Introduction to the WebSphere Application Server performance tests
The purpose of this project was to measure the various release levels of products used in the Trade workload. Our test results and recommendations are specific to our environment. Parameters useful in our environment might be useful in other environments, but are dependent on application usage and system configuration.
Objectives
The objectives of this project were to gather Linux® end-to-end measurements for all of the components in the path from the WebSphere® Application Server user accessing the system to the database. The products in the path are: SUSE Linux Enterprise Server (SLES), WebSphere Application Server, Java™, and IBM® DB2 Universal Database™ on Linux for System z™ or on z/OS®.
We wanted to show how this set of products, which is needed to run the WebSphere Application Server Trade 6 benchmark on Linux for System z, performs on a release-to-release basis.
The release-to-release comparison was intended to show the effect of new releases on performance and identify the cause of any performance degradations.
Executive summary
- Used WebSphere Application Server 6.1
- Divided the network streams from the WebSphere Application Server to the client and to the database
- Optimized the database layout, which demonstrates that the application itself can significantly contribute to performance improvements
- Introduced HiperSockets™ connections to the database server
Overall, we reached an improvement of about 70% even with the decrease in performance that we experienced with SLES10.
With the database on Linux, the separation of the network streams generates a very high throughput, and accordingly, a very high CPU utilization on the WebSphere Application Server. This limits the success of further turning activity. With HiperSockets connections, we reached a full saturation of the CPUs.