« Sharing of data across multiple Virtual Machines | Main | Distributed BTree »

SOA Process Dependency Tracking

Experimenting with ways to track and represent process dependencies in a SOA Integration. If you know any of the processes involved as part of an integration, just type the name and you can get the entire dependency in a graphical mode that has been scaled and designed to handle large number of processes.

You can pan,scale,spatially distort, do anything that is convenient for analyzing dependent processes, fan-outs and fan-ins,cyclic runs, message interaction patterns, without knowing anything about Oracle SOA Suite or having to delve into JDeveloper based BPEL or ESB code.
You can also view parts or whole of the involved processes.

Moreover, the dependencies are represented as GraphML. This means that any third party application that can read and parse XML (like Flex) can hook their front ends to represent the graphical view in any way that they want without knowing a single thing about Oracle SOA Intricacies.

A small video that shows how the dependency trackers crawls through all the processes in an Oracle AIA demo based on Foundation Pack 2.2.

scrawler_small.jpg

Current Enhancements being done:

1) Automated Testing: The information present in the GraphML will be used to gather more information like the messages exchanged between the processes (cardinality,type,etc).The messages will be tested against preconfigured values or ranges in an incremental manner. The GraphML representation will also be used to construct necessary stubs when individual processes require to be tested.

2) Connectivity with WebLogic: Currently sCrawler supports only Oracle Containers for J2EE. Adding functionality for Weblogic and OSB.

3) Automated Deployment: sCrawlers dependency tracking is being extended to incorporate the ability to find dependency of SOA Processes source code. The idea is to deploy processes starting from the edge vertices with the least edges and move into the vertices with maximum edges. Also, sub-graphs may be found which can be independently deployed. This will help in making certain deployments in parallel.

4) JDeveloper Extension: Working on JDeveloper extension plugin. This will help developers to have an end-to-end view while they work on an existing SOA Ecosystem.

5) ANTlibs for sCrawler: For cases where users intend to use a third party graph renderer a separate ANT library will be made available. This library will generate the GraphML documents which will be ready-to-import and use.


Detailed White Paper on sCrawler

TrackBack

TrackBack URL for this entry:
http://blogs.oracle.com/mt/mt-tb.cgi/8954

Comments (1)

Partha:

Good one! Appreciate the effort put in!

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

About This Entry

This page contains a single entry from the blog posted on December 13, 2008 8:29 AM.

The previous post in this blog was Sharing of data across multiple Virtual Machines.

The next post in this blog is Distributed BTree.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type and Oracle