We can calculate the wasted space given the block size, average number of blocks per track, average number of this, that, and the other thing, but wasted space is meaningless and unknowable unless you are using a SLED 3390 which means now at least 15, maybe 20-year old disk drives. With the advent of RAID, the interrecord gaps have been virtualized. Another consideration is that you want to know which of two alternatives is "better". My answer is that it depends - on what variable you wish to optimize, space utilization being only one of many possible variables for which you might want to find a "best" value. Speed of access is another variable, and it could mean either elapsed time, average connect time, channel utilization, etc. Elapsed time can usually be minimized by having all blocks be the maximum possible size. Connect time, conversely, is reduced by transferring smaller block sizes per I/O request. But the more I/O requests you do, the more CPU time your application will eat up in doing the I/Os. And larger block sizes need more real storage to back the larger buffer sizes during each I/O, which impacts the working set size of your application and possibly causing other work in the system to suffer more paging. Bill Fairchild Franklin, TN
----- Original Message ----- From: "Scott Ford" <[email protected]> To: [email protected] Sent: Friday, December 6, 2013 2:38:50 PM Subject: Re: Something to Think About - Optimal PDS Blocking Gerhard and Lionel, How much wasted space is there in the blocks or blksize, slack bytes was the old name many moons ago Scott ford www.identityforge.com from my IPAD 'Infinite wisdom through infinite means' > On Dec 6, 2013, at 3:30 PM, Gerhard Postpischil <[email protected]> wrote: > >> On 12/6/2013 1:41 PM, Dyck, Lionel wrote: >> Which is better - a BLKSIZE of 3120 with 100 members ranging from 3 >> to 30 records or that same set of records in a PDS with BLKSIZE of >> 27920. And similarly what about a PDS with BLKSIZE of 3120 and 100 >> members ranging from 300 to 9000 records compared to the same set of >> records in a PDS with a BLKSIZE of 27920. > > I just ran a test on my macro and source libraries. Each has 80 used > directory blocks. > > MACLIB 98 tks with 3120 94 tks with 27920 > SOURCE 829 tks with 3120 442 tks with 27920 > > In practice I've found the optimal block size to be close to half track, with > one exception (either 3K or 7K record size?), where five block per track > actually gave better usage. > > Gerhard Postpischil > Bradford, Vermont > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
