CPYTOIMPF performance problem
5 pts.
0
Q:
CPYTOIMPF performance problem
Since moving from V5R4M0 to V6R1M0 our program that copies from a PF to an import file on the IFS using the CPYTOIMPF command has had major performance issues. What used to take 30 minutes to copy now takes as much as 24 hours.
ASKED: Nov 11 2008  4:47 PM GMT
0
23625 pts.
0
A:
 RATE THIS ANSWER
0
Click to Vote:
  •   0
  •  0
  • AddThis Social Bookmark Button

Hi,

Are you on the latest PTF level? Do you get any error messages in the joblog or QSYSOPR? Have you tried running the job in debug to see if it gives any extra diagnostics?

Regards,

Martin Gilbert.
Last Answered: Nov 11 2008  4:51 PM GMT by Gilly400   23625 pts.
0
0
Discuss This Answer:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _



_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Tpinky   4165 pts.  |   Nov 11 2008  9:33PM GMT

I responded to a previous question like this back in Sept. Here is the same response I gave then…

You could try using CPYTOSTMF instead of CPYTOIMPF.

CPYTOIMPF/CPYFRMIMPF uses the underlying DB2 engine. IBM, starting in V5R3, got fussy
about getting your CCSID correct. In previous releases I think IBM used
to do a CPYTOSTMF then a CPYTOIMPF. Breaking it down
into two tasks is another perversion that has shown some success. However
getting the CCSID right is best bet. It’s when you do things like
creating a temporary file in QTEMP that some utilities requires not to be
externally defined (CRTPF … RCDLEN(###) type of file) that makes it
tough.

Hope this helps…

 

Slack400   1165 pts.  |   Nov 13 2008  2:33AM GMT

I’m with Tpinky.
We’re doing the pervse method of using the STMF and then IMPF commands.

 
0