I did misunderstand you and yes, you are right, one very common change would be an append, and that would change the last TTR field and thus the DSCB hash -- as would any total rewrite that did not happen to be the same length.
Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Kirk Wolf Sent: Thursday, August 20, 2015 10:46 AM To: [email protected] Subject: Re: Ideas for hash of a sequential data set Charles, I think that you misunderstood me. I'm suggesting that the cheap first hash (used to rule out common changes) would be over a combination of: - the first n bytes of the data set (just like you suggested) - the F1/F8 DSCB (which has stuff like DS1LSTAR and DS1TRBAL which helps to detect common changes to the end ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
