You have my utmost respect but even doctors can be wrong ;)
And it was 2am....just thinking what you think you saw wasn't the whole story...
In other words, maybe it's not that every update of a data area is
forced to disk in an of itself...maybe the data area was journaled and
the journal entry was being forced to disk....or the program doing the
update ended after every update and that cause the data area to be
forced to disk (CHGDTAARA vs. RPG access perhaps?)...
IBM says the following about data areas:
To provide a field that is easily and frequently changed to control
references within a job, such as:
- Supplying the next order number to be assigned
- Supplying the next check number
Again, it doesn't seem to make sense to use them in that manner if
every update to them causes a physical disk IO.
On Mon, May 16, 2011 at 1:37 PM, DrFranken<midrange@xxxxxxxxxxxx> wrote:
It wouldn't seem to make sense to do so if Larry's correct about theNote that it's not truly 'overhead' but rather the way it gets to
overhead of writing to them.
disk. If you've got cache you're good but without it repeated writes to
the same *DTAARA translate to repeated writes to disk where repeated
writes to a Row in a Table don't, UNLESS you close the file every time!
Oh, and *IF*.....? Perhaps you'd better watch over your shoulder
during your next round of PTFs, the Dr Demands respect. :-)
Charles- Larry "DrFranken" Bolhuis
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives