Peak(Peak Requests Queued in the Admin Console) is the highest number of requests that were
everqueued up simultaneously for the use of a native thread since the server was started. This
valuecan be viewed as the maximum concurrency for requests requiring a native thread.
Limitis the maximum number of requests that can be queued at one time to wait for a native
thread,and is determined by the setting of NativePoolQueueSize.
Tuning
Youcan modify the queue size for NativePool by editing the NativePoolQueueSize directivein
magnus.conf.For more information, see “NativePoolQueueSize Directive” on page 67.
NativePoolStackSizeDirective
TheNativePoolStackSize determinesthe stack size in bytes of each thread in the native
(kernel)thread pool.
Tuning
Youcan modify the NativePoolStackSize byediting the NativePoolStackSize directive in
magnus.conf.
NativePoolQueueSizeDirective
TheNativePoolQueueSize determinesthe number of threads that can wait in the queue for the
threadpool. If all threads in the pool are busy, then the next request-handling thread that needs
touse a thread in the native pool must wait in the queue. If the queue is full, the next
request-handlingthread that tries to get in the queue is rejected, with the result that it returns a
busyresponse to the client. It is then free to handle another incoming request instead of being
tiedup waiting in the queue.
Settingthe NativePoolQueueSize lowerthan the maximum threads value causes the server to
executea busy function instead of the intended NSAPI function whenever the number of
requestswaiting for service by pool threads exceeds this value. The default returns a “503
ServiceUnavailable” response and logs a message, depending on your log level setting. Setting
theNativePoolQueueSize higherthan the maximum threads causes the server to reject
connectionsbefore a busy function can execute.
Thisvalue represents the maximum number of concurrent requests for service that require a
nativethread. If your system is unable to fulll requests due to load, letting more requests queue
upincreases the latency for requests, and could result in all available request threads waiting for
anative thread. In general, set this value to be high enough to avoid rejecting requests by
anticipatingthe maximum number of concurrent users who would execute requests requiring a
nativethread.
UsingMonitoringData to TuneYour Server
Chapter2 • Tuning Sun Java System WebSer ver 67