On Jan 21, 2013, at 8:46 AM, Joel C. Ewing <[email protected]> wrote:

> I have never seen any indication that having excessive free DSCBs has any 
> impact positive or negative on new data set allocation, and if you have a 
> VTOC Index defined (which should always be the case) there should be minimal 
> correlation between VTOC size or used DSCBs, and response time when accessing 
> the VTOC.  On the other hand, having no DSCBs left and having unusable free 
> space left on the volume that you would like to use is a serious 
> inconvenience. Throwing extra cylinders at the VTOC at initialization has a 
> minimal cost in DASD space and can avoid future grief.  Having to 
> micro-manage VTOC sizes and/or location is not a profitable use of time.

The only downside of a too large VTOC is that the extra space is not available 
for data sets. Since I've set up the SMS pools by size, I can adjust the VTOC 
size to match each pool. I have an initialization job for each pool. It's not a 
lot more difficult than "one size fits all."

> 
> I always tried to come up with a "worst-case" VTOC size requirement for each 
> DASD size and have a single initialization job that could be used in all 
> cases.  It makes life so much simpler and you don't have to worry about 
> things later if you decide to use the same volume for smaller data sets, or 
> users circumvent your well-planned SMS strategies by allocating many "large" 
> datasets with RLSE that end up small.

I have a daily job that finds these data sets and uses FDRDSF to move them to 
the appropriate pool.

-- 
Curtis Pew ([email protected])
ITS Systems Core
The University of Texas at Austin

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

Reply via email to