In a message dated 9/12/2006 11:24:14 A.M. Central Daylight Time, [EMAIL PROTECTED] writes:
>If the volume is dedicated to JES2 spool (i.e. only one >dataset), then I'd make the VTOC(0,1,14) with no index at all. That >leaves the rest of the volume for the dataset. You could even go to >VTOC(0,1,1). There is no need for an index at all. An index is only >useful if you have a large number of VTOC I/Os going on. On this volume, >once JES2 has OPEN'ed the HASPACE dataset, the VTOC should not be hit >again until the next JES2 restart. If you're going to have only a one-track VTOC in cylinder 0 and the rest of the volume dedicated to SYS1.HASPACE allocated by cylinders, why not put a small VTOC index in cylinder 0 along with the small VTOC? There might be a metadata reason why it's better to have an indexed VTOC on all volumes. E.g., some day all volumes might have to be SMS-managed. Speaking of VTOC hits, does SDSF hit the VTOC when a SYSOUT access begins? Bill Fairchild ---------------------------------------------------------------------- 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
