Sustainable Technology

Unicorn Tamer

2010-07-23T18:55:00.000-07:00

This week at work an email came out raising the discussion of job titles, referencing a series of recent articles sharing the notion of using creative descriptions such as Director of Culture etc. I responded suggesting that mine be Tamer of Unicorns.
Unicorns have been a big deal this week, mainly as a mechanism to acknowledge front and center that some words are overloaded in technology groups – in this particular instance it was Agile, so rather than wrestle with the notion that Agile had been tried already, I started calling the development methodology the Pink Unicorn, clearly differentiating it from anything tried previously here. Agile is a collection of approaches tools and techniques that is used to deliver software iteratively and collaboratively.

My preferred approach is to blend Iterative development with Continuous Integration and a healthy dose of Test Driven development. More to come on those methodologies later. Last week Chad emailed me a link to a really cool project IHMC CMaps which is a very simple tool for the capture of concept maps, I used it to develop a way of talking about all the elements of the Pink Unicorn and show how interrelated they all were. Here is a monster diagram aggregating all the concepts from culture, to tools, to techniques.

NB305 and a battery powered lawn mower

2010-07-17T19:07:00.000-07:00

Many of the people that follow my blog will know by now that I am again gainfully employed. At my new place of work remote desktop access is provided to facilitate productivity. I can create an RDP session directly to my desktop and have full access to my entire desktop environment. Initially I was skeptical about the very high proportion of desktop v laptop machines in the office, but having lived with the option of remote access for a few weeks I can now see how it provides a tremendous productivity boost opportunity. I will explain:

My work desktop is a big tower, 2.4GHz 6GB Ram kick-ass-box, anything I need to do flies, I can run three instances of Visual Studio and it doesn’t flinch.
Last week I purchased a Toshiba NB305 with 1 1.6GHz Atom processor 1GB Ram, it runs like a whipped dog… but… if I connect (RDP) to my work desktop it flies. All it has to do is have sufficient horse power to render the graphics and the connectivity to support moving them, none of the apps are local and so my experience (which btw is how I am typing this blog entry) is as rapid as I could ever hope for – and a 10hr battery life

Similarly I can connect from my home PC to work and from the Netbook to the home PC also now (I will blog on how I have done that later). At home I have a quad-proc-six-core beast with more memory than the Smithsonian library, it has three screens each capable of HD+ quality. The experience of working from my home office connected to work is as if I am there (from a connectivity and technology performance perspective, but not yet from the collaboration perspective).

So what the heck has this got to do with a battery powered lawn mower? About three weeks ago when I became a producer of cash and not a burden on my wife’s income I decided it was time I stopped moving the gas powered mower back and forth between the house in the ‘burbs and the cabin, so I bought an additional one, not a gas one, a rechargeable electric one, J expressed her doubt, and I still have mine…. After being charged for 12 hours I turned the switch and away I went. A couple of learning’s so far: current battery powered mowers are heavy and are not self propelled. The mower is very quiet, the main noise being the blades spinning, when I hit tall grass I can hear the blades slow. It is not clear if the motor has enough power to cut grass in most conditions, however what is clear is that it is now an aerobic exercise now to mow the grass, but at least I am not burning gas to do it.

If you have read my previous posts, you will understand that I think that powerful desktop machines still have a place in providing a “green” working environment (Type 1 hypervisors required), my recent experiences have strengthen that belief. Combine those carbon savings with the absence of yet another gas mower and maybe I managed to decrease my carbon footprint a bit this month.

Development Server Sprawl

2010-05-05T11:37:00.000-07:00

It is rare for a company’s production environment to be fully geographically redundant to such a level as to fully take advantage of remote locations to provide fault tolerant high availability. More frequently some element of the application suite’s stack of technology will be unable to seamlessly survive a site-wide outage, be it the aging ERP or the spaghetti of heritage applications that were glued together over the past decade and now called an integrated product offering with a glitzy web front end. Recognizing the reality that most applications in production are highly available only by the design of the infrastructure residing in a single site, and possible, and hopefully, recoverable over some extended time period in another site, leads most organizations to require their production datacenters to provide considerable resilience. Whilst many organizations will not invest in true highly available datacenter facilities with dual systems capable of being maintained and failing without impacting the technology they support, many organizations are investing in substantially redundant facilities, dual PDU and power sources, multiple network providers, redundant cooling. The proliferation of facilities redundancy has created a scenario where data centers space is costing upwards of twenty five thousand dollars per delivered kilo watt of power available to support the compute load, a typical 2.5MW facility costing in excess of sixty million dollars capital before the first server is racked. The price differential between high and lower resiliency facilities is considerable, so too are the operating expenses if one assumes no incremental cost when the facilities are taken offline for maintenance.There are few large corporations in the US that are not currently facing the tough decision of how to sustain the growth of their technology footprint while surviving an economic downturn. Most companies with substantial datacenter capacity are watching closely the erosion of their excess datacenter capacity and contemplating the next multi-million dollar investment required to establish the next area of raised floor.
Virtualization as an approach to server consolidation is no longer new and exciting; most CIO’s have established programs and are gradually reaping the rewards. However, virtualization only delays the inevitable day that the space power and cooling runs out of capacity.A challenge with many internally driven infrastructure consolidation initiatives is that the very people who are charged with working out how to decrease the scale and complexity of the environment are the very same people whose livelihood depends upon that complexity and scale. Their ability to objectively pursue the full extent of the opportunity must be challenged to ensure they are motivated to implement the appropriate strategies for the company and not their career.
As I have blogged about previously the developer community, of which I am a proud member, is a bunch of server huggers. Each team, and if they got a chance each developer, asks for their own environment to develop, test, debug, stage promote and finally run in production. Expecting developers to efficiently develop in a shared environment is unrealistic; the complexity of single application debugging is sufficiently high that the introduction of multiple application development environments within a single server instance would provide a level of ambiguity that would reduce productivity unacceptably. So whilst I don’t support the notion of every developer needing their own set of servers, I do support the notion that providing teams isolated environments where only their own actions affect the behavior of their technology is paramount if you wish them to perform efficiently. Why though do development environments get hosted in datacenters? And more specifically why do they get hosted in datacenters supporting production environments? And to be even blunter, why are they being supported by resilient facilities? Let’s face it, development environments only need to be working when a developer is working, is your office complex supported by facilities as resilient as the datacenter? I doubt it. If a development server environment is offline does that prevent the developer from continuing to code on their desktop/laptop? No. So I posit that development server environments should be supported on relatively non resilient facilities with a fairly low RTO/RPO, perhaps measured in multiple hours. What is the size and scale of development server environments?
A recent survey by a company I was working at demonstrated that approximately 18% of the server inventory physically residing in the highly resilient data center space was not supporting production equipment. If one conservatively estimates that half of these servers are development servers this suggests approximately 9% of servers in the datacenter do not need to be there.
One might logically suggest that large organizations should build their own low resiliency facilities to house their development environments; this seems the logical next step. However, how many companies truly have the scale to fill a data center with development serves alone? Tiny datacenters tend not to be as financially efficient both in capital and operating expense as their larger counterparts due to the economies of scale of the mechanical and electrical plant.
The scale of development server needs is exaggerated by the general absence of rigorous decommissioning or repurposing policies. Due to the one server-one use mindset that caused the proliferation in the first place few application development environments are decommissioned until the application is decommissioned from production. These behaviors have led to a bloat of underutilized processors sitting idle waiting for an app to break or a feature to be developed.If an organization has one thousand servers in its datacenters then maybe 80 of them are development servers, those eighty are consuming probably between twenty and thirty kilo watts in aggregate, the capex necessary to build the facilities they exist within equates to roughly half a million dollars. The capital depreciation of datacenter investments is complex due to the varying nature of the building, mechanical and electrical plant, a reasonable estimate is a ten year depreciation period. The estimated annualized depreciation of the space occupied by the development server environment could then be calculated to be fifty thousand dollars.
Virtualization results in tangible savings for most organization firstly from its ability to enable consolidation of servers onto fewer CPU. Over allocation, the ability to slice up infrastructure so that the sum of the virtual parts is greater than the physical whole, varies considerably depending upon application type and operating systems, as well as the underlying infrastructure that is the source and destination for the environments. It has been my experience that development environments are especially suitable to substantial over allocating. This is facilitated by several, facts: Most development servers are substantially idle, due to their activity depending upon human interaction by the developers, and the lifecycle of development being sporadic depending upon a longer term product development lifecycle. Additionally, the performance of development environments is far less critical than that of the production environments enabling developers (begrudgingly) to accept lower performing environments than the target production environment. Server consolidation rates of twenty to one are not unreasonable to expect, especially if the target environment is using modern multi-core chipsets optimized for virtualized environment execution.
Applying the twenty to one consolidation approach to the eighty development servers leads to the suggestion that four new physical servers can provide the processing necessary to support the development server environment and still provide adequate performance and isolation to ensure the developer’s productivity is sustained. The capital outlay for four servers is probably 100k, depreciated over 4 years equaling a 25k anual operating expense. The additional operating expense of powering and cooling these four servers is roughly equivalent to 35,000khw per year assuming a conservative PUE, this becomes roughly $2,500. The annualized savings opportunity for this environment is thus conservatively estimated to be $22.5K. This however is a gross over simplification and under estimation as I will continue to discuss later. Hopefully though this potential savings opportunity wets your appetite sufficiently to warrant you continuing to read on.
What would be an ideal development server environment? Many competing demands are at play in development environments, cost, flexibility, isolation, availability. The consequence of attempting to mitigate these demands leads infrastructure teams to establish an ever increasing foot print of servers as each team, project or consulting group comes in and demands that their own unique environment is established early in a projects initiation.
At any one time it is unlikely that every production application is being modified by the development teams. This leads to the obvious conclusion that multiple development server environments are sitting idle at any particular moment in time. Why are these servers not repourposed for the next development effort? Depending upon the capital appropriation practices of your specific organization it is likely that asset ownership becomes an issue. One specific cost center owner has the development assets sitting on their balance sheet, and this inhibits the opportunity for sharing. The unpredictable nature of discovering issues in production that will need rapid diagnostic and development work to take place, coupled with the relatively slow deployment times or repurposing times of physical servers makes the idea of dedicated environments attractive to teams responsible for break fix work.
A number of years ago VM Ware added a product to its portfolio called Lab Manager. Lab Manager provides a rich user experience to establish and maintain complex environments made up of a combination of guest VMs, it provides the ability to quiesce the guests and capture then store the state of entire application development server environments and rehydrate them on demand. This technology then creates the starting point for mitigating several of the factors that have driven technology organization to bloat their development environments. First the rapid deployment of rarely needed server configurations becomes an efficient and practical exercise. Secondly the decommission of retired environments becomes a right click exercise. Thirdly the establishment of entirely new application development environments take minutes not weeks.
It is challenging for many corporations to change their policies for infrastructure investments without it having substantial impact on reported top and bottom lines of individual lines of business. The migration from individual business unit asset ownership towards a shared infrastructure can get bogged down in financial performance restatement and other financial issues, irrespective of the obvious benefits from implementing the strategy. It then becomes attractive to look outside of the organization for a less financially intrusive solution to the challenge. The economics of shared virtual infrastructure provide the opportunity for an external business to provide on a subscription basis access to shared virtual development environments for a lower cost than most business units would experience if they purchased and hosted their own physical servers. If instead of selling a subscription to individual guest virtual servers the infrastructure is sold on the basis of capacity to execute concurrent guests a business can then truly optimize its infrastructure usage to a minimum by providing each development team the ability to hydrate a specific number of virtual guests concurrently. This need then becomes independent of the number of applications actually being supported and instead becomes proportional to the concurrent activity of a development team. Not only does it provide the opportunity to decrease capex and opex, it also provides the opportunity for the rapid spin up of new development initiatives without the delays that can sometimes be imposed by the time taken to establish development server environments, providing a hard to quantify but very real time to market advantage for product development teams. The removal of development environments from expensive highly resilient data center space allows for production server capacity growth, another hard to quantify yet equally important revenue growth supporting opportunity.
My next post will deal with responding to some of the commonly cited objections and resistance to the idea of virtualizing development environments as well as moving infrastructure outside of the confines or security of an enterprise's own datacenters.

90% of businesses should not own a data center.

2010-04-23T08:46:00.000-07:00

Poorly leveraged infrastructure is the least financially efficient model of deployment. The days of needing dedicated hardware for individual applications has passed again as we swing back to shared horse power of virtualization and consolidation reminiscent of the mainframe era.
The logical extension of virtualizing application environments is the sharing of disparate workloads with variable utilization cycles which enable the aggregate CPU/IO/RAM resources to be lower than that of independent application specific hardware. But most businesses, especially small to midsized, tend to have small geographic and time zone coverage, this leads to a lack of diversity in workloads, this can lead to situations where the result of virtualizing environments while reducing the number of physical servers may still result in one resource not decreasing, be that IO, CPU cycles, RAM.
Clearly resource consolidation is just one advantage that operating within a virtual infrastructure environment provides, and it is possible that the business value created through greater automation, improved speed of deployment, rapid repurposing of infrastructure etc can make the transition worthwhile absent of the value of resource consolidation.
By combining individual company workloads together greater workload and resource utilization diversity can result which then leads to a greater opportunity to consolidate infrastructure due to the peak aggregate resource utilization at any moment in time being lower than the sum of the individual peak utilizations. If this math does not hold true for a set of workloads then there is no net improvement in resource utilization through consolidating onto a single infrastructure.
To take this thought process one step further, if there is a resulting advantage in aggregation of workloads then clearly co-location of infrastructure is not sufficient to reap the reward fully. It will be necessary to share infrastructure between companies, sharing storage, server, network etc. this notion of shared leverage infrastructure is marketed today as The Cloud. And I would argue is the appropriate place for any small to midsize (non-global) business to place its workloads.
Okay so the 90% number was pulled out of the air, but, any business which does not have a global geographically dispersed client base which results in almost continuous level utilization will benefit from sharing their infrastructure with other applications – which may be outside of the confines of their own organization.
The good of the many outweighs the good of the few.

Wind and Solar power

2010-04-21T08:12:00.001-07:00

With Earth Day looming ever present I was thinking about power generation in the US, we are it seams heading towards a resurgence of Nuclear power, yet we still lag way behind Europe for Wind and Solar generated power, the map below shows how suited the middle states are for wind generation, I know from experience that Minnesota is well suited and North Dakota ideal, yet few large scale wind farms exist.

I found an interesting article here that details all the available untapped powers sources in the US, I wonder what it will take for us to get with the program and follow Europe into renewable energy generation?

If you are interested to see the solar power growth in the US there is a cool animation of generation deployments of the past decades here.

Spring has sprung

2010-04-18T19:34:00.001-07:00

We planted this tree last year at the cabin, its buds are bursting forth color

Network Services Virtualization

2010-04-15T09:37:00.000-07:00

I had a very interesting conversation with the CEO of Embrane today, they have a clear vision for virtualizing the network services, this is a company to keep an eye on going forward.

"Embrane’s Architectural Vision
Embrane has designed a software-based solution that simplifies the adoption of virtualization and cloud computing for both service providers and enterprise customers, while preserving the existing operational model for the data center. This solution enables the delivery of any network service on a pool of general purpose, x86 processors in the data center. It leverages an architectural design familiar to both the network and systems administration teams, thus making its adoption possible without having to work across organizational boundaries.
It includes the ability to manage and automate the lifecycle of network services through a policy-based, distributed control plane that enables scalable provisioning, monitoring, configuration and reconfiguration of the services as well as elastic changes to their profiles.
Embrane’s solution enables virtualization of network services very much like hypervisor technologies have enabled virtualization of servers."

Setting up gmail with Outlook 2007

2010-03-21T08:10:00.000-07:00

I am a big GMail user, but I have been frustrated lately when using it as a pop3 server from Outlook.
I discovered a great web site which explains how to set up GMail as an IMAP service instead, now I have the ability to keep the GMail folders in sync as well as use staring/flags etc from both the web client, mobile devices and outlook - a much better solution.
I now have all of my email services in one outlook solution, life is good.

The DBA’s role in green IT?

2010-03-19T06:22:00.001-07:00

A number of months ago I was speaking at a conference and I talked about how every person involved in application development has the ability to affect the environmental impact of the application solution. To say this strained the credulity of some members of the audience is being nice about it, some just downright didn’t get it. Clearly I had not drawn the line of sight from their actions to the impact it could have.
The DBA has many roles during the lifecycle of development, establishing a development environment, designing schema, supporting the promotion of application and schema through the multiple environments prior to and into production, the monitoring and tuning of the applications performance, the establishment of production infrastructure, designing appropriate high availability architectures, putting in place backup and recovery mechanisms, establishing a disaster recovery or business continuity methodology. I would assert that every one of these steps provides the DBA with an opportunity to meet the business needs, and minimize the environmental impact of the total solution.
Development environments spawn like rabbits in spring, it used to be that you walked into a development shop and under every developers desk there where a plethora of humming PCs warming their knees. Today the situation is not much better, in fact in some cases is worse. How many application development initiatives have you been on where the first step is to acquire the development platform hardware and get it racked in the data center? And perhaps just as importantly how many of those servers are sitting idle now in the datacenter? Based upon my experience I would estimate between 10 and 20% of power consumed in our datacenters today is as a result of non production equipment.
Having developed applications for many years I can say with authority that developers are server huggers, and not only do we hug servers but we also like them fast. But come on, do we really need them? Virtualization whether server hosted or client hosted guest model provide a powerful flexible economic solution to the developer’s dream of dedicated environments, pristine and free from other developers prying fingers and code.
On the last agile project I worked on we used a single beefy 2 socket quad core machine, sliced into over a dozen VMs running build servers, source control servers, database platforms, development-qa-staging environments, multiple versions of QA environments, replicas of production, you name it we had it; quickly, cheaply and highly power efficient.
Databases are just like any other application, they have specific infrastructure needs, and they CAN be virtualized, yes they might need more memory allocated to them and yes they will probably run faster on their own dedicated native OS platform, but this is a development environment we are talking about, who cares if it takes 90 seconds or 95 seconds to build the application or run the unit tests? In the grand scheme of things we waste more time playing catch over the cube walls than we would gain by having higher performance development environments.
Databases do offer an alternative model of shared infrastructure consolidation; most modern database platforms can support multiple schemas simultaneously. So if your organization has sufficient activity going on you could consider establishing a single larger instance of a database platform and share it across not just a single development teams builds, but across many development teams. This requires good governance, and possible the ability to lock down access to the various schemas using team based ACLs. This approach is less flexible than using virtual platforms as it is harder to tear down and rebuild which can be a useful technique to prove our build and deployment scripts. It also creates a substantial single point of failure which you will then be tempted to make more resilient with clustering or other high availability techniques which then in turn lead to a less efficient environment as the redundancy is just that redundant (most of the time).
Database servers eat RAM and CPU cycles like there is no tomorrow, they also can consume boat loads of disk if you let them. Understanding the appropriate way to ensure the necessary performance characteristics are met with the least infrastructure impact requires the DBA to have intimate knowledge of the way the application intends to use the data. The answer to performance problems is not always partitioning and adding indexes. Consider helping the developers to optimize their data access, help them how to leverage the power of the structures you have created, how to create efficient joins, how to prevent massively inefficient table scans. By partnering with your development teams you can reduce the work load on the server and hopefully reduce the need for additional indexing strategies. Remember every additional index results in additional disk space being used, which results increased spinning media (unless of course you are running on an SSD based storage architecture), and of course those indexes will get backed up which will use yet more media, and your high availability infrastructure will need to be that much bigger etc. etc. hopefully you get my point: Efficient code written in collaboration with DBAs will ensure the most efficient data access which will result in the least amount of infrastructure necessary to support the databases.
One final though before I run out of time: HA and DR/BCP. Datacenters provide resilience in many fashions at many layers of the infrastructure and application architecture, how much do you need? If you have geographic redundancy with a certain recovery time objective, what level of high availability do you need in each datacenter? If you have remotely replicated storage what backup mechanism is necessary? A lot of questions, all of which can have huge consequences on the power consumption of an application solution. More to come on these thoughts.

Choosing a hardware platform for virtualization

2010-03-18T18:59:00.001-07:00

Many server vendors will tell you they have an optimized server platform specifically designed for virtualization, but really, do they? Virtual Machine hosting is just like any other application hosting, the performance characteristics and infrastructure demands will depend heavily upon the type of work load executing on the servers. Now that said many server platforms do have characteristics that lend themselves to perform well under virtual workloads, large memory footprints, fast memory paging/moving architectures large IO capabilities, many core CPU architectures, high levels of component redundancy, electrical power efficiency, cost. These and many more capabilities and features certainly can incent one to choose a specific vendors platform.
In order to make an educated decision you need to understand the burden your typical workload profile is going to place on the infrastructure to understand the efficiency of each platform configuration. What ratio of RAM to Processor Cores do you need? What Storage connectivity is to be used, NAS, SAN, FCoE, iSCSI? What network throughput will you need? How many VMs per physical host are you prepared to allow before the failure domain becomes intolerably large?
Consider also what your priority is. What is driving you to virtualize? Is it space constraints, power running out, opex, capex, the need for a more resilient infrastructure. Most likely it is some combination of all of these priorities. Knowing these drivers will help you decide what your decision criteria will focus around.
Knowing parameters like these allows you to make a real judgment of each vendor’s platform to understand the tradeoffs between chassis types, network port configurations, RAM and CPU selections. And don’t be afraid to get them in a lab to validate your assumptions, it does not take long to install a hypervisor and tryout the performance of your typical workloads on the real configurations, any vendor who wants your business will provide you the equipment for such a study.
Create a matrix of the configuration parameters mapped against the priorities, empirical data can help you then find the sweet spot of vendor platform configuration for your business needs.
Remember, despite what the salesperson may tell you, your virtualized environment infrastructure needs are unique and without a thorough understanding and analysis no one knows the most appropriate server configuration that will deliver the value you need. Do your homework, and repeat it often, as vendors platforms seem to change like the seasons.

Going a bit green in Helsinki

2010-03-18T13:36:00.000-07:00

Speaking of data centers I read in Fast Company magazine today about a data center in Helsinki that is using the heat generated by the infrastructure to heat the cities buildings, it makes interesting reading, better to do something productive with the heat rather than just vent it into the neighborhood sky.
Interestingly enough the data center sits beneath the bowels of a cathedral - blessed be the servers for they shall warm the city?
Apologies to Monty Python fans for the abuse of Dr Karl Grubbers institute's name in this posts title.

Green Data centers?

2010-03-18T13:18:00.000-07:00

It is estimated that 1% of all electrical power in the world is currently supplying datacenters. The Energy Information Administration estimates that in 2010 the worldwide electric power generation equals 20 Trillion Kilowatt hours. Meaning that approximately 200 Billion Kilowatt hours of electricity will be consumed global by the power hungry data centers humming away feeding our ravenous appetites for data and applications. By my estimations this approaches 150 Million Tons of carbon dioxide production annually, this is approximately 800 billion cubic meters of CO2, which if distributed across to United States evenly would represent a blanket of CO2 an inch thick. By anyone’s estimates that is a lot of CO2.
PUE, Power Usage Effectiveness, is a measure that has been widely used to measure data center efficiency, it is calculated as the ratio of total amount of power used by a computer datacenter facility to the power delivered to the computing equipment. The closer the number approached to 1 the more efficient the facility is. Whilst being an interesting benchmark to compare the efficiency of cooling, air movement and power distribution technology it makes no attempt to measure the efficiency of the compute, how many clock cycles within the data center are idle NOPs, how much the memory capacity is utilized, how frequently and how much of the spinning disk is being accessed.
A great deal of focus has been paid on increasing efficiency of data center facilities, and these have reaped tremendous rewards in the ability to pack datacenters with even greater numbers and density of servers and infrastructure. Many organizations are not however paying attention to the efficient usage of the deployed technology. Virtualization can play a key role in driving up utilization, distributing workloads across pooled resources. But many organizations infrastructure has a cyclic utilization dependent upon the usage and access patterns of the end user community. These types of environments do not lend themselves to great efficiency gains through virtualization. Simple server consolidation down to the minimum number of servers necessary to run the peak workload will provide the most efficient single usage environment.
Data centers will only become more efficient if we can drive up the overall utilization of the entire supported infrastructure. Shared infrastructure supporting disparate workloads is one way to enable a greater average workload across the datacenter. And this is where the great C word enters the picture: Cloud infrastructure supporting many applications with differing user access and usage patterns offers the greatest possibilities for increasing datacenter efficiencies.
So the greenest way to use a datacenter appears to be to not own one exclusively but to share one with many other companies and applications. Virtualization provides a great facilitating technology to allow the migration from exclusive single company datacenters to the infrastructure clouds, ensuring isolation, security and resiliency in an economic manner as well as possibly the least environmentally impactful way.
It is commonly stated that average CPU utilization in non optimized datacenters is around 20%, imagine if through consolidation onto cloud infrastructure the average utilization was driven to 80%, resulting in massive electrical savings, massive CO2 production savings, and maybe the CO2 blankets thickens would only grow at a quarter of one inch per year then.
I think I will go and buy some stock in the cloud infrastructure space, and then plant some trees to hug.

Client side virtualization

2010-03-17T13:36:00.000-07:00

In February 2010 VMware announced CVP, previously in October 2009 Citrix announced their XenDesktop 4 platform. Both of these announcements show a growing recognition that centralizing the desktop compute horsepower into the datacenter may not be the most attractive solution to solve the real challenge which is how to wrestle back control of the disparate desktop infrastructure which exists in most corporations today.
Client side virtualization involves running a virtual machine on the local terminal (desktop or laptop), the VM is streamed from the centralized management platform on demand to the end user terminal and cached. Changes are synchronized back to the centralized management platform, and these changes can be cached, enabling disconnected execution – finally solving the disconnected worker virtualization challenge. The centralized management platforms provide many of the parent child image based management efficiencies that the server side VDI implementations provide, however they require far less compute in the datacenter, and similarly require far less bandwidth between the terminal and server once the image has been streamed.
Compared to the Server side guest VM model the end user terminals require more compute power. Consider the mobile workers; the average power draw from a laptop can be approximated at 75W assuming no additional external monitor is provided. This compares favorably with the thin client hardware discussed in my earlier post and the server burden of client side virtualization can approximated at less than one tenth of that required to support a guest VM, reducing the server burden to as little as a single Watt. On the face of it Client side virtualization using laptops appears a very power efficient approach.
Many factors need to be considered in making decisions on changing from a standard desktop to a mobile platform, acquisition costs are higher, support costs can be higher due to equipment failure, anticipated life time will be lower than traditional desktops. So I am not proposing that a wholesale adoption of laptops is the answer. However the adoption of thin clients with server side virtualization for desk bound workers and laptops using client side virtualization for mobile workers becomes a highly power efficient, capex reducing and capital efficient approach to implement a virtual desktop strategy.
But a final word of caution, both products I mentioned in the opening paragraph are very much in their infancy, and not ready for production at the time of writing (3/17/2010). So what is my recommendation? Proceed with thin client server side virtualization programs for your desk bound workers and wait before making any decision on the mobile workers platform. Most major vendors in the virtualization space are aggressively building the client side solutions, and rest assured 2010 – 2011 will be a boom time for mobile worker virtualization.

Is desktop virtualization green?

2010-03-16T20:09:00.000-07:00

Over the past several months I have spent a great deal of time learning about desktop virtualization, and have come to the conclusion that the current server based guest model of virtualization is decidedly un-green if you do it the wrong way.
In a white paper published by Wyse they quote a the power consumption of an average thin client terminal including monitor to be 100W when running connected to a Terminal Server, and compared this to an average PC consuming about 170W under a similar work load. A typical Virtual Server Host, an IBM 3850M2 with two processors, draws around between 650 and 850W depending upon workload, RAM and Processor configuration.
Assuming a conservation 50 guest VMs running each would consume approximately 15W according to an IBM whitepaper. The combination of thin client terminal plus the server burden for the virtual machine nets out to 115W which when compared to the original 170W PC is a great improvement. Few organizations can afford to refresh their entire desktop infrastructure in one effort and so many desktop PCs get repurposed as terminals accessing server based virtual machines.
The graph shows the effect on power consumption of a three year migration of 1000 desktop PC to thin client. These calculations show a saving of approximately 70kW or 38%. Note though that the server burden goes up from zero in the traditional desktop environment to 15kW in the virtual desktop infrastructure environment.
So if you are considering desktop virtualization as a green activity it needs to be combined with refreshing legacy desktop equipment to realize the opportunity to decrease the environmental impact, otherwise you will be taking a step back as the server load increases your overall power consumption.
Few companies are able to fully virtualize their desktop environments due to the need to support disconnected/mobile workers. But several companies have recently announced disconnected virtualization models that leverage the local CPU to run the guest virtual machines. In my next post I will examine how this will be a game changer both from a management perspective as well as open the door to alternative computing models with potentially lower environmental impact.

Time to do some (re)learning

2010-03-11T06:20:00.000-08:00

I am considering developing some deep experience in the following areas to keep my mind working:

iPhone application development
Windows Mobile 7 development
Ruby on Rails

Which should I do first?

Web search and kids

2010-03-05T15:33:00.000-08:00

E just learned that Googling bad words shows bad pictures... any recomendations on helping to protect her on the internet? Net Nanny seems like a highly rated product.

Love her to bits

2010-02-28T16:25:00.000-08:00

The Muppets: Bohemian Rhapsody

2010-02-24T12:12:00.000-08:00

Wont be long

2010-02-23T20:45:00.001-08:00

The Ice will melt soon:

MN Zoo

2010-02-15T15:14:00.001-08:00

More white stuff

2010-02-15T08:19:00.000-08:00

Last night another inch and a half of snow fell in Savage, it cleaned up the crusty looking stuff so it looked pretty. Unfortunately the snow plow came through again, and again and built up a pretty pile at the end of the drive. It seems to me to take as long to clear the snow plowed pile as it does to do the rest of the drive and side walk.

Yesterday evening as a celebration dinner we cooked roast chicken and roasted vegetables (parsnips, carrots, potatoes, corn). We finished the meal with strawberries dipped in chocolate. To drink we picked up a cheap bottle of sparkling wine a Moscato Spumante from Bottega. Despite my scepticism it was very tasty.

Valentines Gift

2010-02-14T11:23:00.000-08:00

J gave me a Nespresso coffee maker today.

so far I made two very nice espressos, unfortunately I had already had two 16oz cups of coffee at breakfast so I am now a little jittery...

The coffee was very smooth thick and flavorful, just like the best coffee I had last month in Sydney. We are now suckered into buying the small capsules of coffee but at least it is now easy to regulate how much coffee I drink, and in a matter of a minute I can have a wonderful espresso to wake me up on the way to work.

Thanks J - love you tons.

Decision time

2010-02-12T08:51:00.000-08:00

For several months I have been contemplating the socio-political climate of blogging about work, technology, family and kayaking. I have reached the conclusion that I am going to separate them and that my original blog will become a kayaking only blog. I have established this blog as a duplicate of the previous posts (I cleaned out the past 6 months kayak posts). I am going through a process of re-branding my first blog to Paddling not politics (you will find that redirection from http://paddlingnotpolitics.com/ works), I am working with a friend Ron (master paddle maker extraordinaire) on some concepts around traditional paddling gatherings and I hope to share them on that blog soon. I am also still considering a technology only bog....
If you experience some weird redirects and cant get to the blogs give them a day to propagate the changes through the domain name servers please.
If you would like to comment on this decision please send me an email and let me know your thoughts.

East coast Blizzard

2010-02-08T11:55:00.000-08:00

Emma had no school today as Baltimore County is digging out from about two feet of snow that fell over the weekend. Apparently more is on the way and they are thinking as many as three snow closure days...... Today was very cool for Emma and I as she finally got Skype video conferencing working and we were able to chat numerous time through out the day.

Homeward bound

2010-01-22T11:47:00.000-08:00

Plaza de Mayo, originally uploaded by Christopher Crowhurst.

Argentina proved to be a beautiful country filled with fabulous architecture and warm hospitable people. The mix of ancient and modern architecture makes for a vibrant lively city, add to that the street café scene and you have a great place to live, or visit.
That said its time to fly home, I can’t wait to see my family again. Europe beckons next with a trip to the UK, Spain and Belgium.