On 10/31/11 13:59, Edward Jaffe wrote:
On 10/31/2011 10:10 AM, Mark Jacobs wrote:
We just started using the zOS Runtime Diagnostics under zOS 1.12 and
while I understand most of what it's telling us, I don't fully
understand what it's trying to tell me, or what I can do about it
when it reports on a HIGH LOCAL LOCK SUSPENSION RATE.
Local lock suspension occurs when some executing code attempts to get
the local lock (which is required for many services e.g., STORAGE) and
another unit of work already has it. The requester is suspended until
the lock is released by the other unit of work.
If the rate is high (above average), it usually means some code is
doing significant processing with the local lock held. It's not
necessarily 'wrong'. Such statistics are a 'heads up' for the
programmer(s). Is it your code that's executing? Or someone elses?
Do you have 'Vsm UseZosV1R9Rules' coded in DIAGxx? Using the new rules
can make STORAGE run faster, thus reducing local lock contention. YMMV.
We're seeing it often for the HSM address space. We have UseZosV1R9Rules
set to No in Diagxx.
--
Mark Jacobs
Time Customer Service
Tampa, FL
----
One of life's greatest mysteries is how the boy who
wasn't good enough to marry your daughter can be the
father of the smartest grandchild in the world.
Yiddish Proverb
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
