Re: Fwd: ILE and locked file

Hi Gad,

Thanks, I didn't catch that you were doing a DELETE FROM; I thought it was DLTF.

That's pretty interesting though.  I wonder what the SQL DELETE FROM is doing that makes it so much faster than CLRPFM?  I remember reading at one time that DELETE FROM would attempt to do a CLRPFM, and if that couldn't be done due to locks, it would attempt to delete all the individual records.

--
*Peter Dow* /
Dow Software Services, Inc.
909 793-9050
petercdow@xxxxxxxxx
pdow@xxxxxxxxxxxxxx /

On 7/22/2022 7:40 AM, Gad Miron wrote:

Hi Peter

Trying to speed up a program that executes recursive-like CLRPFM of a file
in QTEMP
and re-populate it .
We tried executing SQL delete from QTEMP/TEMPF instead of CLRPFM
and it was about twice faster.
We did not use DLTF + CRTPF, only CLRPFM.
I was later told (Was it Birggita?) that CLRPFM is a costly operation
indeed.

HTH
Gad

date: Thu, 21 Jul 2022 11:07:40 -0700
from: Peter Dow<petercdow@xxxxxxxxx>
subject: Re: Fwd: ILE and locked file

Hi Gad,

Did you count the time required to re-create the file?

DLTF + CRTPF < CLRPFM?

--
*Peter Dow* /
Dow Software Services, Inc.
909 793-9050
petercdow@xxxxxxxxx
pdow@xxxxxxxxxxxxxx /

On 7/21/2022 7:59 AM, Gad Miron wrote:

My 2 cnts

In a certain job that repeatedly clears and re-populate a (small) work

file

in QTEMP
we found that delete from QTEMP/WORKF is much faster then CLRPFM
QTEMP/WORKF

Gad