Ted, The time in *MASTER* would be the time spent by ASM moving pages in and out of storage, but I believe the clock is still running for a TCB or SRB when the page fault occurs, and after satisfying it there can be HSB misses if the task is not dispatched to the same IP again.
My observations are that TCB and SRB would increase during periods of high page-in. The master Address Space would jump whether there was high Page-in or Page-out rate, but was relatively unaffected by a high reclaim rate. Admittedly I last looked at this around the time I worked on an ESP for a 3090-200, trying to quantify the benefits of ES, and the cost of taking it away, but I don't recall anything that suggests this would have changed. Ron > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On > Behalf Of Ted MacNEIL > Sent: Monday, February 04, 2008 11:41 AM > To: IBM-MAIN@BAMA.UA.EDU > Subject: Re: [IBM-MAIN] SPAM: CPU time differences for the same job > > >I always thought that page faults and time in the dispatcher were > recorded for the TCB or SRB that incurred the time. > > I was taught they were recorded under SRB for *MASTER* and used that > for justification of a memory upgrade many years ago. > > - > Too busy driving to stop for gas! > > ---------------------------------------------------------------------- > 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
