Below is the discussion from ISPF-L that Mark Zelden was referring to.
Regards,
John Kalinich
Computer Sciences Corp
--
Monitor your ECSA usage. It may increase if you have dozens or hundreds of
users with ISPPROF datasets as PDSEs.
We normally recommend against having ISPF profile datasets being PDSEs
because ISPF uses update-in-place and has padding in profile dataset
members to usually avoid x37 abends while writing to ISPPROF. Also, as
mentioned above, open PDSEs use storage in ECSA, even if they are not
shared. Finally, the performance of PDS is usually better than PDSE for
ISPF profile datasets.
Dave Levish
Former PDSE Developer
Mark Zelden
<mark.zelden To: [email protected]
@ZURICHNA.COM> cc:
Sent by: IBM Subject: Re: Future ISPF
directions
Mainframe
Discussion List
<IBM-MAIN
08/30/2005 09:29
AM
Please respond
to IBM Mainframe
Discussion List
On Tue, 30 Aug 2005 08:04:04 -0600, Paul Gilmartin
<[EMAIL PROTECTED]> wrote:
>I wondered about the value of PDSE here. Alas, however, our
>test systems exceed in number and span of releases both limits
>for SYSPLEX. Using PDSE would aggravate the ENQ logjam.
>
I thought there was a specific recommendation about not using
PDSE for the ISPF profile data set because of the way it is
updated. This was a performance issue, not a breakage issue.
However, I just scanned some ISPF manuals and could not find
anything. I could just be remembering discussions on ISPF-L.
Mark
--
Mark Zelden
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
