I read Pete Massiello's article "How Much Temporary Storage Is In Use And
Who Is Using It?" at
http://www.itjungle.com/fhg/fhg072115-story03.html
While this is a welcome addition to WRKACTJOB it only shows jobs. There
are some big system tasks that consume temporary space also. So I read
this:
http://www.ibm.com/support/knowledgecenter/ssw_ibm_i_72/rzajq/rzajqviewsystmpstg.htm
ran this:
select *
from qsys2.systmpstg
order by bucket_current_size desc
And got:
GLOBAL_BUCKET_NAME JOB_NAME JOB_USER_NAME JOB_NUMBER
BUCKET_CURRENT_SIZE
*DATABASE Segment Cache - - - 30,166,245,376
*MACHINE - - - 15,971,721,216
*DATABASE DSI SQE MTI - - - 2,759,327,744
- QZRCSRVS QUSER 647911
2,443,108,352
*DATABASE SQE Heap - - - 1,680,904,192
- QYPSJSVR QYPSJSVR 309571
1,668,739,072
*IPC - - - 943,480,832
- Q1ACPDST QBRMS 146711
770,519,040
That's about 50GB used by tasks instead of jobs.
What is this "*DATABASE Segment Cache"?
Is there a way to reduce this which doesn't affect performance negatively?
For example, if someone can show me that this is occurring because I have
too many database extents and I need to run a RGZPFM to reduce these
extents and it would be even better if I ran RGZPFM with the appropriate
KEYFILE parameter (or some other reason and possible solution) that is
what I am looking for.
Auxiliary storage:
System ASP . . . . . . : 2218 G
% system ASP used . . : 75.5989
Total . . . . . . . . : 2218 G
Current temporary used : 73985 M
Peak temporary used . : 133076 M
So we peaked out at roughly twice the size of the smallest disk drive
supported by IBM i 7.2? Granted, it is only ~6% of our disk.
Rob Berendt
As an Amazon Associate we earn from qualifying purchases.