Create folder
New-Item $LocalPath -type directory
# Download files
( ‘BootMenuNoKVM.ps1’, ‘PrepareBootFromVHD.ps1’, ‘Unattend.xml’, ‘unattend_NoKVM.xml’) | foreach { Invoke-WebRequest ($uri + $_) -OutFile ($LocalPath + ‘\’ + $_) }

6. Run the PrepareBootFromVHD.ps1 script (confirm required parameters). Reboot will be required since the machine will boot into the VHDX.

.\PrepareBootFromVHD.ps1 -CloudBuilderDiskPath C:\CloudBuilder.vhdx –ApplyUnattend

7. Execute the “Install Azure Stack POC” PowerShell. Here’s one of the examples using one Azure Active Directory:

cd C:\CloudDeployment\Setup
$adminpass = ConvertTo-SecureString “<LOCAL ADMIN PASSWORD>” -AsPlainText -Force
$aadpass = ConvertTo-SecureString “<AAD GLOBAL ADMIN ACCOUNT PASSWORD>” -AsPlainText -Force
$aadcred = New-Object System.Management.Automation.PSCredential (“<AAD GLOBAL ADMIN ACCOUNT>”, $aadpass)
.\InstallAzureStackPOC.ps1 -AdminPassword $adminpass -InfraAzureDirectoryTenantAdminCredential $aadcred

8. Activate administrator and tenant portals. This is an important and new step, if the portals are not activated, only the administrator will be able to use the portals.

For the administrator portal, navigate to https://adminportal.local.azurestack.external/guest/signup, read the information, and then click Accept.

For the tenant portal, navigate to https://portal.local.azurestack.external/guest/signup, read the information, and then click Accept.

9. Connect to the Azure Stack POC using RDP or VPN, following the guideline: “Connect to Azure Stack”.

One could guess this latest update will be the last one before General Availability, but without confirmed date of release yet besides the mid-2017 comment, it would not surprise if we can have additional features showing up in TP3 and closer to GA functionalities. Microsoft stated that there will not be a Azure Stack TP4 release.

For more information about Azure Stack GA features: “Azure Stack: What should we expect for General Availability (GA)”.

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Azure Stack GA is right around the corner, scheduled for mid-2017 this Azure appliance is promising to change the way organizations think about hybrid cloud deployments or, even more, how datacenters are conceived. I wanted to recap some of the most important topics and features we should expect when the Azure Stack GA version is released.

After TP3 release, Microsoft confirmed that there’s not going to be an Azure Stack TP4.

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

In my previous post: “Azure Stack TP3 is Here! And with Pricing” there’s an overview of the latest Technical Preview as well as some details about Azure Stack pricing.

Azure Stack Features and Capabilities in GA

There are still some components that might have some updates until the GA release, but here’s the list of features we can find available in Azure Stack:

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Regarding Azure Stack Scale, even though the minimum will start from 4 servers (no matter the Integrated System selected), customers can add a server at the time if required. For a definition regarding Azure Stack scale, check my post: “Yada Yada Cloud: Azure Stack, what’s your story?”

Pulsant, cloud provider located based in the UK, will start offering starting April 2017, Azure Stack services to their customers, using TP3. Pulsant is one of the few partners available globally to reach customers with Azure Stack before the GA release.

Pulsant has been working with Dell EMC to work on this initiative where partners will give Azure Stack some real world experiences in the immediate future.

Azure Stack Features and Capabilities Post-GA

Also, Microsoft has confirmed there are going to be another features that will become available with upcoming updates to Azure Stack when GA is released. There are still some definitions to be made regarding functionalities/capabilities and dates when those will be available, here’s the list of what we know so far:

Several details available regarding Azure Stack updates and the workflow associated are available in: “Azure Stack TP3 is Here! And with Pricing”

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Azure Stack Licensing and Support

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

As I mentioned before, Azure Stack licensing will not have upfront costs, customers will only be paying for the hardware they are acquiring from the Integrated System (Dell EMC, HPE, Lenovo and available sometime after GA, Cisco).

Update: Huawei and Avanade were added as hardware OEM vendors, for more information: “Azure Stack Welcomes Two New Hardware OEM Vendors“.

The Azure Stack costs will be, as we already described, based on the consumption. It will only be available with Enterprise Agreements (EA) and partners from the CSP (Cloud Solution Provider) program. Review the “Azure Stack TP3 is Here! And with Pricing” article since it provides several key definitions regarding licensing and costs.

Regarding Azure Stack support, here are the important takeaways:

• Each Integrated System will have their own support contract, different from Microsoft’s support.
• The Integrated Systems will have different levels of support, as they have right now for their hardware solutions.
• When support assistance is needed, customers can call the Integrated System or Microsoft. A consistent support experience no matter who you contact for support will be delivered.
• Coordinated escalation and resolution process.
• Hardware solutions support will be provided by the Integrated System and the cloud services support delivered by Microsoft.

Questions Unanswered (at the moment) about Azure Stack

I’ve talked with customers recently regarding Azure Stack and they’ve raised some interesting questions which don’t have an answer yet.

Here’s a summary of concepts and questions to be reviewed later on:

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Is there going to be a “disconnected mode” with Azure Stack?

As we already know, Azure Stack will be integrated with Azure, and Microsoft will offer a unified billing for the services consumed in these two platforms. But what happens for scenarios where the Azure Stack instance needs to be maintained separated and disconnected from the public cloud?

It could be related about security or compliance issues, I’m pretty sure there are going to be several organizations with similar requirements.

A “fixed fee” billing will be offered?

For those organizations with a more static consumption required in the on-premises section of this hybrid cloud model with Azure Stack, can Microsoft offer a fixed fee cost assuming the hardware will stay the same and/or with a scalable fixed fee billing when more hardware is needed?

In this scenario, customers can use all the services they need within Azure Stack without additional costs besides the fixed fee agreed with Microsoft. This scenario could easily apply for the disconnected mode.

Azure Stack Connector for Windows Azure Pack (WAP)

As I reviewed in my initial post about Azure Stack, WAP (Windows Azure Pack) will be maintained as an offer available for customers, which will include an Azure Stack connector. Even more, Microsoft is encouraging customers to obtain WAP or CPS (Cloud Platform Systems) services and then extend those capabilities with Azure Stack.

Several capabilities between these platforms will be overlapping, and we can assume at some point (even though Microsoft has not confirmed this) that WAP and CPS will be replaced entirely by Azure Stack.

In the meanwhile for customers with WAP, Microsoft has released the Azure Stack connector for TP3. Using this connector, customers can enable access between Azure Stack and Windows Azure Pack on-premises platform.

WAP tenants can perform the following activities with Azure Stack:

• Browse resources
• Examine configuration values
• Stop or start a virtual machine
• Connect to a virtual machine through Remote Desktop Protocol (RDP)
• Create and manage checkpoints
• Delete virtual machines and virtual networks

WAP connector for Azure Stack architecture

For more information, review the following article: “Manage Windows Azure Pack virtual machines from Azure Stack”.

Update: WAP connector will be available in Q4 of 2017 according to Microsoft’s roadmap. Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

When Can We Hear More Updates?

There is no formal date for GA release yet, but we can assume we could get some updates regarding dates, functionalities and other topics regarding Azure Stack in Microsoft Build 2017, May 10 – 12 in Seattle, WA.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

This developer conference has been, in the last few years, focusing on cloud capabilities, and if you check Microsoft Build 2016 sessions, you will see topics like building apps for Azure, details about Azure Functions (serverless computing, capability also available in Azure Stack at GA), several tracks related to IaaS and PaaS.

It’s fair to assume we will receive some updates about Azure Stack in this event.

Previous post: [Yada Yada Cloud] Choosing from Exchange On-prem vs Exchange Online: The never ending story? – Part I

Continuing this two set article about Exchange on-prem vs Exchange Online, I’m going to cover the most common conversation points from #5 to #8 and one bonus track.

As in my previous post, I’ll review the Exchange on-prem, Exchange Online and my personal perspective about each topic.

Let’s start.

5. The data security concerns about cloud and Microsoft being the “big brother”

This is a recurrent topic when people from security, legal and auditing are present in the discussion. There are organizations with sensitive and confidential information that are fairly concern about their data being stored and managed by people outside their company.

The Zero Wing reference for “All your base are belong to us“

Let’s review about the Exchange On-Prem view:

• At any moment, Microsoft can say yes to a higher institution and provide our company’s private data.

Microsoft and the Ireland case represents one of the most important affairs regarding digital security. The case began in December 2013 when a judge in New York issued a warrant asking Microsoft to produce all emails and other private information associated with a customer account. This account was involved in a criminal narcotics investigation.

This user data was being stored in one of Microsoft datacenters in Ireland. Microsoft produced account information kept on servers in the United States, but refused to produce any data stored on servers in Ireland, arguing the government’s warrant does not apply in foreign countries.

There was a lot of back and forward from 2013 to 2016, but ultimately Second Circuit Appeals Court ruled in favor of Microsoft and against the US Government, confirming that the Redmond Company did not have to provide the data (here’s the full order available in PDF).

• Microsoft could be selling our private data to other companies, for example data they can obtain from Delve to target users from their preferences to advertise a product.
• We can use the capabilities for e-mail security using Exchange Online Protection (EOP) service, without the need to take all of our data to the cloud.
• There are some scenarios, due to the sensitive information, the emails need to be isolated as much as possible for security reasons.
• If Hillary decided to go on-prem and worked for her, why not us?!

During election times in 2016, as anyone can imagine, this was a recurrent topic to discuss when an Office 365 migration meeting occurred. The full and detailed explanation about this affaire can be reviewed in Wikipedia: “Hillary Clinton email controversy”.

As a quick summary of the incident, Hillary Clinton allegedly used a personal email server (Exchange 2010 platform located in a datacenter in their hometown in Chappaqua, New York) to send and receive classified information from the Secretary of the State, when she was acting as the leading role in that department.

During the several investigations over the years, Hillary Clinton handed to the authorities over 30,000 emails as printed versions to confirm the content was not classified. They also turned over the servers (two Dell PowerEdge 1950 Servers, one for Exchange 2010 and one server for the BlackBerry Enterprise Server) and their drives, which the authorities later mentioned that those disk drives were erased before properly received.

The FBI investigation concluded stating that Clinton was “extremely careless” in handling her email system but recommended that no charges be filed against her.

This is a screenshot of the still active (to this date) Exchange 2010 OWA for Clinton’s email: https://mail.clintonemail.com/owa

Taking a look to the Exchange Online approach:

• Microsoft can only provide information to the government in special exceptions, when there’s a risk of national security.

Have you ever wondered how many times did Microsoft disclose customers’ information to the government? There’s the Microsoft Law Enforcement Requests Report, which is updated twice a year with a complete detail on how many cases and type of responses to the government Microsoft had to provide.

In this global support dashboard, only 2.65% of the cases content was disclosed (majority of disclosed data represents just information about the subscriber). You will also see in the detailed XLS than can be exported, almost 95% of the countries involved had 0 cases where they needed to disclose content.

Microsoft also created the concept of “customer lockbox”, which includes the customer in the decision workflow, if possible by the country’s law, when a request for disclosing information is placed. This way the customer has a vote to decide if they want the information to be disclosed or not.

• The security standards and processes implemented by Microsoft to safeguard the servers and information is usually much higher than most of the companies available.

Providing the high level of security that Microsoft provides by default in Office 365 can be extremely expensive for most organizations and challenging to maintain. And if a security breach occurs, the entire organization loses faith in IT.

Here are some of the links and references available from Microsoft where you can deep dive on the security and legal concerns about data in the cloud, regulatory compliance, etc:

Office 365 Trust Center

https://www.microsoft.com/en-us/TrustCenter/CloudServices/Office-365

Compliant with Global Standards

Standards Conformance & Regulatory Compliance
Security, Audits, and Certifications
Use Office 365 to help comply with legal, regulatory, and organizational compliance requirements
Top 10 compliance areas of Office 365
FISMA/FedRAMP: Frequently asked questions
EU Model Clauses—frequently asked questions
Office 365 & Microsoft Dynamics CRM Online HIPAA/HITECH frequently asked questions

Protected by Lawyers

Law Enforcement Requests Report
U.S. National Security Orders Report
Content Removal Requests Report
Office 365 Privacy Statement

Tested by Security Experts

CodeChat 003 – Ben Godard (Office 365 Red Team)
From Inside the Cloud: What does Microsoft do to prepare for emerging security threats to Office 365?
Security, Audits, and Certifications (including details on Office 365 ISO certifications)

Trust through Transparency

Auditing and Reporting in Office 365
​Controlling Access to Office 365 and Protecting Content on Devices
Data Encryption Technologies in Office 365
​Data Resiliency in Office 365
​Defending Office 365 Against Denial of Service Attacks
​Financial Services Compliance in Microsoft’s Cloud Services
Office 365 Administrative Access Controls
Office 365 Customer Security Considerations
Office 365 Security Incident Management
Privacy in Office 365
Tenant Isolation in Office 365

As my personal opinion on the Office 365 security matter:

The security conversation can also be difficult to have when we have organizations or just a few people with very sensitive information. It can also transform into a philosophical discussion on trying to define who owns the information used by the people and their productivity; or even going further, what are the boundaries of “owning” the information.

Mostly the discussions go around “How secure I feel by allowing Microsoft to store my information?” and “How much do I trust in Microsoft and the use they can have with my data?”

The regulations and standards established as baselines for the cloud services are clear, and the probability on having Microsoft or the government trying to obtain maliciously our private data will always be lower than the risk on having any malicious internal or external party trying to access our information. Considering also the large investment a company needs to make to obtain a similar type of security from what Microsoft offers in an on-premises environment.

6. The challenges of remote locations and inconsistent Internet connections.

The remote locations and slow internet connections discussion can appear in those organizations distributed in large regions. It may not seem all that common in the US since the connections available in most of the country are fairly acceptable, but there are other countries and continents that can experience these issues.

Let’s review about the Exchange on-premises point of view:

• Staying with Exchange on-premises, in case of troubleshooting, we remove the Internet connection as a variable.
• The information handled within email it’s highly confidential, we cannot expose that platform to public connections.
• We have sensitive users that are highly dependent on Outlook and sometimes they don’t have good connections. Outlook connected through Internet requires a good amount of bandwidth.

As we know, Outlook Clients can be an intensive application for CPU, memory and network bandwidth. For the latter, there are several variables than can affect the experience of a “slow network response”. Microsoft developed, several years back, the “Exchange Client Network Bandwidth Calculator” to size properly the necessary bandwidth in one or more locations.

Uses as reference Outlook 2007 and Outlook 2010, they haven’t updated that yet, but the network behavior hasn’t changed all that much in the later versions.

Let’s take a look about an Office 365 example using this bandwidth calculator. Here are the parameters for a “Very Heavy” Outlook client user:

The one item to observe is the “Avg Mailbox Size (GB)”. I’m using the 100GB scenario to include the worst case on all of these users with a full mailbox for O365, which won’t be the case in almost any company. This parameter is actually looking for the OST resyncs to be performed from these clients.

As an exercise, I’m also using the “Medium” profile (average mailbox usage of 20GB, the rest stays the same) and “Heavy” profile (average mailbox usage of 50GB, same for the rest).

Here’s the bandwidth calculation of a mixture scenario: 300 Outlook 2010 (cached mode enabled); 50 Mac users; 50 Outlook 2007; 50 users with OWA; and 300 mobile clients (we can fairly assume most of the Outlook users will be using a mobile at the same time).

For the “Very Heavy” scenario, which is the worst case, we have the 37.51Mbits/sec bandwidth requirement for download.

Going to the other side and reviewing the Exchange Online perspective:

• For the remote places, unless you have the mailbox server in that remote location, your problem will stay the same. The Outlook client will use Internet (Outlook Anywhere or VPN) to connect. The internet connection will always be part of the variables.
• Outlook connectivity might be a problem in some scenarios, but OWA and ActiveSync for mobiles can definitely work with slow connections.
• Microsoft offers several tools and guidelines to assess and optimize client connectivity for Office 365. One of the most valuable ones is “Office 365 Network Analysis Tool” (link included for North America, but there are also available for Europe and one for Asia-Pacific).

Here are some other tools and guidance material available to consider:

• “Best practices for using Office 365 on a slow network”
• “Network and migration planning for Office 365”
• “Network planning and performance tuning for Office 365”
• “Generate Message Profile for the Exchange Calculators”: PowerShell script (updated frequently) to obtain detailed information about the Exchange environment and usage within the organization, works for Exchange 2010, 2013 and 2016. For more information about using this script: “Generating user message profiles for use with the Exchange Calculators”.

But what happens if you get the “My network connection is suitable and I already checked all the best practices for Office 365, but my Outlook connection experience is still poor”, here are a couple of tips to troubleshoot and validate Outlook issues and network problems.

• If you are suspicious on inconsistencies within the applications, “Microsoft Office Configuration Analyzer Tool” (OffCAT) will provide you on any problem related about application configuration, patch level or even data corruption.
• Use “Remote Connectivity Analyzer” to validate issues within an existing Exchange environment, for troubleshooting Outlook issues, certificates inconsistencies, account setup, etc.
• There’s a long history of Outlook add-ons causing issues within the application (iCloud add-on with Outlook 2013 for example). Try to disable all of them for proper testing.
• Graphics acceleration for Outlook 2013 and 2016 can cause problems, you can disable it from Options > Advanced > Disable Graphics Acceleration.
• Limiting the number of calendars in the “My Calendars” folder will improve Outlook performance.
• Having over 5,000 items in any of Outlook folders (Inbox, Sent, etc.) usually cause slow responsive behavior in Outlook.
• It is well known for any iOS or Android users that the native mail applications do not work very well with Exchange (online or on-premises). Try to avoid those for troubleshooting.

As my personal opinion in this matter:

We are already in a world where decent Internet connections represent a basic and common service, like electricity. It’s hard to believe most companies will put a stop into innovation because there are still users who are exclusively Outlook dependent and they also do not have a decent Internet connection.

There are very few cases where they are using this scenario of no public connections for email connections. With smartphones, tablets and so many technologies and devices; user productivity will always go beyond a LAN connection for the messaging service.

Companies should try, if possible, to isolate those particular use cases and offer workarounds to the cloud service, like ActiveSync or OWA (which has great enhancements to make the web experience very similar to Outlook). And of course the hybrid scenario is useful again for this case.

7. Collaboration platform constructed and integrated already on-prem, re-building this scenario might be impossible in Office 365.

There are some customized scenarios customers have in their environment, where several collaborative and productivity platforms are highly integrated, for SAP platforms integrated with Exchange and printers so invoicing for a company runs completely automated. Let’s take a look about each of the perspectives.

Reviewing the approach of Exchange on-premise:

• We already invested time and money to complete these integrations, best case scenario we will have to make a new investment to achieve this migration.
• We don’t even know if our SAP system –or any other- will be able to support Office 365. Microsoft will not provide support for 3^rd parties.
• We cannot install anything on servers if we need a particular customization.
• Office 365 is offering some ways to connect with their systems using APIs, but that integration is limited.

For the Exchange online perspective:

• The systems provide native support for the vast majority of collaboration and productivity platform, like SAP and thousands of others.
• For the case of in-house developments of applications integrated with Exchange, most of companies have developed components integrated with AutoDiscover and EWS (Exchange Web Services) already; which can be easily integrated with Office 365.

Microsoft offers complete guidance on how to achieve these integrations. Here are some links:

Autodiscover for Exchange from MSDN
How to: Use Autodiscover to find connection points
Get started with EWS client applications
Get user settings with Autodiscover, sample application and code (C#)

• There are hundreds of applications available in Azure Market Place, where there are a big number of apps for Azure Active Directory, which is the identity backbone for Office 365. Here’s the link of all the Azure Active Directory applications available and ready to use.

As my personal opinion regarding this subject:

Companies that are constantly requiring in-house customizations on their collaboration platforms might not be suited for Office 365. Personally, I’ve never found an organization that had a large amount of customizations that cannot be implemented within O365 using Azure Market Place apps or taking advantage of the built-in features within the cloud platform. Data Loss Prevention (DLP) capabilities and Right Management Services (RMS) are the services I’ve seen the most as used by companies with 3^rd party products; and those services are already included in Office 365.

If we are talking about a scenario where there was a large customization investment from a company and they don’t want to change their platform, it is also a dangerous scenario because it implies that upgrading the Email platform (servers and/or clients) represents a high risk of losing functionalities. Therefore, in this case, the company is condemned to keep outdated versions of products for a long period of time.

8. Office 365 service availability is hard to track, Exchange on-prem monitoring can provide a clear diagnostic for the problem.

Those responsible about the monitoring of the platform can go either way: Or they can feel relieve that they don’t have to put much effort in monitoring a cloud solution; or can be really hectic about not knowing the service’s components health at all time.

Let’s take a look for the Exchange on-prem approach:

• In case we need to troubleshoot, we can find the solution to the problem and also apply it.
• With Exchange on-prem, we control the complete scenario: Clients and users, network and servers, external connectivity, etc.

One great feature in System Center Operations Manager (SCOM) are Distributed Applications, from which in one simple and clear diagram we can obtain a detailed view about the health status on all the components from our application. Distributed Applications are highly customizable, so your diagram can include anything that SCOM can monitor (which can even include, among other things, the UPS where your servers are connected).

This is an example of an Exchange Distributed Application in SCOM:

Not shown in the diagram but available in SCOM are the Watcher Nodes. You can install an agent (in a VM for example) in remote sites that will constantly monitor connections from the agent to your Exchange server. The diagram will show the ability for the agent to connect to the server from OWA or trying to use MAPI TCP protocol as an Outlook client.

For the Exchange Online perspective:

• Office 365, as most of the relevant cloud services, includes Service Level Agreements (SLAs) and service credits if the service’s uptime is not delivered.

Office 365 has very detailed documentation about SLAs which is described per workload.

Taking Exchange Online as an example, a downtime is described as “Any period of time when users are unable to send or receive email with Outlook Web Access”. Outlook app is not considered as a downtime factor if OWA is available.

Here’s the monthly uptime percentage, the credit associated and how is calculated:

This example refers to the documentation updated in January 2017.

• Office 365 includes the Service Health Dashboard, which can be monitored constantly using System Center Operations Manager (SCOM). Includes also a mobile app.

In this link you’ll find the Microsoft System Center Management Pack for Office 365. This is an SCOM dashboard example monitoring Office 365, using online watchers:

• You can monitor Office 365 in any way you want by using the Service Communication API. Including: Real-time service health; message center communications; and planned maintenance notifications.
• Office 365 includes 24×7 support. Here’s the contact information in case of support needed: “Contact Office 365 for business support – Admin Help”.

Microsoft offers differences from Business and Enterprise plans regarding support. Both include 24×7 support for critical issues (preventing to access your data) and a response time within the hour.

Business plans for high severity issues (events that affect the productivity of users but have moderate business impact) and non-critical (events that have minimal service or productivity impact on the business) are available only on business hours and it does not have a time frame committed to respond.

For Enterprise Plans the support is always 24×7.

As my personal take in this area:

A thorough, detailed and proactive monitoring of an Exchange on-prem platform can be challenging and with a high cost. There are several variables to consider around a server: Disks, memory, network ports, services running, performance counters, hardware components (including the UPS we can monitor with SCOM), etc. In most cases, at the end of the day, the final user is the one that is telling us that the service is not available.

All that cost associated in monitoring is greatly reduced with Exchange Online, moving that operational responsibility to Microsoft. Human error can also appear in cloud services, let’s take the AWS downtime where a S3 admin entered incorrectly a command line and generated a major meltdown; but these scenarios are rarely encountered in cloud platforms like AWS or Office 365.

[Bonus Track] Public Folders dependency within an organization and the challenge to migrate those to the cloud

Public folders is definitely a hot topic when I had the Exchange on-prem vs Exchange Online conversations. Surprisingly I had a large number of customers that assumed public folders is a feature no longer available in Office 365 (a subconscious desire maybe?).

No, Public Folders still exist in Office 365 and there are supported methods available to migrate them from Exchange on-prem. Surprisingly, to this date, there’s no supported method to migrate Public Folders from Exchange 2016 to Exchange Online.

What is Microsoft official stand about Public Folders being a feature that is going to be deprecated? From “FAQ: Public folders”:

Are public folders going away?
No. Public folders are great for Outlook integration, simple sharing scenarios, and for allowing large audiences to access the same data.

For any that participated in Microsoft events with Exchange experts or had personal reviews with any of them, you probably heard a different message than that. Microsoft has been trying to get rid of Public Folders for a long time. They’ve been trying since Exchange 2007 offering to customers migrate their content to SharePoint and collaborative group workspaces.

It is also true that a lot of Exchange admins suffer considerably with Public Folders, based sometimes on people using them for content that is not suited for this platform. That is something we should never negotiate, we know Public Folders has some limitations and considerations regarding sizing and the impact it has on our Exchange platform.

Microsoft offers several features and capabilities that facilitate Public Folders management within SharePoint, OneDrive, Office 365 Teams and even Yammer in some extent. These workloads will offer simpler and more scalable solutions to some of the Public Folders content we see.

Final Thoughts Regarding Exchange On-Prem vs Exchange Online

To summarize this set of articles, the “no silver bullet” statement maintains. Office 365 will not be automatically the solution to your problems, there are going to be different scenarios to consider but for sure it is a very valid and useful option. At the end of the day, Exchange on-prem will give you the ability to control exactly what goes on with your data and environment, but it also makes you fully accountable for it. That is something that not all IT departments are able to do.

Maintaining a complete, complex, secure, highly available and reliable collaboration platform represents a large investment of time, resources and of course money. Would you prefer shifting that spending on more strategic definitions for the business? Instead of dedicating resources on “keeping the lights on” for the messaging platform, start dedicating those resources for enhancing organizations capabilities with new solutions?

We all suffered at least once in our professional lifetimes with organizations that just see IT as an operational cost, with no real value to the business. We are also responsible on changing that mindset.

The cloud path has started for most of us and for the platforms we support, the features and capabilities are getting bigger by the minute. Most of the constraints and limitations we may have today, most likely will not be in the near future. Exchange is embracing that, the hybrid scenario has transformed from a recommended approach to highly necessary for most organizations.

Exchange 2016 might not be the last Exchange on-prem release, but we can assume the on-prem components could start reducing significantly in the next version, focusing on platforms that will be integrated with the cloud.

Are you sure you want to be the last organization with a 100% Exchange on-prem platform?

Previous post: [Yada Yada Cloud] Choosing from Exchange On-prem vs Exchange Online: The never ending story? – Part I

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

The latest and final Technical Preview 3 for Azure Stack has finally been released, this will be the last one before the GA (general availability) is shipped by mid-2017. And the good news, for all anxious customers, is that the pricing model to use Azure Stack has been released as well.

Updated Features for Azure Stack in TP3

Here’s the list of new features available in this Technical Preview:

• Deploy with ADFS for disconnected scenarios
• Start using Azure Virtual Machine Scale Sets for scale out workloads
• Syndicate content from the Azure Marketplace to make available in Azure Stack
• Use Azure D-Series VM sizes
• Deploy and create templates with Temp Disks that are consistent with Azure
• Take comfort in the enhanced security of an isolated administrator portal
• Take advantage of improvements to IaaS and PaaS functionality
• Improved alerting to facilitate Azure Stack operations and management.

Additionally, and maybe to deliver a shippable version of Azure Stack in a deadline, there were some additional features NOT included in TP3, but they will be available in the Technical Preview later: Azure Functions; followed Blockchain, Cloud Foundry and Mesos templates.

And, of course thanks to Internet, we already have people that have deployed Azure Stack TP3 on day 1.

Azure Stack Pricing and Important Takeaways

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

There are several of the assumptions and rumors expressed in my initial post about Azure Stack pricing that are now confirmed. Here’s a table summarizing how Azure Stack cost will be billed in the pay-as-you-use:

Here are some important takeaways from Azure Stack cost:

• There will be a unified billing for what each customer consumes from Azure Public and Azure Stack. To identify Azure Stack consumption, it will use a distinct region.
• Azure Stack fees will be lower than Azure since the customer will be performing the operations on the on-premises Azure platform.
• IaaS and PaaS services are charged on a pay-for-usage basis.
• Azure Stack will be available mainly in Microsoft’s Enterprise Agreement (EA) and Cloud Solution Provider (CSP) programs.
• Since the Azure Stack software will come pre-installed, customers will purchase the hardware and its support from the Integrated Systems (Dell EMC, HPE and Lenovo. Eventually, Cisco will join).
• There will be no up-front costs from Microsoft licensing, just the hardware you are acquiring from the Integrated System.
• Each Integration Systems will offer their support contracts for the hardware included. There will be a separate contract for Microsoft support.
• Integrated Systems will start offering Azure Stack in a 4-server bundle as the smallest unit, which will include storage and networking.
• Marketplace licenses will have the “bring your own license” model. Basically for third party apps, you will need to contact the specific vendor for the license.
• You can use Windows Server licenses in Azure Stack, using that approach you will only pay for the Azure Stack consumption fee. SQL will have the “bring your own license” model as the unique approach.

Azure Stack Availability

Azure Stack will be available in the following countries:

• Americas: Argentina, Brazil, Canada, Chile, Colombia, Mexico, Peru, United States, Uruguay
• EMEA: Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Israel, Italy, Luxembourg, Monaco, Morocco, Netherlands, Norway, Portugal, Russia, Saudi Arabia, South Africa, Spain, Sweden, Switzerland, Turkey, United Arab Emirates, United Kingdom
• APAC: Australia, Hong Kong, India, Indonesia, Japan, Korea, Malaysia, New Zealand, Philippines, Singapore, Taiwan, Thailand

Bolded countries are the ones in which systems will be available from all three partners, Dell EMC, HPE, and Lenovo. Other countries have availability from at least one partner.

This country list is tentative and subject to change in Azure Stack GA release.

Update: Country list updated. Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Azure Stack availability with Cisco UCS is still unknown. Getting late to the party could be costly after all.

Update: Huawei and Avanade were added as hardware OEM vendors, for more information: “Azure Stack Welcomes Two New Hardware OEM Vendors“.

Azure Stack Development Kit

Proof of Concept (POC) deployment model, which is exactly as we know today in the Technical Previews, will be renamed to the Microsoft Azure Stack Development Kit.

For more information on requirements and how to deploy Azure Stack as POC check my initial post: “Yada Yada Cloud: Azure Stack, what’s your story?”

Update: Azure Stack Development Kit available for download now. Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

As the TPs, this deployment model is a single-node server with no dependency on specific hardware and with no cost. It will be completely free for customers that need as their lab environments a simple version of Azure Stack.

This model is intended solely as dev/test environments, it will not have official Microsoft Support.

Azure Stack Updates Workflow

With Azure Stack, we will deliver updates that 1) add cloud functionality for creating and working with applications, and 2) improve Azure Stack itself.

•  Content, services, and tools updates: These updates do not have a regular scheduled release pattern, and can come anytime they are ready. They can include new Marketplace gallery content, updates to Services as well as new optional Azure Services that can be deployed to Azure Stack. They tend to focus on adding new end-user facing functionality to the Azure Stack deployment.
• System Updates: These updates are more structured and regular. They include new and updated Infrastructure Management capabilities, Firmware, Drivers and expanded Scale points. They focus on improving the operational excellence of Azure Stack.

Additionally after GA, Microsoft will include a monthly set of security updates.

Azure Stack it’s just a few months away to be released in GA (General Availability) and Microsoft has just decided they are going to officially support the applications migration (at least part of them) from one datacenter to another in India using the Azure appliance for datacenters.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

This migration scenario it’s not actually very different from what other customers might be already doing with Azure Stack TP2 (Technical Preview 2): Using the POC mode (one node) in their environment, using on-premises applications and workloads which can be easily integrated with Azure public and, of course, using the same Azure Portal to manage this integration from on-premises and cloud.

The migration in this case will involve the Andhra Pradesh state of India, which will be migrating 340 applications in total between different datacenters: Source datacenter is located in Hyderabad (State of Telangana, India) to the destination datacenters in Vizag and Vijayawada (State of Andhra Pradesh, India).

A total of approximately 315 miles from Hyderabad to Vijayawada.

Government of India helped the state of Andhra Pradesh to build a datacenter in Hyderabad, which was part of the same state at some point. But, in June of 2014 the north-western portion of the state was bifurcated to form a new state of Telangana, which generated the need of this project to migrate datacenters between now different States.

You may be wondering: “This is not a big deal, I’m already using Azure Stack for small production workloads, what’s the difference?”, but the catch relies on Microsoft signing a MOU (memorandum of understanding) and committing to migrate 15 apps from their environment.

The migration will still involve a POC implementation of Azure Stack, just one box, but having Microsoft accountable for applications migrations and supporting this project is still a significant step for a real-world validation of this platform.

Recently Microsoft CEO, Satya Nadella met the Chief Minister of the State Andhra Pradesh, N. Chandrababu Naidu. In this meeting they completed the agreement for the Azure Stack migration and some other initiatives. The Chief Minister stated that the mentioned state will be the “cloud-first state” in India.

Satya Nadella was actually born in Andhra Pradesh, India in 1967.

Applications Migration Process with Azure Stack

Microsoft has not released any official comment regarding this agreement, so most are conjectures regarding how they will support this migration. The particularity here is that this is a Government organization, therefore they have strong regulations regarding security of information.

Security protocols in the State of Andhra Pradesh do not allow store any information in the cloud, so they will use Azure Public Cloud as a gateway to transfer the data using Live Migration options for the applications.

Here’s the high level processes that will be used to achieve this migration using Azure Stack:

1. Connect source datacenter (Hyderabad) with Microsoft Azure Public Cloud. Express Route could be an option to guarantee low latency and high speed connectivity.

2. Prepare destination datacenters (Vizag and Vijayawada) with Azure Stack POC implementations and integrate them with Azure Public.

3. Migrate selected applications from source datacenter to Azure Public Cloud. This will be a temporary placement for the applications.

The MOU (memorandum of understanding) between Microsoft and the State of Andhra Pradesh enforces and focus its content about the privacy of their information in this temporal stage.

4. From the Azure Portal in destination datacenters, execute a Live Migration from Azure Public to Azure Stack for the selected applications.

As Azure Stack and Azure Public Cloud share the same code, there will be no additional transformation to perform within these apps.

This is a diagram for the process overview, at least what we assume will be the steps:

Even though this is a small scenario and it should not represent a complex migration to handle 15 applications, it is always good to know Azure Stack will have some real experience with critical data and Microsoft support.

Better late than never, as they say. After Dell EMC, HPE and Lenovo took the lead with Azure Stack developments, Microsoft was able to add a new player in the table: Cisco UCS as Integrated System for Azure Stack.

Update: Huawei and Avanade were added as hardware OEM vendors, for more information: “Azure Stack Welcomes Two New Hardware OEM Vendors“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

“Cisco is accelerating hybrid cloud adoption by gearing up to offer Microsoft Azure Stack on Cisco UCS in Q3 2017“, declared Liz Centoni, senior vice president and general manager of the Cisco Computing Systems Product Group.

Cisco UCS, their server platform, is looking to generate some significant market share in a world clearly dominated by Dell EMC and HPE. Currently in the server market share, Cisco is battling for the third place between Lenovo and IBM; trying also to pick some momentum after last year’s announcement of entering the hyper converged space with HyperFlex systems.

Cisco is showing as the fourth and late(st) player in this game. In September 2016, in the Microsoft Ignite conference, Dell EMC, HPE and Lenovo showcased their Azure Stack platforms to the public with the Technical Preview 2 (TP2) recently released.

Azure Stack, still confirmed to be released by mid-2017, is one of the biggest cards from Microsoft in the Hybrid Cloud offerings. The Redmond based company understands that allowing the customers to offer Azure in their own datacenters, which can be easily integrated with Azure in the public cloud, will generate the concurrent one-stop shop Microsoft wants for their cloud consumption.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Currently Azure Stack is only available in POC mode, completely free (only one server installation, no support from Microsoft for production environments). Additionally, the product will only be released for production environments using one of the Integrated Systems: Dell EMC, HPE, Lenovo and now Cisco UCS. Azure Stack will be part as a pre-deployed component in that hardware, not requiring for customer to install it from scratch.

The good news is that the POC mode for Azure Stack will still be available after GA (General Availability), completely free and the alternative to install it on any hardware.

For detailed information about Azure Stack like requirements, step-by-step deployment, review on the Integrated Systems approach and more, check my previous blog post: Yada Yada Cloud: Azure Stack, what’s your story?

Continuing with the cloud discussions, an interesting theme to review is the Office 365 suitability and readiness for each organization. I want to expand this discussion with a comparison between Exchange Online vs Exchange On-Prem, reviewing the most common conversation points when I get to evaluate the Office 365 migration alternative.

This is going to be a two set article, distributed with the top 8 discussions topics when organizations need to choose from Exchange On-Prem or Exchange Online. Each of these topics will include the Exchange on-prem side of the discussion as well as the Exchange Online, I’ll add my opinion and thoughts regarding each discussion.

As an initial disclaimer I must say that it is my firm believe that Office 365, as pretty much any technology available, does not represent a silver bullet. There’s no one and final truth all organizations will apply, that is what makes this conversation valuable and rich.

“No Silver Bullet – Essence and Accidents of Software Engineering” is one of the most important and fundamental papers related to software engineering, it was released in 1986 by Frederick P. Brooks.

“Not only there are no silver bullets now in view, the very nature of software makes it unlikely that there will be any –no inventions that will do for software productivity, reliability, and simplicity what electronics, transistors, and large-scale integration did for computer hardware.”

To download the full publication, access this link (PDF).

1. Office 365 is forcing customers to update the software frequently to be functional.

This is the initial complain I hear from customers when we are talking about a new Office 365 environment.

Let’s make a quick review about the Exchange on-premises perspective:

• Updates are out of our control in the environment, if Microsoft releases an update that is required for clients to work with O365, we have no option.
• Critical updates might be released during a company’s end of fiscal year or any sensitive week, might generate a huge business impact.
• Desktop clients’ updates could be time consuming and an expensive task for IT admins, Help Desk, etc. Microsoft released over 300 updates within the same year for Office 365.

Microsoft’s update model for Office 365 ProPlus on monthly basis

“Admins—get ready for Office 2016, rollout begins September 22!”

And for the Exchange online perspective:

• Updates are required for on-premise scenarios as well. Any security update release by Microsoft should be addressed in a timely fashion.
• Microsoft updates management platforms WSUS and SCCM already support Office 365 clients, you can update them within your methods and schedule.

Check the following links for more information “Office 365 Client Updates via WSUS” and “Manage updates to Office 365 ProPlus with System Center Configuration Manager”.

• If you have a dependency with Outlook 2003 or 2007 and you are not able to update it, this might be because you have a different problem to face better than Exchange Online.
• Clients’ deployments are easily managed with Click-to-Run for Office 365.

Click-to-Run is a technology developed from the App-V principles of application virtualization and streaming. The Click-to-Run process can be executed on demand from the user (accessing the website and/or installer), or deploying it with an automated method for the application streaming to start, making the Office 365 client components available rapidly.

Following the application virtualization model, the Click-to-Run Office 365 deployment runs isolated from other Microsoft Office versions already installed in the client OS, therefore no conflicts will appear.

Click-to-Run deployment process, the applications are ready as the streaming and preparation is completed.

Click-to-Run, same as previous versions of Microsoft Office, provides the alternative to use the configuration.xml file for customizations. For more information: “Reference: Configuration options for the Office Deployment Tool”.

Also, Click-to-Run offers the alternative to allow an automated deployment for one or several computers using local content for the installation files (the components that are streamed to the client), instead of each client downloading from the Internet these components. Check the following link: “Deploy Office 365 ProPlus from a local source”.

A couple of considerations regarding Click-to-Run are: Click-to-Run products are installed only on the system drive, which is typically drive C; you cannot customize the installation location. Also, Click-to-Run has enabled the automatic updates for their products, this option can be modified using the configuration.xml file.

• Migrating to Exchange Online should be the only big and final migration each organization will experience. Migrating from different Exchange On-prem versions can have larger impacts.
• Tie the productivity experience to old applications versions reduce the chance of innovating and offering improved experiences.

As my personal opinion:

Most organizations that mentioned this topic usually are experiencing challenges on updating their client platforms based on problems that go beyond their productivity suite (for example, as mentioned, depending on legacy applications no one can update). It is never a good idea to depend on old and outdated software.

Additionally, for any change to succeed, user adoption is critical. Therefore, organizations need to assess and understand their productivity use cases in order to create the proper training. If your organization can’t or won’t face this change and training, then migrating to Office 365 will transform more into a problem than a solution.

2. Larger mailboxes, our salvation or downfall

No matter if you are talking about going to Exchange Online or you’ve been on-premises for several years, this is for sure a discussion that most of messaging admins had and still have to this day.

Reviewing the Exchange on-premise arguments:

• Large mailboxes offered by Exchange Online can be a liability, the IT / Exchange admins end up being accountable for all the data in their inbox. Let’s force users to have only work stuff in their mailbox.
• Forcing users to manage their own mailbox and deleting their own data is a far simpler solution.
• In case we absolutely need to, we can provide large mailboxes on our terms. Buying more disks for my Exchange it’s not that expensive.
• We can provide an on-premise archive solution to mitigate large mailboxes or compliance requirements.

And if you talk with the Exchange Online approach:

• Too much corporate data is spread across PSTs in local or external drives or shared folders. PSTs are the real liability!
• Having online data always available offers users a simpler way to handle their history mailbox data.
• Compliance is easier with Office 365.
• Implementing an archive solution can be really complex and costly to operate and maintain.

Office 365 offers several options regarding compliance, the most important ones are related to In-Place Archive, In-Place Hold and Litigation Hold. Exchange 2016 also offers pretty much same capabilities, but usually harder to implement and maintain than just enabling these features within O365.

In-Place Archive provides additional storage space for the mailbox data. Appears alongside the users’ primary mailbox folders in Outlook or Outlook Web App. Microsoft offers 50GB in addition to the 50 or 100GB mailbox (depending the plan), but additional space can be provided if it’s required. The archive cost represents $3.00 per user / month.

Litigation hold is normally used when you want to put an entire mailbox on hold. Preserves all data in this mailbox, including deleted items and original versions of modified items. This option is usually what legal departments need. For more information: “Place a mailbox on Litigation Hold” (option also available in Exchange 2016: “In-Place Hold and Litigation Hold in Exchange 2016”).

In-place hold allows administrators to create hold based on different criteria, like keywords. Which can be applied to several mailboxes with the same criteria.

Combined with these options, Office 365 includes eDiscovery, when in-place hold is enabled, allows the administrators for a single pane where the searches are defined and results are constantly being updated. For more information: “Manage eDiscovery cases in the Office 365 Security & Compliance Center”.

The eDiscovery searches are defined as “cases”, where we can define keywords and where. The “Search everywhere” option includes mailboxes and SharePoint sites.

This option is also available in Exchange 2016: “In-Place eDiscovery in Exchange 2016”.

Curious fact: Microsoft states that every year they save $4.5M on eDiscovery by reducing manual work from discovery requests: “Microsoft saves $4.5 million annually using Office 365 eDiscovery”.

My personal take regarding the large mailbox dilemma between on-prem and online:

Being accountable for every email and item on each user mailbox could be madness, the same applies for PSTs, it is almost impossible to track and own for an IT department these files scattered everywhere. The mailbox size discussion is usually related to each organization culture, several times I’ve heard IT admins complaining about most users demanding mailboxes quota increases without any specific reason besides “if you don’t increase the mailbox size, I’m not productive and you are disrupting the business”.

Exchange on-prem, no matter the organization big or small, require clear policies, defined and respected mailboxes quota profiles.

The questions back usually are: “How much time, effort and money are you spending for that accountability?” “Would you rather to transfer the responsibility to Exchange Online to keep that data?”

Granted there are some scenarios and organizations that indeed have compliance or security issues to use Office 365 (we’ll talk about them also). Sometimes those cases are suitable for a hybrid approach on Exchange, some particular mailboxes need to keep on-prem and some others are easier and cheaper to maintain online.

3. Basic email is the only feature needed, no need to pay additional features or workloads available with Office 365

Let’s take a quick look about these scenarios where organizations are just happy with the email functionality provided by Exchange on-prem.

Here are the Exchange on-prem arguments:

• Why do we need additional workloads if the organization just needs email? Why pay extra for features we will not use? Exchange on-prem already provides High Availability scenarios and reliable parameters to guarantee a solid Service Level Agreement (SLA).
• Microsoft pushes collaboration platforms like SharePoint or Yammer when our organization it’s not interested, everything they need is provided by email.
• Collaboration activities can be provided by Facebook, all of our users are already using it and talking with that platform.
• Exchange on-prem represents a reliable platform, we know how to operate it correctly and it does not surprise us with inconsistent behavior.

For the Exchange Online comments:

• Exchange Online includes high availability, load balancing, security, Data Loss Prevention (DLP), anti-spam and anti-malware; etc. All of those messaging related components within one price.

Exchange Online Protection (EOP) offers all these security capabilities included by default with all Exchange Online subscriptions. EOP can be acquired as a standalone service and used with an Exchange on-prem implementation.

• Office 365 offers predictable cost. Even if you just need email, there are variable costs associated to maintaining, operating and upgrading when necessary an on-prem platform.
• There are going to be, in the near future, some innovations into the email platform that will only apply for Exchange Online and not on-prem.
• Regarding security concerns, Office 365 has built-in parameters making it more secure than most Exchange on-prem implementations.
• You don’t have to worry to ensure the Exchange backend platform is functional, that’s Microsoft’s responsibility.

Here’s my personal view about this topic:

I’ve engaged with organizations with a genuine lack of interest on renovating their messaging platform. Sometimes because their users are fairly static and simple email consumers or they just have a user count fairly small. In most of those cases they usually rely on an old Exchange version and also this messaging platform does not represent a core and critical service for their business.

Trying to convince these type of business that Office 365 is the right answer for them might not be true. Having said that, for those organizations running outdated versions of Exchange, running an unsupported version represents a clear risk that they need to assume. Sometimes the “low operational cost” mentioned by them about the Exchange platform it is not 100% accurate; the probability of those risks (like security breach, unstable platform, etc.) and its impact should be quantified and included in this cost.

And we should also always remember that organizations can just acquire Exchange Online without the need to pay for other features like SharePoint, Yammer, Delve, etc.

4. Office 365 has an inexistent backup strategy, Exchange on-prem backups provide an additional guarantee

When this topic appears and there’s someone related to the auditing process present, it’s usually a difficult topic to handle if Office 365 shows up as the main alternative.

The Exchange on-prem admins arguments:

• Offsite backup policy is established within our organization, we are required to have a backup available in one of our sites due audit compliance or fraud investigation.
• Auditors tend to have a “boolean” review about policies: Either you have a backup or you don’t.
• We had cases from an Executive requiring to recover an email he/she deleted 3 months ago. Our cold backups can provide that functionality.
• We just need the comfort feeling on having a backup available.
• Human error for Exchange Online admins apply as well. We don’t have any control over that infrastructure.

For the Exchange Online discussion:

• That’s correct Office 365 doesn’t make backups, Native Data Protection is the answer to that. Exchange Online does not need backups when multiple databases are replicated to different datacenters.

So, what is exactly Native Data Protection included in Exchange Online? This feature appeared in Exchange 2010 and basically resides in Database Availability Groups (DAGs) databases replication. The databases and the logs associated to those are replicated between the Exchange servers member of the DAG group. The idea behind it was to reduce costs from Exchange backups and providing the alternative to eliminate them completely.

Native Data Protection definition for Exchange Online uses also the database lagged copies. This feature allows that any change in a database will have a delay (specified by a customizable period of time for Exchange on-prem) until these changes are fully committed. Exchange Online, unofficially, has a 7-day delay for lagged copies. This permits that any database corruption (logical or hardware based) will have a point in time prior to that corruption or loss.

Sample diagram of an Exchange DAG with active, passive and lagged databases.

Unofficially, Microsoft has 6 database copies for the Exchange Online servers. This databases are distributed, by default, within 2 datacenters within the same region. But there are some regions that have the databases replicated into 4 different datacenters.

• Exchange Online offers the capabilities to configure deleted item retention to restore emails deleted; single item recovery; and retention tags and policies to easily customize the platform.

So, what happens when a user deletes an email? If the mailbox does not have In-Place Hold or Litigation Hold enabled, when the user deletes an item, it is kept in the Deletions subfolder of the Recoverable Items folder. Items remain in this folder until the user manually removes them, or until they are automatically removed by retention policies. For Exchange Online the after an email is deleted, by default, it stays for 14 days. It can be modified up to a maximum of 30 days, for more information check: “Change how long permanently deleted items are kept for an Exchange Online mailbox”. Exchange on-premises allows a maximum of 24,855 days or the equivalent of 68 years (?!) to be set as a retention period.

One of the features that were evaluated by the Exchange Online team in Microsoft was to allow an unlimited amount of days to recover items, unless the user manually removes them. It’s still unconfirmed if this feature will be enabled at some point.

But if the mailbox has enabled the In-Place Hold or Litigation Hold, stops the processes on purging deleted items, all the content in that mailbox will be preserved. Additionally, copy-on-write page protection is also enabled for the mailbox. Copy-on-write page protection creates a copy of the original item before any modifications are written to the Exchange store.

Workflow overview of a message life cycle:

• The In-Place Hold and Litigation Hold are the easy way to resolve backup issues. With that approach, you are guarantee to maintain all items for selected or all mailboxes.
• Destructible PowerShell cmdlets are not accessible for Office 365 tenant admins. No one in Microsoft has standing admin access.

As my personal review about this topic:

As mentioned, this could be a difficult topic to engage when there’s a hard requirement about offsite backups and/or audit compliance; because depending on the point of view, some could assert that Office 365 doesn’t have the same capabilities as an on-prem solution.

The initial response to that requirement is that you can enable In-Place and Litigation Hold and you are fulfilling the need on having backups for mailboxes (I’ve heard this recommendation mentioned from Microsoft as well). This solution still does not fit for organizations requiring to own the backups in a different site.

Sometimes I hear questions from organizations asking: “What happens if there’s a massive earthquake that takes down a whole Microsoft region with all their datacenters?” If that’s the case, we may have a world catastrophic event, considering that these datacenters are built and regulated to support earthquakes, flooding, fire and most of intentional and unintentional disasters. It is definitely most likely that a simultaneous event affecting your datacenter and your offsite location for backups will occur before that catastrophic scenario.

The protection offered by Microsoft within Office 365 and Exchange Online is one of the highest and most reliable in the market; that concept alone is the backbone of their cloud services. If customers’ data is lost or becomes unavailable, how Microsoft and other companies can profit with these solutions?

In the second post I’ll cover some of the additional discussions related to the Exchange On-Prem vs Exchange Online, including: The challenges of inconsistent Internet connections; on-prem collaboration platforms built and integrated already; Office 365 service availability and monitoring; compliance and legal issues. And yes, we will have some comments regarding the “Public Folders” scenarios.

Second Part: [Yada Yada Cloud] Choosing from Exchange On-prem vs Exchange Online: The never ending story? – Part II

It’s been a while since my last post, a whole year to be exact. Also, before my last post I share some thoughts regarding “writers and bloggers block” and how we usually get stuck without developing or writing any material, even though we should be able to.

I decided to create a new set of articles related to cloud, which is one the main topics I’m currently working on my day job. The “Yada Yada Cloud” concept represents the way I’ll try to simplify these subjects, and avoiding all the “chitchat” or platitudes we sometimes get when we are researching some new cloud concept.

Five points if you already got the Seinfeld reference: “The Yada Yada”.

There’s a lot of buzz related to Azure Stack (formerly called Windows Azure Pack vNext) within the last few months, and the tendency clearly states that it’s not going to change. So let’s start reviewing the concepts behind Azure Stack without getting lost with all the excitement.

Update: With TP3 being released, there are some important updates about features, pricing, and other topics. For more information: “Azure Stack TP3 is Here! And with Pricing“.

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

In this post I’ll review the following topics:

• Azure Stack Definition: Simple as you can get.
• What can Azure Stack provide: Review about the services we can expect to use from Azure Stack.
• How can I install Azure Stack?: Requirements review and the step-by-step to deploy Azure Stack.
• Azure Stack Integrated Systems and the operational model: Implications about the operations within Azure Stack and the integrated systems.
• What’s the representation of an Azure Stack instance?: Simple definition about how Azure Stack instance is represented.
• Azure Stack cost: What we know so far about Azure Stack cost.
• Where’s Windows Azure Pack (WAP) and Cloud Platform System (CPS) in all of the yada yada?: What are the differences and how WAP and CPS fit in the Azure Stack world

Azure Stack Definition

Simplifying, Azure Stack represents the way to run Azure in your datacenter. You will have an Azure implementation within hardware you own, allowing you and other companies to offer the public cloud services within your own datacenter.

One of the key components in Azure Stack is about extensibility and elasticity. Here are some of the important features around Azure Stack:

• Azure Stack can be easily integrated with Azure (they are using the same code and binaries), therefore customers can expand their resources as they need it and still be using the Azure Stack portal to provide their services.
• Customers can build and offer their own type of applications (like PaaS platforms), services and customizations in the environment as their own catalog with charge-back possibilities, same as in Azure.
• Azure Stack will be offered, initially, on what’s called “integrated systems”. Dell EMC, HPE and Lenovo will be the only vendors to offer the platform pre-installed on their hardware.

Update: Cisco joins as one of the Integrated Systems, more info: “[Yada Yada Cloud] Azure Stack: Cisco joins the party”

Update: Huawei and Avande are joining as Integrated Systems as well, more info: “Azure Stack Welcomes Two New Hardware OEM Vendors“

What Can Azure Stack Provide

Since we are talking about Azure Stack offering the same capabilities as Azure, here are the services we can consume using this platform:

• Compute: Virtual machines (supporting Windows and Linux) and any customizations we can provide with these virtual machines. Provides Azure Virtual Machines (offering Windows and Linux VMs on demand) and VM Extensions (allowing VMs customization).

Not all sizes will be possible initially Azure Stack (very small or very large). Instances available will be A (0 to 7), D (1 to 4 and 11 to 13) and D v2 (1 to 4 and 11 to 13).

• Storage: Built on Windows Server 2016 SDS (Software-defined-storage), it offers blobs (what we usually use as the OS or data disks in VMs), Tables (NoSQL key/value store), and Queues (letting cloud software communicate via messages).

• Networking: Built based on Windows Server 2016 SDN (Software-defined-networking), provides Virtual Networks (allowing the creation of isolated networks in the cloud, including the integration with Azure), Load Balancers (layer-4, to balance load between different VMs), and VPN Gateway (allowing connections among virtual networks and more).

• Platform as a Service (PaaS): Provides App Service (supporting Web Apps, Mobile Apps, and API Apps created using .NET, Java, PHP, or other technologies) and Service Fabric1 (offering a foundation for micro services applications).

• Security: Key Vault (for securely storing encryption keys). This is basically the “secrets repository”, where all certificates and passwords are stored.

• Azure Resource Manager (ARM): Capabilities for automated deployment and a vast number of features and services. ARM exposes RESTful APIs to Azure Stack services and allows the creation of templates to automate the deployment of Azure resources.

• Management and extensibility: Azure Stack Portal and support via Azure Resource Manager (ARM) for other clients (Visual Studio, PowerShell, and a command-line interface (CLI) for Linux, Macintosh, and Windows).

Update: Features update for TP3, more information: “Azure Stack TP3 is Here! And with Pricing“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

How Can I Install Azure Stack?

Currently you can find available Azure Stack Technical Preview 2 (TP2) ready for download, General Availability (GA) does not have a definitive date just yet but is expected to appear in mid-2017 (near end of Microsoft’s fiscal year).

Microsoft only supports the Azure Stack deployment in a Proof-of-Concept mode (POC) in a single-node. Saying the obvious, Microsoft will not support production deployments of Azure Stack in this stage.

Azure Stack Requirements

Here is the recommended configuration mentioned by Microsoft:

Unofficially Microsoft stated that Azure Stack Technical Preview 3 (TP3) will come with an integrated version of Active Directory.

Update: Azure Stack TP3 will not include an integrated version of Active Directory, but it will include the option to use ADFS for disconnected scenarios. More information here: “Azure Stack TP3 is Here! And with Pricing“.

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

Microsoft offers the Deployment Checker for Azure Stack Technical Preview 2 to confirm that your hardware meets all the requirements.

Update: The deployment checker works for TP3 as well, the script is updated constantly. The same requirements from TP2 apply for TP3.

To get a full list of requirements and recommendations for the Azure Stack POC, access the following link: “Azure Stack deployment prerequisites”.

Azure Stack Deployment

Once you have all pre-requisites in place, the recommended steps for deployment from Microsoft is the following: “Deploy Azure Stack POC”.

Here’s an overview about the step-by-step to implement Azure Stack (it may take up to 2 or 3 hours to complete):

1. Download Azure Stack Technical Preview, accessing this link.

2. Extract Azure Stack files and copy the CloudBuilder.vhdx file into the C:\ drive.

3. Download the Azure Stack TP2 support files using PowerShell script.

# Variables
$Uri = ‘https://raw.githubusercontent.com/Azure/AzureStack-Tools/master/Deployment/&#8217;
$LocalPath = ‘c:\AzureStack_TP2_SupportFiles’
# Create folder
New-Item $LocalPath -type directory
# Download files
( ‘BootMenuNoKVM.ps1’, ‘PrepareBootFromVHD.ps1’, ‘Unattend.xml’, ‘unattend_NoKVM.xml’) | foreach { Invoke-WebRequest ($uri + $_) -OutFile ($LocalPath + ‘\’ + $_) }

4. Run the PrepareBootFromVHD.ps1 script (confirm required parameters). Reboot will be required since the machine will boot into the VHDX.

.\PrepareBootFromVHD.ps1 -CloudBuilderDiskPath C:\CloudBuilder.vhdx –ApplyUnattend

5. Execute the “Install Azure Stack POC” PowerShell. Here’s one of the examples using one Azure Active Directory:

cd C:\CloudDeployment\Configuration
$adminpass = ConvertTo-SecureString “<LOCAL ADMIN PASSWORD>” -AsPlainText -Force
$aadpass = ConvertTo-SecureString “<AAD GLOBAL ADMIN ACCOUNT PASSWORD>” -AsPlainText -Force
$aadcred = New-Object System.Management.Automation.PSCredential (“<AAD GLOBAL ADMIN ACCOUNT>”, $aadpass)
.\InstallAzureStackPOC.ps1 -AdminPassword $adminpass -AADAdminCredential $aadcred

6. Connect to the Azure Stack POC using RDP or VPN, following the guideline: “Connect to Azure Stack”.

7. And now you are ready to start working with Azure Stack, you can try “Provision a virtual machine”.

When Azure Stack gets GA, the installation process is going to change?

Yes, significantly: There’s not going to be any. Azure Stack will be offered with the “Integrated Systems” as an OEM. Azure Stack will come pre-installed in the hardware that you are buying, these Integrated Systems will be Dell EMC, HPE and Lenovo. There will be no option on buying on different hardware, for now.

Curious fact: In 2010, in the World Partner Conference (WPC) event, Microsoft mentioned the existence of an “Azure appliance” that will be offered by Dell, HP and Fujitsu. They even stated that eBay was one of the early adapters, servicing with that appliance some of the web applications to the public.

Update: Cisco joins as one of the Integrated Systems, more info: “[Yada Yada Cloud] Azure Stack: Cisco joins the party”

Update: Huawei and Avande are joining as Integrated Systems as well, more info: “Azure Stack Welcomes Two New Hardware OEM Vendors“

Azure Stack Integrated Systems and the Operational Model

Microsoft mentioned, without any guarantee, that in the future there could be a scenario where customers can actually install on their own Azure Stack on the hardware they want to, with specific hardware recommendations from MS. The release of Azure Stack on these Integrated Systems is mainly because two reasons:

1. Microsoft wants to offer Azure Stack on a platform that is guaranteed to function properly. The number of hardware vendors, components and combinations are far too many to get a newly product in the market and make it compatible with all of those variables.
2. Customers need to focus on providing services and administering the Azure Stack platform, instead of worrying about compatibility issues.

Azure Stack will have, of course, updates being released periodically for the platform. Microsoft will deliver these updates and all the details on the operational model supported for Azure Stack.

Update: Cisco joins as one of the Integrated Systems, more info: “[Yada Yada Cloud] Azure Stack: Cisco joins the party“

Therefore, as they have for Azure, Microsoft and the integrated systems will have a detailed manual and processes defined for: Patching operating systems, disk controllers, drivers, and firmware; replacing hardware components; and any other operational task required.

The Azure Stack patching and updates releases, same as Windows Azure, will be pre-validated for software and firmware and designed to not disrupt tenant workloads.

Microsoft introduced its Patch and Update Framework (P&U) with the standard edition of its Cloud Platform System (CPS). CPS was designed to run Microsoft’s previous Azure Pack software, and has been re-engineered to be an on-ramp of sorts for Azure Stack.

Update: Microsoft has released the updates workflow following up TP3 and GA release for Azure Stack. It will divided in content, services and tools update; system updates and security updates. More information: “Azure Stack TP3 is Here! And with Pricing“.

What’s the Representation of an Azure Stack Instance?

We just discussed about the integrated systems and Azure Stack deployments, but what’s exactly is the Microsoft definition for an Azure Stack instance? There are several components to define this, but let’s start:

An Azure Stack instance is defined by the following:

• Single instance of Azure Resource Manager (ARM)
• 1 or more Regions under management of ARM
• 1 or more Scale Units within a Region
• 4 or more servers within a Scale Unit

Let’s break down these concepts.

Azure Stack Region

• Set of Scale Units that share same “physical location”
• Under one physical and logical “administrator”
• Networking requirements: High-bandwidth / Low Latency

Azure Stack Scale Unit

• Associated with a single Region. 1 or more Scale Units can exist in that Region
• Unit of Capacity Expansion. The smallest scale unit will be 4 servers.
• Fault Domain (Azure Consistency)
• Alignment of Hardware SKU within the integrated system (Homogenous within Scale-Unit)
• Servers must share top-of-rack (ToR) switch
• Servers part of the same Failover Cluster
• Each scale unit can have different hardware generations

With those definitions we can say then:

Azure Stack Scale = Scale Unit x Scale Units per Region x Number of Regions

Unofficially Microsoft says the maximum number of servers in an Azure Stack instance will be 65,000. It is also expected that number will increase in time.

Currently Azure Stack POC in Technical Preview only supports a single-node for deployment, which of course is not supported for production environments (this means you won’t be able to initiate a support ticket with Microsoft). The good news stated by Microsoft is that the POC single-node option will be always available after GA for free.

Update: The POC single-node mode will be called Azure Stack Development Kit in GA. It will have the same characteristics and with no cost. For more information: “Azure Stack TP3 is Here! And with Pricing“.

Azure Stack Cost

There is no official statement from Microsoft about the cost model on Azure Stack, but there were some hints here and there about it:

Update: Most of the assumptions made here were confirmed by Microsoft on the release of TP3, and some additional details on pricing. For more information: “Azure Stack TP3 is Here! And with Pricing“.

Update: Azure Stack final pricing and packaging released, more information: “Azure Stack Final Pricing Available, Disconnected and Fixed Fee Model Confirmed“.

Update: Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

• Microsoft wants to align the cost model to the hybrid cloud scenario for customers.
• Pay-as-you-go is the preferred method for Azure Stack. The idea is that you won’t be paying for an operating system license per server you are buying, but the cost will be related to an ownership model around the hybrid cloud the customer is providing with Azure Stack.
• Microsoft wants a unified billing for public and private cloud.
• Microsoft stated that customers “ready to deploy an Azure-consistent cloud on their premises now should buy Microsoft Cloud Platform Solution (CPS)”. Customers will be able to use Azure Stack to manage CPS resources thereby preserving investments in CPS.
• The integrated systems (Dell EMC, HPE and Lenovo) will still offer their portfolio (or at least some variations of it) of support contracts and subscriptions on their hardware where Azure Stack will be deployed.

Update: Cisco joins as one of the Integrated Systems, more info: “[Yada Yada Cloud] Azure Stack: Cisco joins the party”

Update: Huawei and Avande are joining as Integrated Systems as well, more info: “Azure Stack Welcomes Two New Hardware OEM Vendors“

Where’s Windows Azure Pack (WAP) and Cloud Platform System (CPS) in all of the Yada Yada?

Windows Azure Pack (WAP), appeared in 2012 with Windows Server 2012 and System Center 2012 releases, is Microsoft’s software-defined hybrid cloud and bundles Windows Server, System Center and more into a package that can run VMs created in Azure and then downloaded to your own datacenter.

Cloud Platform System (CPS) debuted in 2014 with the Azure appliance approach, pre-installed and configured hardware with the integrated systems, offering automation and integration with Windows Azure. WAP appears also as one of CPS components to achieve this similar hybrid cloud scenario with Windows Azure. CPS overall objectives and features are very similar that we can find now in Azure Stack. Although all signals are pointing that Azure Stack will be the preferred Azure appliance or Azure-in-a-box offering.

WAP and CPS are meant to be maintained (at least for now) as a complement for the Azure Stack offering, even though there is going to be some significant overlapping. Microsoft intend is to maintain and encourage all customers that already have WAP and/or CPS to extend capabilities with Azure Stack.

To that notion, Microsoft will be releasing the WAP/CPS connector with Azure Stack. This connector will have the following features:

Update: The WAP connector will be available in Q4 of 2017, according to Microsoft. Azure Stack has been released in GA (General Availability) with updated features, Azure Stack Development Kit free download and more: “Azure Stack Is Here! Development Kit (Single-Node) Free Download Available and More“.

• Connecting any existing WAP/CPS platform with Azure Stack in order to allow customer to preserve any investment already made in these platforms.
• Enabling tenants to access VMM IaaS resources from Azure Stack portal through seamless integration.
• Will be released through WAP/CPS Update Rollup at Azure Stack General Availability (GA).

As you can see, Azure Stack has a large number of concepts and features that can be extended far beyond this article. The idea is to set the “Yada Yada Cloud” as the place I can share and discuss these type of topics.

We’ll see each other soon!

Do not be afraid; meet this new (or not so new) player in town: Software Defined Storage (SDS). Moreover, why is it important for you to know about its existence? Why should you start considering evaluating it in your solutions deployed within your organizations or customers? Let us take a quick look into the technology, what represents, and the main benefits.

Before getting into Software Defined Storage, let us recap about the storage definitions we had before SDS:

• DAS: Direct-attach storage, like the ones you have on your notebook, desktop or tablet, hard drives attached to a mainboard. No complications here, DAS types differ from each other on the way they are connected to that mainboard.
• SAN: Storage Area Networks also connected to the hosts (mainly servers) but via a backplane using adapters and cables. These type of connections vary from Fiber Channel, iSCSI (which is SCSI, Small Computer Systems Interface, but over internet) and Serial Attached SCSI or SAS. SANs still are by far the main storage deployment for servers we can find in organizations.
• NAS: Network Attached Storage, is what we commonly understand for “file servers”. Servers with storage attached (which can be DAS, SAN or SDS) prepared for file sharing. Accessing a NAS it is not performed as DAS or SAN work, using SCSI protocol, but using a network file system like SMB from Microsoft, which works over TCP/IP.

So, when actually RAID technology starts to play?

RAID (Redundant Array of Independent Disks) appeared a long time ago as an answer to achieve some protection and improvement within the data we stored in the hard drives, showing as a single drive a combination of several disks.

RAIDs are configured using software installed in a server or within an array controller. Some people call the latter “hardware arranged RAID”, but it is software nevertheless installed on an array controller.

This software dependent storage called RAID introduced improvement on how we considered storage, but with a large complexity costs. Specialized hardware is needed to achieve RAID, interconnections and arrays prepared (vendors specific); software to integrate with all those components. Each components has to be compatible with each other, if any problem appears, the data within that storage could not be accessible.

Outlining Software Defined Storage

SDS introduced the concept of separating completely the hardware from the software that managed this storage. This idea of decoupling the hardware from the solution, easily relates with the virtualization concept: SDS does for storage what virtualization does to servers, which is why we also call SDS as virtual SANs.

There are varies definition about experts that define storage virtualization as not the same concept as Software Defined Storage. We will tackle the differences along the way in the series of posts I have prepared.

SDS is no other than a software layer handling the data access to hardware, but we can even say, “That’s exactly what RAID technology does!” and we will be somehow right. However, RAID will use software to transform several disks into one, that is just one characteristic we can achieve from SDS. Using Software Defined Storage, we can define different features, ways of access the data, and policies like QoS, integrate several different components (hardware and software) which is usually not possible with RAID, and so on.

Image used from http://headintotheclouds.com

Regarding the technologies and vendors we can find offering SDS solutions, there will be differences as well. StarWind concepts about SDS differs from what you VMware has to offer; and Microsoft concept is different about VMware’s conception. We will take a closer look also of those in the following articles.

Main Benefits about Using SDS

Before getting to know the alternatives, vendors and technologies within SDS we can find, let us take a review about the main benefits on using Software Defined Storage for our solutions:

• Decoupling software from hardware gives you more possibilities when handling your storage/s solutions: As mentioned before, using different disks as one is just one section in SDS, we can define different ways on how to access data, policies applying to different workloads, hypervisors; automation possibilities; and several other features.
• Heterogeneous storage is possible: Related to the benefit mentioned above, adding a software layer also allows us to use different type of storage, different type of disks, arrays and even hardware vendors to manipulate the same SDS we define. You will no longer depend on a specific vendor when considering to improve or expand your storage solution.
• Control performance and capacity: Much easier to control on how the storage is used by the solutions hosted; managing the performance our workloads can access depending on our definitions; dynamically expanding capacity and performance as our systems needs it is one of the main benefits within Software Defined Storage.
• SDS solutions can provide the possibility to reduce storage costs immensely: Software Defined Storage concept, depending on the vendor you select, work on commodity and inexpensive hardware. TCO (Total Cost of Ownership) is greatly reduced since we reduce the cost not only for the hardware, but also about maintenance and operations.

As we can see, SDS represents the dynamic solutions our customers require daily, in a market where 24x7x365 represent no longer as an “added value” but a necessity. Without Software Defined Storage, the solutions we implement in our datacenter will continue maintaining dependencies and will be restricted on scaling out.

In the following articles, we will review on the alternatives we have with SDS, vendors, comparing solutions and so on.

Or being able or not being able to write, those are the questions. It’s being a while since my last post, so I thought it could be a good idea to dedicate it to the “hard time” that represents for me writing. I’m pretty sure that there are plenty of us that experience the same issues.

In this post what I would like to review (more like a retrospective to myself I must say) about the common issues that bloggers might find when we try to convince that we “should being writing/posting/blogging something”. And also some of the best practices that I’m currently finding on trying to avoid writers or bloggers block.

Image used from article: “Unblocking Writer’s Block”

The Importance of Writing

For me, writing has being always an important part of my life and, being honest, having a blog gave me tons of opportunities and for sure made me a better professional and communicator. And that’s the keyword you should think of, it’s not about writing something you know, it’s about sharing and communicating.

Communication is a key element in every professional’s life, you can be the best in what you do but if you are not able to communicate then you will not get the recognition you deserve. And I’m not mentioning “recognition” as a narcissism quality, if you work with customers (and I do believe everyone works with customers in different types, shapes and forms), effective communication is the one thing that will keep customers satisfied about your service, product or both.

I’m planning on having a set of posts regarding effective communication and other topics later on, but I just want to share one quick example on the professional differentiator for having a blog. A few years back, I had a small project for a customer, they wanted to try out a new and emerging technology as a proof-of-concept: App-V.

Back then, the App-V current version was the 4.5 and there were not available a lot of documentation about it, just quick start guides. So, I started working and of course a lot of troubleshooting started from installing and sequencing applications. Finally the proof-of-concept ended really well, the customer and myself loved the technology. And I decided, since I had a “hard time” getting all the pieces right, to write a set of posts to share what I’ve learned (Part I, II, III and IV of the App-V Step by Step set of articles).

Those set of posts, for a long time were actually the first search result when someone typed “app-v step-by-step”. That lead to Packt Publishing offer me to write a book about App-V, the “Getting Started” guide which also led to the “Advanced Guide” for App-V one year later. And a customer found my blog by a search and requested for me to provide a specialized and customized App-V training course in the US; that was my first trip to NY City and the US.

Writing those books or teaching a course were never a goal of mine, I just wanted to share what I’ve learned about the technology and improve my communication in the process, eventually I ended up gaining a whole lot more than that.

The issues that I found in the last couple of months to get back on track on some of the writing were many. Here’s a list that I’m sure there will be lots of you with the same problems:

• I can never find time: Writing takes time, for some of us much more time than others, and with a demanding day job and trying to use the weekends to get some rest, sometimes seems just impossible.
• Words and ideas just don’t flow out of my system: The simpler way to explain it is the “bloggers block” and the more you delayed your writing, the harder it gets.
• Procrastinating soon as I sit down and start writing: Since it’s always harder to get back on the writing motion, soon as I start trying, I start thinking and doing other things not related to my main goal.
• There’s too much information I’m interested to review all and write about it: Like most of you geeks out there, I have several interests in the IT world that I try to keep up and with that generating my own opinion, contents and such. But sometimes seems too much information all the time to try to read it all that is available.

10 Tips and Best Practices to Avoid Bloggers Block

Again, these are just a few I’m trying to apply in my own personal life, hopefully 2015 will work out better than the year before regarding writing and publishing.

1. In order to write, you need to read

And no, this is not a catch-phrase using the same method as the Sphinx in “Mystery Men”: “He who questions training only trains himself at asking questions” (?)

Enhancing your capabilities to elaborate more easily requires for you to train your mind in the right direction. Reading blogs, articles, forums and comments gives some traction to your mind to think regarding opinions and perspective, which will lead for your writing to flow without much problems.

All of you should already know Feedly to organize your RSS feeds, it is the best tool available since Google Reader disappeared. I also personally use Nextgen Reader in my desktop and Windows Phone to keep up with the reading.

2. Tweet, Comment and Interact

This is also a great way to motivate your mind into perspective and writing, the more you interact the easier will be to start writing. Putting your opinions in motion will help you on finding the things you want to write about; Twitter, as the “micro-blogging” network, is a great option to start.

Interacting in other people’s blog in their comments is going to be really useful as well. I always try to ensure that my profile shows up in every comment I make, people interacting with me in comments can review my blog and provide ideas as well on the things I’ve written.

3. Free writing. Write along

Free your mind from any structure, forget about titles, subtitles. Never mind trying to get the right introduction or a catchy phrase that can lure readers. Just write, write along.

This is a great technique I’ve been trying lately when the ideas don’t come into my head. I just start writing about anything that comes into my mind. This post actually started like that, thinking about “I don’t know what to write about. I need time to read and understand something new, but always takes me too much time”.

And if you are already working with an article, write the things you have in your head without giving it too much formality or shape. The idea is to make the writing flow, and good ideas will come eventually.

4. Start the article backwards or start from the middle

This is something related to the last recommendation about free writing. Once you have a solid idea about what you want to write about, don’t structure yourself to get the entire format of the article at the beginning. Start with the part that suits you better, the “low-hanging fruit”.

It’s quite common for me to get an idea about what I want to write, but getting trapped just in the start because I can’t find a right introduction. It happened quite often when I wrote the books, starting a chapter was too hard.

When that happens, I try to start by the easy part. Sometimes could be the ending, sometimes could be the middle of the article. For example, if it’s something technical, start writing with the step-by-step process; the rest will come along later.

5. Commit to consistency and deadlines, but be realistic

When you start writing commit yourself to two things: Consistency/periodicity regarding your writing and deadlines, and only do it if it’s going to be a realistic commitment.

If you have a bloggers block don’t start saying “I’ll dedicate two hours every day and publish one article per week”. Sounds like a great plan, but being honest with yourself, if you didn’t write at all for weeks (or even months like me), don’t try to jump into something that is going pretty unlikely to be completed and will only frustrate you.

Maybe a good place to start would be writing on Saturdays and a monthly post. Once you get the consistency, you can move forward with stronger commitments.

6. Write your thoughts at any place and any time

I often find myself during the day, in any type of activity, with ideas that comes into my head, and if I don’t write them down, just as they are, raw and simple, that thought is gone. I must admit that I do have memory problems, so this is a big issue for me.

What I’m trying to do is to write those ideas in my phone (try Evernote if you didn’t already), or write them down in a notebook. Later on I review them to keep on a train of thought I had hours or days ago.

“The moment a man sets his thoughts down on paper, however secretly, he is in a sense writing for publication” – Raymond Chandler

7. Short posts work as well as any other

If what you want to write is concise, simple and short, then so be it. A short post can have the same or even better impact than a larger one.

This is something that it’s hard for me even to this day, I have my mind set already that when I want something to write should be using an extended topic. But if you want to share an opinion, something you learned recently, and it doesn’t take much writing, then post it.

Using this will also facilitate your focus and you will practicing your writing. Sharing frequently in your blog will also encourage readers to interact more often.

8. Maintain a healthy environment while you are writing

This is a broad topic and could easily be a post on its own. Having a healthy environment it refers to several meanings and options, here’s just a summary about my best practices on that:

• Keep your desk neat, clean and organized. Order inspires order, an organized workplace will keep your mind focused.

Can you actually be motivated to write if your desk looks like this?

• Your writing place should be a distraction-free space. It could sound obvious, but sometimes we don’t notice that a TV nearby or street sounds if you sit by the window could be quite distracting.

• Always keep a bottle of water in your desk and have healthy snacks. Drinking water regularly helps your digestion, circulation, absorb nutrients more easily and maintaining the body temperature. All of those things are important to maintain your mind sharp. Healthy snacks as fruits also gives the necessary components to your body and brain to function properly, eating trash meals and snacks will do the exact opposite.

• Coffee it’s not my main option while writing. Even though caffeine is a must for several of technology guys, I try to stay away during writing hours. Coffee forces the body to eliminate water and nutrients are harder to absorb while you are drinking.

• Ergonomic position while you in front of the monitor. Not only sitting correctly but also having your keyboard, mouse and monitor in the right position. You usually don’t notice, but getting used to an unhealthy position can lead for back, neck, wrists pains and chronic problems. Here’s a nice and complete overview of some of the best practices regarding sitting correctly: “How to Sit at a Computer”.

• Take regular and scheduled brakes. And don’t stay in front of the computer while lunch or dinner, leave the working area.

• Try to be fully dressed while you are writing. Doesn’t matter if you write during early morning, or late at night, if you are wearing pajamas or just your underwear, it’s most likely that you won’t be completely focused on writing but more about your bed or laying in your couch instead.

9. Use Microsoft Word and thesaurus within

I always start writing my articles in Microsoft Word, and it’s not only to use review the spelling and grammar, the other important tool I have is the thesaurus providing synonyms and related concepts.

My native language it’s not English, so finding the right words could be difficult sometimes and could be a stopper if you are in the middle of an idea that you cannot set the right words. Using Shift+F7 in any word written, you’ll get in the right panel the synonyms you need for that thing you wanted to express differently.

10. Sometimes it just going to take more time that you thought so

Don’t get frustrated, please don’t. And don’t get too perfectionist, once you start reading your article again and again, you will always find something you want to change, add or delete; don’t get yourself into an infinite loop about getting the right article.

Writing sometime takes time, for some of us longer than others, it’s just the way it is. And there’s one thing certain: Practice is the only medicine to solve that.

That’s pretty much it for now.

How do you handle your own bloggers block? What type of issues do you find the most while trying to write?

If you don’t know StarWind iSCSI SAN product and you currently handling clusters that require a shared storage (not necessarily Windows), I highly recommend to take a look around to this the platform. To summarize, StarWind iSCSI SAN represents a software which allows you to create your own shared storage platform without requiring any additional hardware.

I created a post a while ago about “Five Easy Steps to Configure Windows Server 2008 R2 Failover Cluster using StarWind iSCSI SAN” to explain how can a Failover Cluster can be easily configured with the help of StarWind iSCSI SAN. Since there has been some changes in the latest releases of Windows Server and StarWind iSCSI SAN has a brand new v8 of its platform, I thought it would be a good idea to create a new article to achieve an easy way to create our own cluster.

As I did, for the previous post, the main idea about this article is to show a simple step-by-step process to get a Windows Server 2012 R2 Failover Cluster up and running, and without requiring to use an expensive shared storage platform to complete it. The steps involved are:

1. Review and complete pre-requisites for the environment.
2. Install StarWind iSCSI SAN software.
3. Configure and create LUNs using StarWind iSCSI SAN.
4. Install Failover Cluster feature and run cluster validation.
5. Create Windows Server 2012 R2 Failover Cluster.

1. Review and Complete Pre-Requisites for the Environment

Windows Server 2012 introduced some changes into the Failover Cluster scenarios, even though those are important and improved changes, the basic rules of Failover Cluster has not changed. Here are the requirements for a Windows Server 2012 R2 Failover Cluster.

Requirements for Windows Server 2012 R2 Failover Cluster

Here are the requirements in Windows Server 2012 R2 for Failover Clusters:

• Two or more compatible servers: You need hardware that is compatible with each other, highly recommended to always use same type of hardware when you are creating a cluster. Microsoft requires for the hardware involved to meet the qualification for the “Certified for Windows Server 2012 logo”, the information can be retrieved from the Windows Server catalog.
• A shared storage: This is where we can use StarWind iSCSI SAN software.
• [Optional] Three network cards on each server, one public network (from which we usually access Active Directory), a private for heartbeat between servers and one dedicated to iSCSI storage communication. This is actually an optional requirement since using one network card is possible but not suitable in almost any environment.
• All hosts must be member from an Active Directory domain. To install and configure a cluster we don’t need a Domain Admin account, but we do need a Domain account which is included in the local Administrators of each host.

Here are some notes about some changes introduced in Windows Server 2012 regarding requirements:

We can implement Failover Cluster on all Windows Server 2012 and Windows Server 2012 R2 editions, including of course Core installations. Previously on Windows Server 2008 R2 the Enterprise or Datacenter Edition were necessary.

Also the concept for “Active Directory-detached cluster” appears in Windows Server 2012, which means that a Failover Cluster does not require a Computer object in Active Directory, the access is performed by a registration in DNS. But, the cluster nodes must still be joined to AD.

Requirements for StarWind iSCSI SAN Software

Here are the requirements for installing the component which will be in charge of receiving the iSCSI connections:

• Windows Server 2008 R2 or Windows Server 2012
• Intel Xeon E5620 (or higher)
• 4 GB of RAM (or higher)
• 10 GB of disk space for StarWind application data and log files
• Storage available for iSCSI LUNs: SATA/SAS/SSD drive based arrays supported. Software based arrays are not supported in iSCSI.
• 1 Gigabit Ethernet or 10 Gigabit Ethernet.
• iSCSI ports open between hosts and StarWind iSCSI SAN Server. The iSCSI ports are 3260 and 3261 for the management console.

General Recommendations for the Environment

In this scenario, there are several Microsoft and StarWind recommendations we must fulfill in order to get the best supportability and results. Keep in mind that each scenario could require different recommendations.

To mention some of the general recommendations:

• NIC Teaming for adapters, except iSCSI. Windows Server 2012 improved significantly the performance and of course supportability of network adapters teaming and is highly recommended to use that option for improved performance and high-availability. But we must avoid configure teaming on iSCSI network adapters.

Microsoft offers a very detailed document about handling NIC teaming in Windows Server 2012: “Windows Server 2012 NIC Teaming (LBFO) Deployment and Management” and also check this article “NIC Teaming Overview”.

• Multi-path for iSCSI network adapters. iSCSI network adapters prefer handling MPIO instead of NIC teaming, because in most scenarios the adapter throughput is not improved and moreover there could be some increases in response times. Using MPIO is the recommendation with round-robin.
• Isolate network traffic on the Failover Cluster. It is almost mandatory that we separate iSCSI traffic from the rest of networks, and highly recommended to isolate the rest of traffic available. For example: Live Migration in Hyper-V clusters, management network, public network, or Hyper-V replica traffic (if the feature is enabled in Windows Server 2012).
• Drivers and firmware updated: Most of hardware vendors will require prior to start any configuration, like a Failover Cluster, to have all drivers and firmware components updated to the latest version. Keep in mind that having different drivers or firmware between hosts in a Failover Cluster will cause to fail the validation tool and therefore the cluster won’t be supported by Microsoft.
• Leave one extra LUN empty in the environment for future validations. The Failover Cluster Validation Tool is a great resource to retrieve detailed status about the health of each cluster component, we can run the tool whenever we want and it will not generate any disruption. But, to have a full “Storage Validation” it is required to have at least one LUN available in the cluster but not used for any service or application.

For more information about best practices, review the following link: “StarWind High Availability Best Practices”.

One important new feature introduced by StarWind iSCSI SAN v8 is the use of Log-Structured File System (LSFS). LSFS is a specialized file system that stores multiple files of virtual devices and ensures high performance during writing operations with a random access pattern. This file system resolves the problem of slow disk operation and writes data at the speed that can be achieved by the underlying storage during sequential writes.

At this moment LSFS is experimental in v8, use it carefully and validate your cluster services in a lab scenario if you are planning to deploy LSFS.

2. Install StarWind iSCSI SAN software

After we reviewed and verified the requirements, we can easily start installing StarWind iSCSI SAN software, which can be downloaded in trial-mode. This represents the simplest step in our list, since the installation does not have any complex step.

In the process, the Microsoft iSCSI service will be required to add to the server and the driver for the software.

After the installation is complete we can access our console and we will see as a first step necessary is to configure the “Storage pool” necessary.

We must select the path for the hard drive where we are going to store the LUNs to be used in our shared storage scenario.

3. Configure and create LUNs in StarWind iSCSI SAN

When we have the program installed, we can start managing it from the console and we will see the options are quite intuitive.

We are going to split the configuration section in two parts: Hosting iSCSI LUNs with StarWind iSCSI SAN and configuring our iSCSI initiator on each Windows Server 2012 R2 host in the cluster.

Hosting iSCSI LUNs with StarWind iSCSI SAN

We are going to review the basic steps to configure the StarWind iSCSI to start hosting LUNs for our cluster; the initial task is to add the host:

3.1 Select the “Connect” option for our local server.

3.2 With the host added, we can start creating the storage that will be published through iSCSI: Right-click the server and select “Add target” and a new wizard will appear.

3.3 Select the “Target alias” from which we’ll identify the LUN we are about to create and then configure to be able to cluster. The name below will show how we can identify this particular target in our iSCSI clients. Click on “Next” and then “Create”.

3.4 With our target created we can start creating “devices” or LUNs within that target. Click on “Add Device”.

3.5 Select “Hard Disk Device”.

3.6 Select “Virtual Disk”. The other two possibilities to use here are “Physical Disk” from which we can select a hard drive and work in a “pass-through” model.

And “RAM Disk” is a very interesting option from which we can use a block of RAM to be treated as a hard drive or LUN in this case. Because the speed of RAM is much faster than most other types of storage, files on a RAM disk can be accessed more quickly. Also because the storage is actually in RAM, it is volatile memory and will be lost when the computer powers off.

3.7 In the next section we can select the disk location and size. In my case I’m using E:\ drive and 1GB.

3.8 Since this is a virtual disk, we can select from either thick-provision (space is allocated in advance) or thin-provision (space is allocated as is required). Thick provisioning could represent, for some applications, as a little bit faster than thin provisioning.

The LSFS options we have available in this case are: “Deduplication enabled” (procedure to save space since only unique data is stored, duplicated data are stored as links) and “Auto defragmentation” (helps to make space reclaim when old data is overwritten or snapshots are deleted).

3.9 In the next section we can select if we are going to use disk caching to improve performance for read and writes in this disk. The first opportunity we have works with the memory cache, from which we can select write-back (asynchronous, with better performance but more risk about inconsistencies), write-through (synchronous, slow performance but no risk about data inconsistency) or no cache at all.

Using caching can significantly increase the performance of some applications, particularly databases, that perform large amounts of disk I/O. High Speed Caсhing operates on the principle that server memory is faster than disk. The memory cache stores data that is more likely to be required by applications. If a program turns to the disk for data, a search is first made for the relevant block in the cache. If the block is found the program uses it, otherwise the data from the disk is loaded into a new block of memory cache.

3.10 StarWind v8 adds a new layer in the caching concept, using L2 cache. This type of cache is represented in a virtual file intended to be placed in SSD drives, for high-performance. In this section we have the opportunity to create an L2 cache file, from which again we can select to use it as write-back or write-through.

3.11 Also, we will need to select a path for the L2 cache file.

3.12 Click on “Finish” and the device will be ready to be used.

3.13 In my case I’ve also created a second device in the same target.

Configure Windows Server 2012 R2 iSCSI Initiator

Each host must have access to the file we’ve just created in order to be able to create our Failover Cluster. On each host, execute the following:

3.14 Access “Administrative Tools”, “iSCSI Initiator”.

We will also receive a notification about “The Microsoft iSCSI service is not running”, click “Yes” to start the service.

3.15 In the “Target” pane, type in the IP address used for the target host, our iSCSI server, to receive the connections. Remember to use the IP address dedicated to iSCSI connections, if the StarWind iSCSI SAN server also has a public connection we can also use it, but the traffic will be directed using that network adapter.

3.16 Click on “Quick Connect” to be authorized by the host to use these files.

Once we’ve connected to the files, access “Disk Management” to verify we can now use these files as storage attached to the operating system.

3.17 And as a final step, just using the first host in the cluster, put “Online” the storage file and select also “Initialize Disk”. Since these are treated as normal hard disks, the process for initializing a LUN is no different than initializing a physical and local hard drive in the server.

Now, let’s take a look about the Failover Cluster feature.

4. Install Failover Cluster feature and Run Cluster Validation

Prior to configure the cluster, we need to enable the “Failover Cluster” feature on all hosts in the cluster and we’ll also run the verification tool provided by Microsoft to validate the consistency and compatibility of our scenario.

4.1 In “Server Manager”, access the option “Add Roles and Features”.

4.2 Start the wizard, do not add any role in “Server Roles”. And in “Features” enable the “Failover Clustering” option.

4.3 Once installed, access the console from “Administrative Tools”. Within the console, the option we are interested in this stage is “Validate a Configuration”.

4.4 In the new wizard, we are going to add the hosts that will represent the Failover Cluster in order to validate the configuration. Type in the server’s FQDN names or browse for their names; click on “Next”.

4.5 Select “Run all tests (recommended)” and click on “Next”.

4.6 In the following screen we can see a detailed list about all the tests that will be executed, take note that the storage tests take some time; click on “Next”.

If we’ve fulfilled the requirements reviewed earlier then the test will be completed successfully. In my case the report generated a warning, but the configuration is supported for clustering.

Accessing the report we can get a detailed information, in this scenario the “Network” section generated a warning for “Node <1> is reachable from Node <2> by only one pair of network interfaces. It is possible that this network path is a single point of failure for communication within the cluster. Please verify that this single path is highly available, or consider adding additional networks to the cluster”. This is not a critical error and can easily be solved by adding at least one new adapter in the cluster configuration.

4.7 Leaving the option “Create the cluster now using the validated nodes” enabled will start the “Create Cluster” as soon as we click “Finish”.

5. Create Windows Server 2012 R2 Failover Cluster

At this stage, we’ve completed all the requirements and validated our configuration successfully. In the next following steps, we are going to see the simple procedure to configure our Windows Server 2012 R2 Failover Cluster.

5.1 In the “Failover Cluster” console, select the option for “Create a cluster”.

5.2 A similar wizard will appear as in the validation tool. The first thing to do is add the servers we would like to cluster; click on “Next”.

5.3 In the next screen we have to select the cluster name and the IP address assigned. Remember that in a cluster, all machines are represented by one name and one IP.

5.4 In the summary page click on “Next”.

After a few seconds the cluster will be created and we can also review the report for the process.

Now in our Failover Cluster console, we’ll get the complete picture about the cluster we’ve created: Nodes involved, storage associated to the cluster, networks and the events related to cluster.

The default option for a two-node cluster is to use a disk as a witness to manage cluster quorum. This is usually a disk we assign the letter “Q:\” and does not store a large amount of data. The quorum disk stores a very small information containing the cluster configuration, its main purpose is for cluster voting.

To perform a backup for the Failover Cluster configuration we only need to backup the Q:\ drive. This, of course, does not backup the services configured in the Failover Cluster.

Cluster voting is used to determine, in case of a disconnection, which nodes and services will be online. For example, if a node is disconnected from the cluster and shared storage, the remaining node with one vote and the quorum disk with also one vote decides that the cluster and its services will remain online.

This voting is used as a default option but can be modified in the Failover Cluster console. Modifying it depends and is recommended in various scenarios: Having an odd number of nodes, this case will be required to use as a “Node Majority” quorum; or a cluster stretched in different geographically locations will be recommended to use an even number of nodes but using a file share as a witness in a third site.

For more information about quorums in Windows Failover clusters, review the following Microsoft TechNet article: “Configure and Manage the Quorum in a Windows Server 2012 Failover Cluster”.

More Resources

To review more information about Windows Server 2012 R2 clusters and StarWind iSCSI SAN review the following links and articles:

• “What’s New in Failover Clustering in Windows Server 2012 R2”
• “Failover Clustering Hardware Requirements and Storage Options”
• “Validate Hardware for a Failover Cluster”
• “Deploy a Hyper-V Cluster”
• “Use Cluster Shared Volumes in a Failover Cluster”
• “Network Recommendations for a Hyper-V Cluster in Windows Server 2012”
• “Cluster-Aware Updating Overview”
• “StarWind Software Technical Papers”

After the general overview about application packaging, benefits and best practices in Part I of this series, it is time to start working understanding existing installation packages, handle some examples and reviewing step-by-step processes we will need to start packaging applications.

In this second post, the focus will be packaging applications without the use of any 3^rd party tool, basically achieving applications deployment with silent and customized installations. What we will review in this article:

1. MSI Files are your New Best Friends
a) Reviewing the Windows Installer
b) Using MSIEXEC.EXE
2. Understanding and Identifying Installer Vendors
a) InstallShield Overview
b) Wise Installation System Overview
c) Inno Setup Overview
d) SetupBuilder and ActualInstaller Overview
3. Packaging Examples for Silent and Customized Installations: First Approach
a) Finding our MSI when there’s no MSI
b) Creating and Using “.iss” Answer Files
c) Troubleshooting InstallShield Deployments

MSI Files are your New Best Friends

So, after we’ve decided that silent installations is going to be our first stop in packaging applications, we must know that MSI files provides the best way to achieve that.

An MSI file is a container that works with an embedded engine in the OS (Windows Installer service) in order to facilitate applications installations and uninstallations. This platform (MSI + Windows Installer service) contains all necessary instruments to accomplish a customized, automated and silent installation.

Later in the next following posts we will discuss on the components in an MSI file, but initially we can take a quick overview about the Windows Installer platform, the MSI files and how we should interact with them.

Reviewing the Windows Installer

The Windows Installer service appeared embedded in the Windows OS since Windows 2000 and with the solution to several problems about inconsistencies in the software vendors on how they were using the application installers, typically in setup.exe files.

The Windows Installer services function is to support the application life-cycle: installation and customization, maintenance (auto-repair and patching for example) and retirement. To process all of these steps is msiexec.exe. MSIEXEC takes all the information contained in the MSI file (installation database, a summary information stream, and data streams for various parts of the installation) and other optional (in most cases) files, to perform the installation or desired process.

These other files can be MST (Windows Installer Transform), which contains customizations for the installation (for example, features to be installed); MSP (Windows Installer Patch) representing application update/patch.

The complete list for installation files available:

It is important to remember that MSI files support silent installations natively; meaning that every application containing one will be suited for this type of installation.

We will take a closer look to the MSI components in the following posts when we start building our own. For now, we will focus on how to use them in silent installations.

Using MSIEXEC.EXE

As we were talking, msiexec.exe represents the engine from which the Windows Installer performs any of the life-cycle tasks of each application. Using msiexec, if you haven’t before, it’s simple and should not represent any problem.

Once we have our MSI file, using it for silent installations should be in most cases as the following example:

msiexec.exe /I <path_to_MSI.msi> /qn

• Using “/I” parameter at the beginning we are configuring MSIEXEC to install or configure the application (also known as “product”).
• The “/qn” parameter represents for a “quiet” (q) installation (no user interaction) and “No User Interface” (n). This way, the application will complete its installation without prompting any notification for the user.

If everything about packaging and deploying applications would be represented by MSI files, then these articles won’t be necessary. We’ll see that most of the times, we will have to deal with EXE files for deployment.

Also, using these EXE files could have different methods according the vendor that created the installer. In the following section we will review how to handle those situations.

Understanding and Identifying Installer Vendors

Even though there are plenty of EXE files which contain MSI files within (to be reviewed later in this post), there are other files with no MSI file available but supporting a silent installation method. To identify how we can silent install the application we must first understand which installer vendor (if any) was used for this app. Identifying is not always easy, the installer icon available could be helpful and sometimes we will just have to guess.

The most common vendors involved in installation files are:

• InstallShield, from Flexera Software
• Wise Installation System (formerly InstallMaster), from Symantec (previously from Wise)
• SetupBuilder, from LinderSoft
• ActualInstaller, from Softeza Development
• Inno Setup, from JR Software (open source project)

InstallShield Overview

This is probably the most common vendor involved for developing installation files. We can easily identify their installers by reviewing the logo involved.

There are also two ways we can deploy InstallShield packages, which actually depends if the MSI file was included in the installation package:

1. If the MSI is included in the package, the InstallShield installer should support a silent installation using this example (the /v parameter is optional for verbose):

setup.exe /s /v"/qn"

2. If the MSI is not included there’s an interesting method for some legacy installers to use “answer files”, InstallShield calls this “Package for the Web”. We will need to run an installation with specific parameters to generate our answer file (“.iss”) that we can later use for deploying the application silently and with specific parameters. We will discuss the details later in this article.

The “Package for the Web” has been discontinued by Flexera Software, and it will no longer be available for InstallShield packages. But, most likely, in most organizations we can still find this alternative as a valid method.

To review the complete parameters available for InstallShield check this link: “Setup.exe Command Line Parameters”

Wise Installation System Overview

You will find this installation file type in some legacy applications in your environment. Wise Solutions was the creator of this platform, which originally started with CompuServe; later on was acquired by Altiris Inc, which also had been acquired by Symantec on 2007. The last release of this suite was 9.0 but has been already discontinued by Symantec.

The applications created with this platform support the classic “/s” parameter for silent installations.

setup.exe /s

But unfortunately there are no other options available for customization and the exit status when you executed the process was inconsistent, so it might get a little bit confusing.

Inno Setup Overview

This open source project has been receiving significant contribution of the community in all the years it has been available (since 1997), supports all Windows OS, including Windows 8 and Windows Server 2012.

The command line used for supporting silent installations in this suite are:

setup.exe /sp- /verysilent /norestart

• “/sp-“: Disables the “This will install… Do you wish to continue?” prompt at the beginning of Setup.
• “/verysilent”: Will execute silently the installation process. The “/silent” parameter is available but, even though it will not require users’ intervention will show a progress bar.
• “/norestart”: Prevents automatic restarts, which can be very useful if the “verysilent” parameter is used since it will automatically restart the OS without the user intervening.

For more information about Inno command lines available: “Setup Command Line Parameters”.

SetupBuilder and ActualInstaller Overview

SetupBuilder and ActualInstaller represent a couple of somehow new installation packagers’ vendors available in the market. Like many of these recent suites, both platforms support Windows 8 and Windows Server 2012.

One of the most common standards for recent installation packagers is the use of the same type of parameters for the silent and customized installations. SetupBuilder and ActualInstaller have the same silent installation method:

setup.exe /S

Among other providers we will find the same parameters that can be used, we have to be aware that sometimes the parameters are case-sensitive.

Packaging Examples for Silent and Customized Deployments: First Approach

Let’s review some of the examples we can find when we want to package applications using silent and customized deployments.

Finding our MSI when there’s no MSI

One common situation that we might find is that the application we are trying to deploy does not contain an MSI file, instead we are provided with an exe file that could not support silent installations. But, fortunately, most times there are some secrets hiding in plain sight.

When we are handling EXE installer files (usually as setup.exe), several times we are using an installer file that is actually a packaged set of files that contain an MSI, installation instructions and other optional files that could be used during the installation.

Companies use this alternative to have more scalable installation procedures that cannot be achieved with one MSI file. For example: Adding several MSI files which can be activated depending on the type of installation selected by the user. One example for this situation is Apple’s iTunes.

There are a two ways to extract the needed MSI file/s. Let’s review them using Google Earth as our application:

In the particular case of Google Earth, we need to download the standalone installer for the application available in this link.

1. Double click the EXE installer file and when the installation wizard starts access %TEMP% folder to locate our MSI. This process is called “expanding EXE packages”.

Most of the times, the files are located in a folder using a random name (in this case %TEMP%\._msige61).

2. Use 7-Zip File Manager to browse inside the EXE file. This free tool is quite useful, all we need to do is use the option “Open Inside” and we can browse around the installer and handle all the files within.

3. Some software vendors provide a command line to extract the files without the need to double click them or using 7-Zip File Manager.

Once we have our MSI, the process for a silent installation is simple:

msiexec /i “Google Earth.msi” /qn

Creating and Using “.iss” Answer Files

As we reviewed earlier in this article, InstallShield represents one of the vendors commonly used by software companies to create their installers. When we handle this type of installer, we get the simple option to use the “/s /qn” parameter or we might need to create an answer file.

This answer or response file is represented by our “.iss”. This file must be created running a command line which will install the software normally with all the desired options, but creating in the process the .iss file to be used in other deployments. Since every time we need to create an answer file we will need to install the software, using virtual machines and snapshots will be quite useful at this point.

InstallShield also supports the use of different compiled scripts, .ins files. Only applies for InstallScript and InstallScript MSI projects only, InstallShield version 12 and earlier.

The process for creating response files and then deploying the package is the following:

1. Locate your setup.exe (or the installer’s name) and a target machine to deploy the application (ideally a virtual machine).

2. Open a command line and execute the following to create the “setup.iss” response file:

setup.exe /r

The response file will be created in %WINDIR% (typically C:\Windows), but we can alternatively change the path by using the following command line:

setup.exe /r /f1”C:\setup.iss”

Note that there’s no space between –f1 and the “<>“ with the destination path.

3. Install the application with all the customization’s necessary.

4. Once is completed, locate the setup.iss file created.

5. To deploy the application, if we are using the .iss file with the default name (“setup.iss”) and is located in the same folder as the installer, we can deploy our package using the following command line:

setup.exe /s /v/qn

If we changed the name of the response file or we have it located in a different folder, the command line should be the following:

setup.exe /s /f1”<.iss name and location>”

Keep in mind that there’s no space between –f1 and the “<>“ with the file location.

6. [Optional] We can use /f2 as a parameter to create a log file for the deployment:

setup.exe /s /f1”<.iss name and location>” /f2”<.log name and location>

If we don’t set log file name and location, it will automatically generate “setup.log” file.

Troubleshooting InstallShield Deployments

In case there’s a deployment issue with our package, creating the setup.log would be mandatory in order to find out the exact problem we have.

The Setup.log file contains three sections.

• [InstallShield Silent]: Identifies the version of InstallShield Silent used in the silent installation.
• [Application]: Details Installed application’s name and version, and the company name.
• [ResponseResult]: The result code indicating whether or not the silent installation succeeded.

The result codes can be one of the following:

-0 Success.
-1 General error.
-2 Invalid mode.
-3 Required data not found in the Setup.iss file.
-4 Not enough memory available.
-5 File does not exist.
-6 Cannot write to the response file.
-7 Unable to write to the log file.
-8 Invalid path to the InstallShield Silent response file.
-9 Not a valid list type (string or number).
-10 Data type is invalid.
-11 Unknown error during setup.
-12 Dialogs are out of order.
-51 Cannot create the specified folder.
-52 Cannot access the specified file or folder.
-53 Invalid option selected.

In the next article we will review more details about silent and customized deployment, including more examples in more complex scenarios.