In a recent note, Eric Bielefeld said: > Date: Tue, 3 Oct 2006 10:33:24 -0500 > > Does the PDSE blocksize work like the blocksize on a load module? As I > understand it, if you allocate a loadlib with 32720, and there is a > But why not 32760?
> large enough load module, the 1st block written will be 32720 bytes > long. Probably slightly less, as a couple of other small blocks are > It would seem that the existence of a few initial small blocks shouldn't preclude writing a full 32720 blocks as long as sufficient space remains in the track. > written first. Then, if I understand correctly, the next block is > written out and will fill the track. > Does each member begin on a fresh track? If not, copying a member or compressing the library spoils all this clever planning. > Does the PDSE work similarly, or is my speculation all wet? > My understanding is that PDSEs are always written in 4KiB pages; the specified BLKSIZE is an emphemeral fiction; you might write a member with BLKSIZE=80, then read it with overriding BLKSIZE=32720 and perceive apparently full blocks; or vice versa and never experience "READ WRONG LENGTH RECORD". The access method aims to please you. There ought to be a similar facility for DSORG=PS. Of course, one could always do everything with one-member PDSEs. -- 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
