I will be presenting the following sessions at JavaOne 2013 India:

Session ID: HOL1186
Session Title: Hands-On Introduction to Integration and Oracle SOA Suite 11g
Venue / Room: Flexi 1 HOL Room – Develop
Date and Time: 5/8/13, 14:00 – 16:00

Session ID: CON1177
Session Title: Oracle SOA Suite: the Most Capable Tool for Every Possible Integration Challenge
Venue / Room: Hall 2
Date and Time: 5/9/13, 10:15 – 11:15

Hope to see you there!

Visit the Oracle booth #608 at Cloud Connect next week (April 3 and April 4) to learn about how you can use Oracle SOA Suite to provide a comprehensive cloud integration solution. I will be at the booth on April 3rd starting at noon. Looking forward to see you there!

Update Nov-2012: The session pages now have the slide pdf documents for download. Simply click on each session link below to get started.

—-

I have 4 sessions at Oracle OpenWorld 2012.

Oracle Fusion Applications: Customizing and Extending Business Processes

I will join my colleague Ravi Sankaran (outstanding architect and advisor to the Fusion applications development team) to talk about the approaches in customizing and extending Oracle Fusion Applications with Oracle SOA Suite. CON8719

When: Monday, Oct 1, 4:45 PM – 5:45 PM

Where: Palace Hotel – Twin Peaks North

Managed File Transfer Use Cases and Oracle Fusion Applications Cloud

As part of the “Birds of a Feather” session, I will join Dave Berry and Sundar Shenbagam to discuss usage scenarios for integrating managed file transfer with Oracle SOA Suite, Oracle B2B applications, Oracle Healthcare applications, and Oracle Fusion Applications. BOF8326

When: Monday, Oct 1, 6:15 PM – 7:00 PM

Where: Marriott Marquis – Golden Gate C3

How to Integrate Cloud Applications with Oracle SOA Suite

I will be joined by Geeta Pyne (Director, Middleware at BMC Software) to address cloud integration challenges and how Oracle SOA Suite can help with a consistent approach to integration, whether on-premises or cloud. I am quite excited about this session as we will tackle the hype and myth of “simple” cloud integrations and share real-life application integration experiences. Don’t miss this one! CON8968

When: Tuesday, Oct 2, 11:45 AM – 12:45 PM

Where: Moscone West – 3003

Oracle Fusion Applications: Best Practices in Integration Design Patterns

I will be joined by Ravi Sankaran to provide a high level view of the Oracle Fusion Applications integration strategy and showing the best practice integration design patterns. You will learn how to discover integration assets, invoke web services and use cloud data integration. The session is not just limited to SaaS deployments, but will be useful for on-premises customers as well. CON8685

When: Tuesday, Oct 2, 1:15 PM – 2:15 PM

Where: Palace Hotel – Telegraph

I will also be at the SOA Customer Advisory Board on Thursday, in case you are attending. See you at OpenWorld!

A few months back, a key telecom customer needed to describe the semantic Enterprise Business Object (EBO) structure in a hierarchical/indented model to a flat spreadsheet so that designers can perform application mapping for the Process Integration Packs (a.k.a. PIP, basically pre-packaged integration accelerators for specific business process flows). I pointed them to a not-well publicised (and highly underestimated tool) called the XSL Mapping Analyzer (a.k.a XMAN).

Below is an email from the customer describing how useful they found the tool, in fact just one of the utilities! Thanks to the OCS Team (Balaji S G and Kurt) and KT (Sungwoo Yu) for this valuable feedback!

Side note: XMAN is actually a set of utilities my team built to parse XSL and EBO XSDs into various formats for semantic analysis of attribute level mapping; including HTML, CSV, flattened CSV and XML. It’s real power is in providing tactical SOA governance if you are using AIA canonicals, as it can help flag functional interoperability issues across projects using the same EBO, thereby encouraging service reuse and providing the ROI on your SOA. To learn more about XMAN usage, see my earlier blog post and OOW 2009 session.

— Email below published verbatim with prior owner’s permission —

Hi Balaji,

I’m glad to share my experience with Rajesh’ blog

Specially, I use to visit and learn a best practices using aia from that popular blog.

I’d like to briefly introduce my situation.

I’m working at telecom Project aims to build next generation BSS/OSS for 35 mil subscribers utilized by Oracle RODOD(Siebel,BRM,OSM, OPH and etc) .

I’m in charge of Integration team.

As we try to fit OOTB PIPs with our business, we need to analysis OOTB PIPs and design extension of PIPs.

These days our team cooperation with OCS for Communication PIP extension design on AIA Technical Document Design.

Although we cannot totally understand what existing input/output message and mapping in complex process integrations, we have to provide explain how prebuilt aia integration build interoperability with Siebel, BRM.

Basically, we can refer PIP implementation documents. However, if we closely looked at message level input/output and mapping details for reviewing service detail communication with edge application. It had been challenge.

At that time, I concerned that below so that I requested the OCS guidance

For we describe edge ABM structure or complex EBMs on spreadsheet, how can we describe different level elements on spreadsheet automatically?

Why we need to describe on spreadsheet, most of edge application function designer is not familiar with complex related xsd and different level of structure specially ebo.

So, when designed for extension of pips, Edge application function designer is unable to consider EBO structure.

If AIA team share used each operation EBM by spreadsheet, they will be easy to approach understanding EBO

Do you have any idea for that?

1.       How to automatically convert message structure xsd to table. (It could be different structure distinguished by indentation)

2.       If we describe different level elements distinguish using XPath, How we automatically extract XPath of all nodes from EBM.

Thanks to OCS(Balaji, Kurt) helps

We can utilize AIA Development Tools such as XMAN, XSD Flatter and generated XMAN reports RV11.1.

Particularly, XSD Flatter is right to solve above two problems.

For instance, I’m going to explain how I can extract element tree on spreadsheet from SalesOrderEBM.xsd

source $AIA_INSTANCE/bin/aiaenv.sh.

cd $AIA_Installed Directory/DeveloperTools/XSDFlattener/bin/

xsd2csv.sh -inputSchemaURL $AIA_Metadata_Directory /AIAComponents/EnterpriseObjectLibrary/Industry/Communications/EBO/SalesOrder/V2/SalesOrderEBO.xsd -rootElement ProcessSalesOrderEBM

As converted, I’ve got a csv file with x-path, cardinality, annotation and element tree with display hierarchy using indentation, and then reformatted below.

Even, I can convert ABM of Siebel Integration Object Message as well.

xsd2csv.sh -inputSchemaURL $AIA_Metadata_Directory /AIAComponents/ApplicationObjectLibrary/Siebel/V1/schemas/SWIOrderIO.xsd -rootElement ListOfSWIOrderIO

In addition, I aware generated RV11.1 XMAN report already provided whole EBM csv information by support.oracle.com

Also I can generate csv report from all of EBM using eol2csv.

eol2csv.sh -inputDir $AIA_Metadata_Directory /AIAComponents/ApplicationObjectLibrary/Siebel/V1/schemas -outputDir ~/eoloutput/AIAComponents/ApplicationObjectLibrary/Siebel/V1/schemas

Thanks

Sungwoo

As described in the paper, these services are used by customers/partners to extend and integrate with Oracle Fusion Applications, whether in on-premise or SaaS mode. This post describes the keyword in detail by addressing five frequently asked questions.

Q1. What does the EXTERNAL keyword imply for integration scenarios and SaaS deployments?

To understand this, let’s look at the Fusion Applications deployment topology below Note: this shows only SOA Composite services, but is also applicable for ADF Services.

The topology splits visibility to web and service resources into two categories – internal and external. This is achieved by fronting all HTTP requests through two different Oracle HTTP Servers (OHS), one for internal traffic routed between Fusion Applications servers, and another for external sources, which has only a small subset of the routing rules to prevent unauthorized access by external systems. This approach facilitates unhindered access within the Fusion Applications domain and acts as a firewall to restrict access to external systems.

In the example shown above, internal clients (i.e. those deployed in the same Fusion Applications domain) would have access to all composite services - A, B, and C and any service endpoints that they provide. In contrast, all other external clients are restricted to accessing only services provided through composites A and B.

Another characteristic of services that are exposed externally is that they also enforce a more stringent Oracle Web Services Manager (OWSM) policy using SSL or WS-Security message protection. By contrast, services NOT marked as EXTERNAL are protected using the OWSM Global Policy Attachments (GPA) feature, which in the case of Fusion Apps do not enforce message protection and rely only on username token or SAML identity propagation.

As you may have guessed by now, services tagged with the EXTERNAL keyword in OER, are the only ones provisioned with routing rules on the External OHS and protected with a stringent message protected OWSM policy. This is what makes them suitable for application integration use cases; even more so in the cloud where deploying custom services to the SaaS WebLogic Server domain may not be possible.

Q2. The service I want is not marked as EXTERNAL. Can I just tag it with the keyword for my integration project?

As explained above, making a service “EXTERNAL” is more than just tagging the keyword in OER. The service also needs to be secured using a message protection enabled local OWSM policy and requires modification of the OHS routing rules to allow service URI access to the external world. For Oracle-shipped services, this is done by the Provisioning framework. For custom services, this can be possible in on-premise and hosted installations, but may not be possible in the SaaS mode.

Most likely, if an Oracle-shipped service has not been marked as EXTERNAL, it is due to specific functional and/or security reasons, which is usually not addressable at the customer site.

Q3. If I can’t change non-EXTERNAL services, why document them in OER?

While it is true that non-EXTERNAL services cannot be consumed by external clients, they can still be consumed by internal clients, such as custom implemented composites that are deployed in the Fusion Applications SOA domain in on-premise or hosted deployments. Moreover, apart from service invocations, many composite services also have capabilities for layered customizations for BPEL processes and Oracle Business Rules. The OER entries for these non-EXTERNAL services therefore serve as API documentation.

Q4. Are EXTERNAL services the same as “Public” services?

The two terms — “EXTERNAL” and “Public” — are frequently used interchangeably, however, they are not synonymous! “EXTERNAL” refers to service visibility in the topology, while “Public” basically amounts to the level of API support provided by Oracle. In fact, there is no attribute called “Public”, instead it is referenced by the Compatibility attribute value of Supported.

While most EXTERNAL services tend to be Public, there are exceptions. For instance, mobile-enabled services accessed by iPad or iPhone apps (external to Fusion Applications) will be tagged as EXTERNAL. However, if the only intended supported client is the pre-certified mobile app, then the service may be marked with a Compatibility value of Not Supported to effectively make it “Private”.

The Fusion Applications OER white paper also lists the various combinations of the Keyword and Compatibility tags and their implications for service use in integration projects.

Q5. How do I find all available EXTERNAL services?

Using the cloud hosted OER instance. Simply search for EXTERNAL and limit the Asset Type to ADF Service or Composite Service to filter the results.

IMPORTANT NOTE: This is a simple text search and will result in services that have the word EXTERNAL in any metadata, including descriptions. Therefore, you may get false positives. To confirm the service accessibility, always check the Keyword value on the Taxonomy tab of the service detail page as shown in the screenshot at the top of this post.

Hopefully this post has addressed most of the questions about the EXTERNAL keyword usage. If you still have some, feel free to sound off in the comments.

One of the common queries I get is how on-premise customers without a dedicated OER can find a concrete service WSDL URL for their specific environment using the cloud hosted OER instance.

To answer this, let’s understand the two attributes on the OER service details screen.

In the Cloud OER instance, the Abstract WSDL URL link points to the Oracle-shipped non-customized service definition, which can be used by partners/ISVs for developing tenant-agnostic apps (more on that in future posts).

The concrete WSDL URL can be found in the Service WSDL URL link (highlighted in the screenshot above). In an on-premise deployment, this link points to the runtime WebLogic Server where the service has been deployed. Since the cloud OER instance is not linked with on-premise customer-specific runtime environments, the link does not work (as expected). However, you can still derive the on-premise environment-specific concrete Service WSDL URL via a simple substitution.

Replace

rep://FUSIONAPPS_HCM/

with

https://host:port/ for the specific Fusion Applications environment.

The Fusion Apps repository team is working on making usability improvements to document this in-place, along with adding additional service metadata that you are sure to find very handy when consuming the services. Stay tuned!

AIA Essentials Book

I was recently asked by Packt Publishing to review the above mentioned book. I have worked in AIA development for three years (and still work closely with the team) driving many cross-functional Process Integration Packs (PIPs) architecture initiatives, developing Foundation Pack developer productivity tools, managing SAP PIP development and leading pre- and post sales solution architecture support for strategic accounts to enable successful implementations. Having seen issues faced with complex SOA based integration projects first-hand, it was an interesting opportunity to review a book that attempts to explain the “essentials” of my product from an implementors perspective.

Swiss Army Knife. Image courtesy of Josh Linkner.

In enterprise business applications, complexity stems from complex enterprise needs. To give you a flavor of these, here are just a few examples of requirements that lead to creation of complex platform infrastructures that make up the complex enterprise software.

User Interface

• Translation for multiple languages
• Localization for different areas (e.g. currency, dates etc)
• Complex Query Needs (custom fields, saved queries etc)

Customizability/Extensibility

• Adding custom attributes, objects
• Modifying out of the box functionality e.g. business processes
• Modifying the look and feel for personalization by end users (e.g. MyYahoo)
• Verticalization needs (e.g. industry specific flavors)

Operational

• High Availability (for planned downtimes)
• Selective feature uptake (to avoid re-training thousands of users)
• Performance requirements for high volume, latency, throughput etc.

Functional

• Complex security needs (users, roles etc)
• Organizational setup (e.g. business units, divisions etc)
• Error handling and compensation for rolled back business processes

In addition, there are a few additional factors to consider for large enterprise software products:

Development Scalability
For multi-product application suites, economies of scale is achieved by consolidating common functions into a common platform, which then have the same complexities that the applications face when trying to develop products for a wide range of customers. This adds a layer of “knobs” to tune over and above the functional knobs that the application software provides.

Integration
No enterprise software lives in an island and integration with other systems is usually one of the big costs in application deployments. Integrations are inherently complex due to the nature of trying to tie together heterogenous applications with different data models, granularity, cardinality, semantics and protocols.

Standards
To allow for a plug-n-play model, most platform level APIs are driven with facade style interfaces that can plug-in to various implementations e.g. jazn; that add a layer of complexity. Use of standards-based technologies (BPEL, ESB, Web Services) also contributes to some level of complexity, as those standards are aimed towards satisfying requirements from all the participating members who created those standards.

Perception
I do not pretend that all enterprise software are absolutely user-friendly, but do realize that large chunks of enterprise software are targeted towards specific roles, which may seem very unfriendly to users who are not in that role. For example, a snazzy, graphical UI is not necessarily “simple” for data-entry clerk as it is to a knowledge worker.

When you think about it, a large percentage of the world economy depends on enterprise business software (ERP, CRM, HCM etc), and requirements such as above tend to add complexity.

Note that while the software can be complex, there is no excuse for not making the end-user experience as smooth as it can be. Although in my experience, some complexity does tends to bleed into the user experience, especially in on-premise deployments. The SaaS model, to some extent, shields end-users from most of the complexity, but once you start getting into requirements such as integration, complexity does find its way through during implementations.

Quora: Why is enterprise software often so complicated?

Update 5-June-2012: Great answer by Jignesh Shah and my response.

Oracle Fusion Applications are designed and built from the ground up using a service-oriented architecture (SOA). As I have discussed in an earlier post, visibility and control are the key tenets to SOA governance, and the first step in integrating with Oracle Fusion Applications is to find out what are the integration options available. Oracle Enterprise Repository, an industry-leading metadata repository, provides this visibility.

Oracle Enterprise Repository for Oracle Fusion Applications can be used to discover functional and technical information about integrating with other applications, including web services, business events, interface tables and data model documentation among other artifacts.

Oracle Enterprise Repository for Oracle Fusion Applications

To maximize the usage of the repository, I have posted a white paper “Making the Most of the Oracle Enterprise Repository for Oracle Fusion Applications” - also available from the Help link on the top right of the repository. I thank Nigel King, Tim Hall and Kris Doherty for their valuable inputs and reviews on the paper. I hope that our customers and partners will find this a useful resource to mine the repository content. In future posts, I will discuss various best practice integration patterns i.e. how to integrate with Oracle Fusion Applications.

Note: Access to the repository and the white paper requires an Oracle account.

The task of setting up my new iPad and iPhone recently was an eye-opener to the dependency I now have on cloud services. I expected it to take a few days to totally get all my stuff (email, contacts, calendar, photos etc) on the devices, however, it was a pleasant surprise to get it done in less than a few hours.

No, the credit does not go to the devices…it goes to the cloud! This is what my current setup looks like for the various personalized services and how I keep the iDevices and other laptops in sync.

Email Accounts: Gmail …setup using Google Sync on mobile devices. Work email is setup using IMAP which is auto configured via a custom profile. Thunderbird IMAP setup on the laptop.

Calendars: Google Calendar …setup using Google Sync on mobile devices. Work calendar is set up using CalDAV which is auto configured via a custom profile. Thunderbird Lightning CalDAV setup on the laptop.

Contacts: Google Contacts …setup using Google Sync on mobile devices. Work contacts directory are available via a special VPN based app. Thunderbird SyncML extension Zindus setup on the laptop.

Passwords and Personal Data: KeePass …the encrypted password database is shared via DropBox and accessed via HTTP using the MyKeePass app on the iDevices.

Chat: Skype, Yahoo! Messenger…iMO app connects to all the common messaging platforms. Work IM is configured using the Oracle Beehive app provisioned via a custom profile. Pidgin IM setup on the laptop.

My Documents: Dropbox …although the native app is quite good, what’s even better is the GoodReader app for the iPhone/iPad that can sync with Dropbox as well as email attachments, Google Docs and WebDAV servers. Secured TrueCrypt containers also synchronize well with Dropbox, however I haven’t found any mobile app that can securely browse and access those documents. Password-protected zip files (with AES encryption) have file names exposed and I have still not found any iPad app to open encrypted 7z files. Please comment if you have found any way around this. Update 07/29: iUnarchive app on the iPhone/iPad does open password protected zip files.

Not My Documents: Google Docs …this is stuff I want to read/keep but not use up my Dropbox storage and I could afford to lose e.g. pdf articles, free ebooks etc. Again, the GoodReader app does a great job browsing and downloading content from Google Docs. What could be made better is a sync feature as well as ability to view documents without downloading. Update 07/29: Using iUnarchive, I also sync the eBooks on the server with the built-in iBooks app.

Browser Bookmarks: Firefox Sync …using Firefox Home app on the mobile devices. Tighter integration with Safari or a native FF browser would have been ideal.

Pictures:  Picasaweb …browse and upload using Piconhand app or show slideshow using the iShowPhoto HD app (both free). Picasa setup on the laptop.

Videos:  YouTube and Picasaweb …native YouTube app. Upload using Picasa on the laptop.

Music: Pandora, ShoutCast and various other streaming sites and apps. For purchased/ripped content, sync via iTunes library (eagerly awaiting wireless sync in iOS5).

TV/Movies:  Netflix, Hulu, YouTube (yes, I am a CordCutter) along with iTunes Home Sharing, along with YuppTV for international news and music.

Telephony:  Google Voice, Skype …enough has been said of them.

News: Flipboard, Feedly …the apps use Google Reader RSS subscriptions as well as Twitter lists.

Thanks to the cloud, personalizing and setting up any new connected device or laptop now only requires a few settings and app installation. Now that’s what I call magical.

1. Get management buy-in.
2. Plan for entire releases, not just one sprint.
3. Plan sprints with specialized/shared resources in mind.
4. Complex inter-dependencies are a reality – deal with it!
5. Keep reasonable sprint lengths – at 3 or 4 weeks.
6. Don’t expect all sprint deliverables to be production ready.
7. Define “done” consistently across the teams.
8. Reserve at least two “hardening” sprints.
9. Be pragmatic about usable design documentation.
10. Adopt continuous integration principles.

Agile Development Methodology

Read the entire article here.

UPDATE: To clarify, the tips above specifically address the challenges put forth in the earlier post around enterprise application integration projects. For other enterprise projects, vanilla Scrum approach may work fine.

Image via Flickr

Enterprise integrations are complex, both functionally, due to implementation of a business process; and technically, due to introduction of one or more runtime layers between applications. Since these integrations typically represent end-to-end business flows, developers need to ensure that the performance meets the business need.

Here are some considerations when planning for performance testing of service oriented architecture (SOA) projects that integrate enterprise applications, such as Oracle’s Application Integration Architecture (AIA).

Update April 21, 2011: AIA specific tuning details can be found in Chapter 28 of the Developer’s Guide for AIA 11gR1 (E17364-02).

1. Define the End Goal. Clearly.

It may sound obvious, but it is the main cause of performance testing efforts going awry – lack of a clear end goal.

Note: “make it run faster” does not count as a clear goal!

Quantify desired metrics in an objective manner by setting Key Performance Indicators (KPI). Here are some KPIs you may want to check for:

• Throughput of the end-to-end business flow by users, payload size, volume
• Response Time for the end-to-end business flow by users, payload size, volume
• Throughput of integration layer only (legacy application interactions stubbed out)
• Response of integration layer only (legacy application interactions stubbed out)

2. Use Metrics Relevant to the Business

System performance KPI should be derived from business metrics so that it involves both business and IT. This results in a more realistic goal than arbitrary benchmarks set by developers or vendors. For example, the throughput KPI could be derived based on a formula that uses software cost and peak order volume to result in a “minimum orders per CPU core per minute” indicator that satisfies the business needs.

When looking at transactions, always consider “peak” spikes vs the average. For example, orders coming in usually have peak periods (e.g. holiday season sales), wherein the system will be subject to transaction load that is a magnitude higher than on non-peak times. Defining KPIs based on peak transaction volumes will not only help in setting realistic goals, but ensures true success of the project when it actually handles the load when it is most needed by the business.

Finally, don’t try to boil the ocean – identify a subset of the integration use cases which are prone to performance bottlenecks and meet all the KPIs before attempting other ones.

3. Do you REALLY Need Production Grade Hardware for Testing?

Using dedicated hardware is always better than sharing existing development or QA environments. However, every business has different needs with their enterprise applications and even this changes by business process. For example, an order-to-cash process may have a need for consistently high target performance metrics with medium-high load; as compared to the financial close process, which may need it once every quarter with high load.

Instead of buying or configuring hardware that necessarily matches every possible target scenario, consider the use of commodity hardware with matching “normalized” KPIs that are downsized from the target business scenario. For example, say the production hardware uses a given compute unit (CPU/memory/cache specification); and the commodity hardware is determined to be one-fourth the compute unit. If the business KPI target is 40 orders/CPU core/minute on the production grade hardware, then the internal, normalized KPI would be one-fourth of that i.e. performance testing would need to achieve 10 orders/CPU core/minute on the commodity hardware to be considered successful.

Of course, the benchmark may not scale as linearly, but this can be easily factored into the equation, providing a good educated estimate of the integration performance. Compared to the alternative of not testing due to hardware unavailability and discovering issues in production, use of commodity hardware and normalized KPIs can be a very viable performance testing approach.

4. Choose a Consistent Testing Strategy

For integration scenarios, a bottoms up testing strategy may be useful to consider, i.e. optimize a single use case fully (to reach desired KPIs) before introducing additional artifacts or flows.

Plan on the sequencing of the use cases appropriately, which can save some cycles e.g. between a Query and an Insert use case, the Query may look simpler, but it needs data which can anyway be seeded by the Insert use case, so it may make sense to proceed with Insert first. Also, identify the “data profiles” for the use cases and create representative sample data e.g. B2B orders may have 50-100 lines per order whereas B2C orders may only have 4-5 lines/order.

For each use case, once KPIs are met with for a particular number of users, payload size etc., run longevity tests for at least 24 hours to ensure that the flow does not have memory leaks or other issues. Check the desired metrics e.g. JVM garbage collection, database AWR reports etc. and purge data after each run to ensure consistency between tests.

When the above passes, gradually increase number of users and increase payload on the same use case to identify system limitations when under load. Once the specific use case is optimized to KPI for concurrent users / payload, add new flows to the mix and tune.

While the above may again seem obvious, the temptation to “switch gears” when one use case is not fully working can cause a lot of overhead in switching context by the project teams and setting up data for the new use case etc. It is better to complete one full use case successfully before targeting others.

5. What about Standalone Testing for Integrations?

Standalone testing – stubbing out enterprise applications – is useful strategy to identify integration hotspots and remove the unknowns of the enterprise application performance from the integration scenario. However, be aware that it will not identify all performance issues. Developing stubs requires substantial investment to emulate the edge applications and may be non-trivial for enterprise applications that typically have complex setups. Furthermore, some integration settings on the SOA server will typically change when the applications are introduced, so avoid over-tuning the solution when performing standalone integration testing.

Performance testing and tuning is still somewhat of an art that requires a good understanding of the technologies, its limitations, and all the available tuning “knobs” in each technology to achieve the KPI requirements of the integration flow. At the same time, the non-technical, project related aspects of the testing exercise is also essential to the success of the initiative as a whole.

There are three areas that are challenging for Agile in the enterprise:

1. Complex inter-dependencies between projects - a reality in any enterprise

2. Handling of Specialized and Global Project Resources such as expert architects in geographically distributed teams

3. Sprint Overhead caused by complex project tasks such as initial architecture design, that would typically not fit in any reasonable sprint duration.

Read the entire article here.

UPDATE 05/04/2011: Followup post on tips for adopting agile in the enterprise. Full article on AgileScout.

Solution Architecture is a design blue print that satisfies a given enterprise problem and a solution architect needs to apply architectural principles to the solution to create that blue print. This is important because success of any architecture is dependent on its uptake.

That is the core difference between a Development Architect and a Solution Architect – a Development Architect CREATES the architecture and a Solution Architect APPLIES it to solve a problem. It takes a different mind-set and is not just some terminology issue. Having done this role; and working with other solution architects, I like to use this movie analogy…

• Development Architect is like the movie equipment producer e.g. camera.
• Solution is the end product — the movie itself.
• Solution Architect is the Director knowing enough of both to make movies.

One cannot produce blockbuster movies with just technical knowledge of the equipment otherwise every movie technician would be winning Oscars for their directorial ventures. What is needed is a James Cameron who has a broad understanding of the solution – market, audience, story, screenplay, direction, background score, cinematography etc – as well as deep knowledge of the tools – 3D, film formats, sound formats, color coordination, special effects, sync-sound, foley sound effects etc. That is the role of a solution architect – someone who has the deep know-how as well as broader vision of creating an Avatar!

In enterprises, solution architects need to understand the architecture – applications, data models, middleware etc – and how to leverage them to satisfy the solution needs – working with stakeholders, identifying the correct need, negotiating scope etc.

The above does not imply in any way that any one type of architect is more important than another. Even Cameron had to wait more than a decade before the technology became mature enough for his vision, so both types of architects are crucial to the success of the solution – the analogy is just to highlight the differences.

The session was quite well attended for a Thursday afternoon (and final day of OpenWorld) with almost 100 attendees. I shared some of the best practices we have learned from our customer implementation experiences, focusing on three areas:

1. REUSE – if you are not reusing services, you might as well not use SOA
2. BUILD NEW – using a holistic approach to SOA with governance utilities
3. PLAN ROLLOUT – to ensure success in your SOA implementation

Here is the webinar of that session.

You can consider this session as a follow-up to the Oracle OpenWorld 2009 AIA Best Practices session.

Update: September 30, 2010: Video of the team’s Saturday day out in Paris, with background score from Shammi Kapoor‘s Bollywood movie “An Evening in Paris“.

Paris

View entire Paris set.

Frankfurt

View entire Frankfurt set.

Hyderabad

View entire India set.

Also, check out the 5 must-see AIA sessions, which includes my session “S316877 – Practical Insight on using Application Integration Architecture” (yes, we do have to spell out AIA in full, but that’s another story). That’s at Moscone West L3, Room 3011 on Thursday, September 23 at 1:30pm. I will be presenting on the usage of various AIA features and tools in a typical implementation, so if there is anything specific you would like to hear about, drop me a comment below.

See you at OpenWorld!

Correction Sep 14, 2010: There are six sessions, not five listed on the page.