PI63856: THREAD POOL REUSES THREADS INSTEAD OF DISPATCHING TO NEW THREADS

Fixes are available

APAR status

Closed as program error.

Error description

In some cases, the thread pool is waiting for worker threads to
complete their work and reusing them, rather than dispatching
to an idle thread, leading to performance delays.  There is no
stack or anything for the problem, and the "failures" caused by
the issue are specific to the customer's testcase.

Local fix

Problem summary

****************************************************************
* USERS AFFECTED:  Users of IBM WebSphere Application Server   *
*                  experiencing latency in thread pool         *
*                  dispatch times                              *
****************************************************************
* PROBLEM DESCRIPTION: Threads handling work items are not     *
*                      dispatched as quickly as you may        *
*                      expect                                  *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
The WebSphere Application Server thread pool utilizes a thread
pool to dispatch work items to worker threads.  The work items
are placed on a bufffer from which they are pulled by the
pool.  In some cases, particularly when a large number of work
items are dispatched at once, the mechanism used to lock
buffer slots may acquire the locks in an inefficient order,
leading to slight delays in dispatch of new threads (or reuse
of existing threads when a new thread should be created).
Note that this issue has only been observed on AIX, and the
latency was only detected in rare cases, within a strict test
scenario requiring 100ms response times; this is not a
pervasive performance issue.

Problem conclusion

An optional, optimized version of the buffer locking mechanism
was added to the thread pool infrastructure.  To enabled it,
add "-Dcom.ibm.websphere.threadpool.alternateBufferLock=true"
as a generic JVM argument on the server process.

The fix for this APAR is currently targeted for inclusion in
fix packs 8.5.5.12 and 9.0.0.4.  Please refer to the
Recommended Updates page for delivery information:
http://www.ibm.com/support/docview.wss?rs=180&uid=swg27004980

Temporary fix

Comments

APAR Information

APAR number
PI63856
Reported component name
WEBSPHERE APP S
Reported component ID
5724J0800
Reported release
850
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2016-06-09
Closed date
2016-12-22
Last modified date
2017-04-12

APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:

Fix information

Fixed component name
WEBSPHERE APP S
Fixed component ID
5724J0800

Applicable component levels

R850 PSY
UP
R900 PSY
UP

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSEQTP","label":"WebSphere Application Server"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"8.5","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
03 May 2022

Tips

PI63856: THREAD POOL REUSES THREADS INSTEAD OF DISPATCHING TO NEW THREADS

Fixes are available

Subscribe

APAR status

Closed as program error.

Error description

Local fix

Problem summary

Problem conclusion

Temporary fix

Comments

APAR Information

APAR number

Reported component name

Reported component ID

Reported release

Status

PE

HIPER

Special Attention

Submitted date

Closed date

Last modified date

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Fix information

Fixed component name

Fixed component ID

Applicable component levels

R850 PSY

R900 PSY

Document Information

Share your feedback

Need support?