TIBCO Business Works is one of the most widely used module of TIBCO Software Suite. TIBCO Business Works is considered highly scalable, robust and reliable platform for developing and designing Integration business processes. Business Works Engine runs on JVM (Java Virtual Machine). Following configuration changes for JVM can greatly help in improving and fine Tuning TIBCO BW Performance.
BW Engine Heap Size
The amount of memory to be used by BW engine is defined in bwengine.tra and designer.tra configuration files. By changing the value of this parameter can help tuning TIBCO Business Works performance.
For example, in my TIBCO installation, I changed this value from 256M to 512M which greatly improved my BW Engine performance.
tibco.env.HEAP_SIZE=512M
Java Objects Garbage Collection
Garbage collection is the process of destroying Java objects to release the memory being used. Well optimized garbage collection helps in efficient utilization of memory resources. You can tune performance of Tibco Business Works by changing Java garbage collection parameters in the configuration file.
You can set the Java memory parameters by using the java.extended.properties option in the bwengine.tra configuration file.
For example, specify the following in the bwengine.tra to set the heap size to 512M and retrieve the garbage collection metrics for the engine.
java.extended.properties -Xms512m -Xmx512m -verbose:gc
Additionally, refer to my video tutorial about TIBCO BW Design & Development Best Practices which will help you to apply best practices in your project implementations.
Certain deployment configurations are also essential for performance improvement which are covered in An Overview of Using MAX JOBS, FLOW LIMIT and ACTIVATION LIMIT For Process Configuration tutorial and you may also benefit from those.
Hello Ajmal,
Need you help to set property to explicitly call to Garbage Collection. My application is not returning back memory to OS after going to 7 GB of use. We have given 8 GB max heapsize and 32 MB as minimum.
thread count as 16 and Java Thread Stack Size (KB) as 3072