On Jul 5, 2006, at 9:25 AM, Paul Gilmartin wrote: ---------------------SNIP--------------------------
I don't understand "unused space at the end of each block". Surely the interblock gap isn't larger by that amount, so the unused space must exist within the block. The only way I know this can occur is if RECFM=VBS and the block ends with a null segment. I don't believe this is commonly done.
At some time in previous history (pre MVS?) QSAM did allow for short blocks at anyplace in the dataset (including the last block). UNLESS it was created with FBS as the recfm=. I *think* that with SAMe it became standard practice for all blocks (except the last one) to be "full" (ie the blocksize of the dataset). I believe that today blocks are written so that the are "filled up" (ie FBS). VBS is another animal entirely.
I think part of SAM-e original design was to buffer up (more than the then 2 buffers) the file as the cost of IO has always been high.
Our Se (way back in the 70's) authored an orange book to show what a large savings (cpu wise and channel wise) what buffering could save.
Ed
-- gil -- StorageTek INFORMATION made POWERFUL ---------------------------------------------------------------------- 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
---------------------------------------------------------------------- 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
