Subject: | |
From: | |
Reply To: | |
Date: | Tue, 14 Feb 2006 10:10:12 -0800 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Forget all the rules-of-thumb. On MPE, memory utilization is best
viewed from the perspective of how effective it is in acting as a cache.
The two top metrics in evaluating this is Read Hit Percentage (the
percentage of all read I/O's satisfied in RAM) and CPU Busy on Memory
Management.
You want a Read Hit Percentage in excess of 95% and CPU Busy on Memory
Management less that 4-5 percent.
Finally, the best axiom about memory on MPE is to buy all you can.
Very, very rarely is there such a thing as too much memory.
Bill Lancaster
-----Original Message-----
From: HP-3000 Systems Discussion [mailto:[log in to unmask]] On
Behalf Of Dave Waroff
Sent: Tuesday, February 14, 2006 10:03 AM
To: [log in to unmask]
Subject: Re: [HP3000-L] Memory usage on HP3000
-----Original Message-----
From: HP-3000 Systems Discussion [mailto:[log in to unmask]] On
Behalf Of Tony Summers
Sent: Tuesday, February 14, 2006 12:00 AM
To: [log in to unmask]
Subject: Memory usage on HP3000
Can anybody point me to papers on how we review whether our HP3000s are
suffering from memory shortages.
----
Some information is at
http://jazz.external.hp.com/mpeha/papers/off_white_2004.html
MPE performance strikes me as almost the inverse of a traditional
virtual storage system where plentiful I/O is substituted for
scarce memory.
I remember from somewhere a rule-of-thumb like:
64M/cpu + 16M (XM) + 8M/batch job + 4M/session
to get a minimum memory configuration.
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
|
|