In a recent note, Ted MacNEIL said: > Date: Fri, 26 Aug 2005 00:00:00 GMT > > >He's got it down (now) to a pretty simple formula: "the maximum number > of SAM buffers (BUFNO=) should be MIN ( 30 , 240K/BLKSIZE). > ... > > 1996. Still a little old, but much closer to reality. > And, I am suspicious of any formula with apparently arbitrary and fairly large constants.
Why 30? Does it depend on some queuing parameter built in to z/OS? What if that changes in a future release? Why 240K? Does that depend on device geometry? Or in some memory management parameter built in to z/OS? What if those change with new hardware, or with the next z/OS release? Why is the nuclear force 137 times stronger than the electromagnetic force? I confess I haven't read Barry. (Nor Feynman.) -- 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