Monday, June 29, 2009

Best practices for BPM, SOA and EDA

While visiting ODTUG Kaleidoscope 2009 in Monterey and talking to fellow BPM, SOA and EDA adepts I got this idea about creating a best practices and lessons learned blog series. This first blog is dedicated to best practices in the BPM and SOA-space based on cases from a presentation by Lonneke Dikmans. Subsequent blogs will dive into best practices and lessons learned for a specific product, methodology or technology.

Case I: Introducing BPM. Mistake: Organizational impact underestimated. Explanation: Successful delivery of BPM project, business heavily involved. Never used because they realized after delivery that changes in both the organization and the software were needed. Best practice 1. BPM and SOA are about business, IT and humans. Observe how people work, don’t just ask them.

Case II: Notifications. Mistake: Dependencies between processes modeled directly in the processes itself. Explanation: Process flow sometimes is influenced by other processes. This was modeled into every process: this makes processes tightly coupled to each other and hard to change. It resulted in deadlocks. Best practice 2. Use events to notify running processes. Best practice 3. Monitor & avoid exceptions.

Case III: New technology. Mistake: Use BPEL as a general purpose language. Explanation: BPEL is a domain specific language; it was designed to orchestrate (web) services. Someone coming from a homogeneous -for example PL/SQL environment- in their back office, could decide to rewrite everything in BPEL, even the service implementations. The progress of such a project is very slow, and doing things that used to be easy becomes very hard. Best practice 4. BPEL is a Domain Specific Language (DSL); use BPEL for orchestration only. Best practice 5. Use an Enterprise Service Bus (ESB) to expose services to consumers including BPEL. Best practice 6. Use Java for service implementation. Best practice 7. Use PL/SQL for persistent data manipulation and data integrity rules. Best practice 8. Use rules when you need customization, inference or when business rules are volatile.

Case IV: Quality of Service. Success: involve administrators early. Explanation: Someone designed their first SOA project with quality of service in mind. In production all the non-functional demands were met. Best practice 9: Design architecture for Quality of Service from the start … but only what you really need! Not everyone needs clustering, fail-over, high-availability, and so on.

Case V: Domains. Success: combine a top down with a bottom up approach. Explanation: By defining 6 business domains and one supporting domain, the service taxonomy and event definitions were easier to keep track of. Also defining an owner for some of the services and design guidelines for services that cross domains become possible. Best practice 10. Use domains and layers to facilitate making a taxonomy of services and defining design guidelines.

Conclusion. Think big, start small. Meet in the middle requires aligning Business, IT and People. Architects can be intermediaries. Sharing knowledge and experience is necessary.

The next blog in this series will dive into Web Services best practices.

Friday, June 26, 2009

ODTUG Kaleidoscope 2009

ODTUG Kaleidoscope 2009 is almost coming to an end at the time of this writing. After some chilly days the sun started to shine in Monterey and turned this great event into an even better one. As mentioned almost every day, ODTUG is one of the few conferences that has grown compared to last year. This was my first visit to ODTUG. I actually thought it would be as big as Open World :-) But it’s about a hundred times smaller. And actually that’s cool; much more intimate. Here you really get the chance to speak to product managers, interact with peers, and meet lots of new and interesting people!

So what were the highlights?

SOA and BPM Symposium on Sunday. For the first time there was a separate symposium and track dedicated to SOA and BPM. It was put together by ACE Directors Lonneke Dikmans, Lucas Jellema and Mike van Alst. Although the APEX and database tracks attracted more audience, we had a very interesting and interactive day with a mix of newcomers and SOA-adepts. The day was split into a business and technology part. Breakout sessions were mixed with great presentations by Demed L’Her, Geoffroy de Lamalle, and Clemens Utschig. Read about the results -SOA and BPM approaches- on Oracle Wiki.

Fusion Apps demo. The first official demo of the upcoming Fusion Apps. It looked really smooth! It’s build on top of the new Oracle Fusion Middleware 11g stack (WebCenter, ADF, SOA Suite). Lot’s of social networking capabilities and interaction. Expect to see more of this. Some technology stats: approximately 11,000 task flows, between 5 and 6 thousand tables, tens of thousands of ADF BC View Objects, and so on.

Oracle ACE dinner. Great dinner followed by a bonfire and s’mores on the beach!

Presentations in the SOA and BPM track. Lots of interesting presentations here. I was an ambassador for Lonneke’s presentation and did a presentation myself on SOA in a database-centric environment. Also great presentations by Roman Dobrik on BPEL development patterns, Chris Judson on canonical data models, Mauricio Naranjo on a government SOA project in Latin America, Samrat Ray on SCA in SOA Suite 11g, Mark Simpson on tools for business processes, and Lucas Jellema on SOA in an Oracle classic stronghold.

Meeting fellow geeks. You want to meet people that drive cars with license plates like “BPEL” or “WEB 2 OH”? Find them at ODTUG!


Great conference! Thanks to everyone and looking forward to meet everyone again at Oracle Open World 2009!

Thursday, June 18, 2009

Passive adapters in Oracle ESB that won’t be activated

Configuring SOA Suite 10g for high availability (HA) isn’t the most easy thing to do. Several administrators I spoke with and worked with in projects brought this up. I really hope that FMW 11g -besides all the new functionality, enhancements and support for new standards such as SCA- also makes things like HA easier to configure.

One particular issue we recently ran into in one of our projects has to do with the use of non-concurrent adapters in Oracle ESB when upgrading our clustered environment from 10.1.3.3 to 10.1.3.4. Non-concurrent (or singleton) adapters are adapters that cannot run in an active-active configuration since the underlying infrastructure does not provide a good locking mechanism. Examples are file and FTP adapters. JMS and database adapters on the other hand support concurrency. For non-concurrent adapters you have to ensure that there is only one adapter instance active at runtime. Otherwise you could have two active file adapters both reading the same file, starting two ESB flows instead of one. Futhermore, you want to have fail-over. If the ESB RT (runtime) node on which the active file adapter is running (or adapter itself) fails, the passive adapter on another ESB RT node should be activated. In earlier SOA Suite 10g releases you had to install and configure a separate ESB RT for this (ESB Singleton) and deploy non-concurrent adapters to this separate node. Real overkill. Fortunately, in later versions you could deploy non-concurrent adapters to the existing ESB RT’s and configure these adapters in an active-passive configuration by setting the clusterGroupId property. The jGroups protocol is then used so that only one instance of all adapters that have the same clusterGroupId value will be activated.

When we upgraded to SOA Suite 10.1.3.4 none of our file adapters in the acceptance environment was activated anymore! After some investigation it seemed that ESB 10.1.3.4 uses its own jGroups configuration instead of the jGroups configuration as specified in the global jgroups-protocol.xml file (as was the case for ESB 10.1.3.3). That isn’t a problem by default. However, in our case both our test and acceptance environment are clustered and both run in the same network. The internal jGroups configuration of both test and acceptance by default probably use the same ip and subnet addresses. Meaning all adapters of all ESB projects in the same network with the same clusterGroupId are all put in the same active-passive configuration. For ESB project “A” only one file adapter instance for test was active, the same file adapters for ESB project “A” for acceptance were all in passive mode. Luckily you can specify the useJgroupConfigFile property for an ESB endpoint and set it to true to enforce using the jgroups-protocol.xml configuration file; as was the case in ESB 10.1.3.3. Then configure a different ip and subnet address combination for test and acceptance. That way the non-concurrent adapters in the same ESB projects but in different environments are separated when they have the same clusterGroupId. Another workaround would be to include the environment name in the clusterGroupId value, e.g. MY_ESB_TEST_ID and MY_ESB_ACCEPTANCE_ID.

Tuesday, June 2, 2009

Web Services article on OTN

Oracle recently released Oracle Enterprise Pack for Eclipse (OEPE) 11g. OEPE is a certified set of Eclipse plug-ins that is designed to help develop, deploy, and debug applications for Oracle WebLogic Server. I wrote an article on OEPE’s Web Service capabilities, and more specifically its support for the JAX-WS and JAXB standards. The article includes a step-by-step tutorial, explains different approaches to Web Service development, and concludes with several best practices. You can find the article on Oracle Technology Network (OTN).