On our HSM MCDS, BCDS, etc, we have 0 for the secondary extent.
If we specified TRACK(50000 1) so we get 1 track as the Control Area,
1. How much would the Index size go up?  15X?
2. Wouldn't CAs be reused a lot more frequently, since you only have
to get 1 track empty instead of 15?
3. Which would be better for an end user?
4. How would throughput be affected?

On Tue, Mar 15, 2016 at 6:07 AM, John Eells <[email protected]> wrote:
> Norbert Friemel wrote:
> <snip>
>>
>>
>> CA_RECLAIM?
>> http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dgt2d4a0/2.5.3.3
>
> <snip>
>
> Norbert beat me to it (smile).  The key to this being useful is that you
> have empty CAs to reclaim.  One could have a pathological application that
> left one record per CA behind after deleting the rest, so that no CAs ever
> emptied out to be reclaimed, for example.  Or, one could have an even *more*
> pathological application that wrote a lot of records into specific key
> ranges, deleted them all (causing reclaims) and then rewrote a lot of
> records into the same key ranges, causing CA reclaim/CA allocation thrashing
> to the detriment of performance.
>
> That said, it's very likely that using CA Reclaim will get you out of the
> reorg business, permanently, with no ill effects, for all of your KSDSs.  CA
> Reclaim is available on all supported releases now, but I have heard of only
> one pathological application so far.  Of course, it was one of ours here at
> IBM.  (You can't make this stuff up.)
>
> --
> John Eells
> IBM Poughkeepsie
> [email protected]
>
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN



-- 
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to