View previous topic :: View next topic
|
Author |
Message |
MFSubbu
New User
Joined: 16 Feb 2008 Posts: 19 Location: Bangalore
|
|
|
|
Hi,
My Requirement is to get the job name, job start time, job end time, job run time(duration it takes for the job to complete).
I tried using SMF type 5. There were aroung 50K records. I splitted them into 10K each and then using REXX i filtered the required information. The start time (SMF5RST) and the end time(SMF5RSTT) remains the same for all job, I am not sure why it is so. I also thought of achieving this through SMF type 30. but in this i have 500,000 records. Below are my questions
1) Rexx is not capable of reading large file and it is not efficient as well.. is there any other way to achieve the output other than using REXX
2) Why the start time and end time of the output remains the same. Manually i checked few jobs which run for longer time, even for those both the time remains the same.
3) I parsed SMF5TRAN and SMF5TTAT to findout which gives the job run time information, but both didnt give me the run time. |
|
Back to top |
|
|
Bill Dennis
Active Member
Joined: 17 Aug 2007 Posts: 562 Location: Iowa, USA
|
|
|
|
Those fields are the time spent getting the job through the INTRDR. That is very quick.
The field you want to use are
SMF5JIT - job initiation time
SMF5TME - job end time
The difference between them is runtime. Don't forget the job many end on the next day or several days later. Check those fields, too. |
|
Back to top |
|
|
Robert Sample
Global Moderator
Joined: 06 Jun 2008 Posts: 8696 Location: Dubuque, Iowa, USA
|
|
|
|
Also note that when IBM recommends something, wise programmers pay heed. From the SMF manual:
Quote: |
Note: IBM recommends that you use record type 30 rather than record types 4, 5, 20, 34, 35, and 40. Use of record type 5 may cause you to miss key workload indicators. |
Have you checked with your site support group to see if they are using MICS or MXG or another package that processes SMF data into easily usable SAS data bases or flat files? |
|
Back to top |
|
|
MFSubbu
New User
Joined: 16 Feb 2008 Posts: 19 Location: Bangalore
|
|
|
|
Bill,
Now it is working fine..
Robert,
We donot have any package to convert SMF data.
After fetching the result only I noticed the output includes all the jobs submitted by the scheduler and also by the users.. Is there any way i can filter the jobs submitted by the scheduler alone. I mean to get the OWNER of the job from SMF data. I could see SMF30JNM will fetch that info... but as i said SMF typer 30 fetching 5 lakh records and REXX will not be able to read the file. Is there any way to filter only the jobs submitted by the scheduler.
Thanks in advance... |
|
Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006 Posts: 19244 Location: Inside the Matrix
|
|
|
|
Hello,
Why do you believe this all must be done with rexx? |
|
Back to top |
|
|
CICS Guy
Senior Member
Joined: 18 Jul 2007 Posts: 2146 Location: At my coffee table
|
|
|
|
Doesn't DFsort have some good facilities to read and prune some of those SMF type 30 records prior to feeding them to REXX? |
|
Back to top |
|
|
MFSubbu
New User
Joined: 16 Feb 2008 Posts: 19 Location: Bangalore
|
|
|
|
Thanks everyone...
I used Syncsort to filter the records and then used REXX to generate the report and it worked.... |
|
Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006 Posts: 19244 Location: Inside the Matrix
|
|
|
|
Good to hear you have a solution - thank you for letting us know
d |
|
Back to top |
|
|
PeterHolland
Global Moderator
Joined: 27 Oct 2009 Posts: 2481 Location: Netherlands, Amstelveen
|
|
|
|
SMF5Rxxx is the reader start/stop time. And that will be almost always
be the same. |
|
Back to top |
|
|
|