|
View previous topic :: View next topic
|
| Author |
Message |
saravananj
New User
Joined: 23 Nov 2007
Posts: 17
Location: Chennai
|
|
|
|
What is the difference between CPUTIME reported in JESMSGLG for each step and the CPU time & SRB times reported in IEF374I?
For eg: for a test job in our MF,
JOBNAME STEPNAME PROCSTEP RC I/O CPUTIME
TESTJOB STEP1 0 380K 00:17:02.29
TESTJOB STEP2 0 190K 01:21:40.21
TESTJOB STEP3 0 21 00:00:00.27
For same job IEF374I message reports,
IEF374I STEP/STEP1 /STOP 2008068.0129 CPU 1MIN 08.08SEC SRB 0MIN 04.53
IEF374I STEP/STEP2 /STOP 2008068.0139 CPU 5MIN 56.11SEC SRB 0MIN 01.27
IEF374I STEP/STEP3 /STOP 2008068.0139 CPU 0MIN 00.00SEC SRB 0MIN 00.00
As seen above the CPUTIME in JESMSGLG for STEP2 is 01:21:40.21 (HH:MM:SS.SS), where as in IEF374I message it is only 5MIN 56.11 SEC.
How are these values comparable? do they have any relationships. |
|
| Back to top |
|
 |
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
| Quote: |
JOBNAME STEPNAME PROCSTEP RC I/O CPUTIME
TESTJOB STEP1 0 380K 00:17:02.29 |
check with Your support,
do the lines quoted have a message identifier,
those lines are printed by the user written IEFACTRT step/job termination exit
( a properly written exit shoud anyway put a message identifier )
they look more like elapsed times
just check with the timestamp at the beginning of the line and see if there is a rough match
in my system those line look like
| Code: |
13.07.49 JOB00025 HTRT01I CPU (Total)
13.07.49 JOB00025 HTRT02I Jobname Stepname ProcStep RC I/O hh:mm:ss.th
13.07.49 JOB00025 HTRT03I ENRICOI GENER0 00 26 00.06
13.07.50 JOB00025 HTRT03I ENRICOI BATCHPDF 00 375 00.67
13.07.50 JOB00025 HTRT03I ENRICOI DELPROF 00 3 00.01
|
|
|
| Back to top |
|
|
saravananj
New User
Joined: 23 Nov 2007
Posts: 17
Location: Chennai
|
|
|
|
do the lines quoted have a message identifier,
No the lines quoted do not have a message identifier.
they look more like elapsed times
just check with the timestamp at the beginning of the line and see if there is a rough match
in my system those line look like
[code]13.07.49 JOB00025 HTRT02I Jobname Stepname ProcStep RC I/O hh:mm:ss.th
13.07.49 JOB00025 HTRT03I ENRICOI GENER0 00 26 00.06
13.07.50 JOB00025 HTRT03I ENRICOI BATCHPDF 00 375 00.67
13.07.50 JOB00025 HTRT03I ENRICOI DELPROF 00 3 00.01
The elapsed time comes right after the CPUTIME in our MF. I will check with support on how to interpret CPUTIME |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
there are some discrepancies in Your data
could You pleas post also the IEF373I message
You posted
| Code: |
IEF374I STEP/STEP2 /STOP 2008068.0139 CPU 5MIN 56.11SEC SRB 0MIN 01.27
IEF374I STEP/STEP3 /STOP 2008068.0139 CPU 0MIN 00.00SEC SRB 0MIN 00.00 |
if the second step had an elapsed of 1h 21m
the ief374 for the third step should report an end time of 20008068.0250 |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
the cpu time is the time spent executing instruction...
in PCese heating up the CPU and the motherboard
roughly... elapsed = cpu + wait ( system, i/O, operator intervention ) |
|
| Back to top |
|
|
saravananj
New User
Joined: 23 Nov 2007
Posts: 17
Location: Chennai
|
|
|
|
| enrico-sorichetti wrote: |
the cpu time is the time spent executing instruction...
in PCese heating up the CPU and the motherboard
roughly... elapsed = cpu + wait ( system, i/O, operator intervention ) |
This should clear all confusions,
01.29.34 JOB21922-JOBNAME STEPNAME PROCSTEP RC I/O CPUTIME ELAPSED
01.29.34 JOB21922-TESTJOB STEP1 0 380K 00:17:02.29 01:18:58.04.
01.39.19 JOB21922-TESTJOB STEP2 0 190K 01:21:40.21 00:09:44.58.
01.39.20 JOB21922-TESTJOB STEP3 0 21 00:00:00.27 00:00:00.40.
The elapsed time comes right after the CPUTIME in our MF |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
| Quote: |
This should clear all confusions,
01.29.34 JOB21922-JOBNAME STEPNAME PROCSTEP RC I/O CPUTIME ELAPSED
01.29.34 JOB21922-TESTJOB STEP1 0 380K 00:17:02.29 01:18:58.04.
01.39.19 JOB21922-TESTJOB STEP2 0 190K 01:21:40.21 00:09:44.58.
01.39.20 JOB21922-TESTJOB STEP3 0 21 00:00:00.27 00:00:00.40.
|
nahhhh  it adds confusion
what I cannot understand is the 1:21:40.21 bolded in the quote
/rough numbers on
in the fiirst step You have 17 minutes of cpu and 1hout 18 minutes elapsed
in the second step Yo have 10 minutes elapsed which matches the timestamp at the beginning of the line
but.... where do 1hour 21 minutes of cpu time come from
that would be a heck of parallel processing
ok for the step 3
if You posted the ief373i and the ief 374i again and checked again the data
we might get to the final conclusion
( at a first sight looks like Your iefactrt might be displaying the wrong data)
please postusing the code tag ( after cut and paste all the columns will be aligned ) |
|
| Back to top |
|
|
saravananj
New User
Joined: 23 Nov 2007
Posts: 17
Location: Chennai
|
|
|
|
| Code: |
00.10.36 JOB21922 IEF403I TESTJOB - STARTED - TIME=00.10.36
01.29.34 JOB21922 -TESTJOB *INIT* 08.068 0:10:36 * - -TIMINGS (HH:MM:SS.SS)- -.
01.29.34 JOB21922-JOBNAME STEPNAME PROCSTEP RC I/O CPUTIME ELAPSED
01.29.34 JOB21922-TESTJOB STEP1 0 380K 00:17:02.29 01:18:58.04.
01.39.19 JOB21922-TESTJOB STEP2 0 190K 01:21:40.21 00:09:44.58.
01.39.20 JOB21922-TESTJOB STEP3 0 21 00:00:00.27 00:00:00.40. |
It was the same 01:21:40.21 that confused me as well. All the IEF374I messages posted earlier are correct. I have checked them again. Thanks for the "code" stuff too. |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
my iefactrt prints more info on the right...
| Code: |
CPU (Total) Elapsed CPU (TCB) CPU (SRB) Service
RC I/O hh:mm:ss.th hh:mm:ss.th hh:mm:ss.th hh:mm:ss.th Units
00 26 00.06 00.22 00.06 00.00 415
00 375 00.70 01.37 00.67 00.03 4634
00 3 00.01 00.06 00.01 00.00 89
|
can You look if You get something similar and check that cpu_total = cpu_tcb + cpu_srb ( roughly ) |
|
| Back to top |
|
|
saravananj
New User
Joined: 23 Nov 2007
Posts: 17
Location: Chennai
|
|
|
|
| Our IEFACTRT does not print anyother information in the right, other than the ones posted in my last post. Thanks |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10873
Location: italy
|
|
|
|
Looks like Your support group will have to fix the IEFACTRT code
once upon a time the IBM sample had errors, somebody should check the source of Your exit
the are lots of samples around very well written and worth evaluating
with good formatting, and a little bit more info
good luck |
|
| Back to top |
|
|
|
|
