Performance considerations in multithreaded applications

There are several performance considerations for multithreaded jobs on i5/OS™.

The concepts that are described here pertain to all programming languages. For details about how each language enables a concept, refer to the Language topic in the information center for that specific language.

• Multithreaded server recommendations
  When you write a typical server to take advantage of threads, a common design is to have a single listener thread waiting for client requests and multiple individual worker threads to perform the client's requested operations.
• Job and thread priorities
  An i5/OS thread competes with threads throughout the system for scheduling resources in addition to competing with other threads in the same job.
• Contention among threads
  Contention occurs when one thread has to wait for another thread to finish using a resource.
• Effects of storage pool sizes on threaded applications
  The storage pool size specified for a subsystem affects the performance and number of threads you can create.
• Activity levels of storage pools
  The activity level of a storage pool refers to the number of active threads within that storage pool.
• Performance and threaded applications
  Job-level performance counters are updated by all the threads in a job.