You can tune a process model for quality of service and performance. Quality of service is a measure of reliability, visibility, and control. In general, you achieve high quality of service by persisting data as a process executes. Performance is a measure of the time it takes a process to complete and the number of processes completed per time period. In general, you achieve good performance by using volatile data storage (server RAM) while a process executes. An increased quality of service is achieved at the expense of performance and vice versa.

In a maximum quality of service/minimum performance environment, processes are guaranteed; that is, a process automatically recovers at the appropriate step in case of system failure, and the process is guaranteed to run to completion. However, in this case, each process takes longer to complete and fewer instances complete per time period.

In a maximum performance/minimum quality of service environment, processes complete more quickly and more processes complete per time period. However, processes are not guaranteed and might not automatically recover in case of server failure; that is, if a server fails, the process might not run to completion. In addition, there is no visibility or control through webMethods Monitor. You cannot take any of the actions through webMethods Monitor that are available in a maximum quality of service environment.

The level of quality of service and performance that you require depends on your business needs. The Process Engine quality of service and performance settings are highly configurable and can be set on a process model by process model basis; you can tune your environment to a quality of service or performance extreme or anywhere in between.