Marty's post on this thread reminds me a posting Lynn Wheeler put onto
something 10-20 years ago when he was talking about the advances that
had been made in CPU speed in relation to the relatively slow dasd speed
advances by then. I've been totally surprised in watching DASD response
time lately however. We have a SHARK, which by now if kind of out of
date, I guess. Nevertheless, Performance Toolkit reports Dasd response
time in the 1 ms range vs. the 30 ms. time I was used to in the spinning
Dasd days
Jim
Marty Zimelis wrote:
Phil,
Unless there was something else out there (a poster or whatever), that
would have been me doing a riff in my VM Performance classes, first for
Amdahl, then for Velocity. Your buddy's time frame is about right (15 years
ago). I was attempting to emphasize the impact of an RPS miss (show of
hands: who remembers what that was?) on response time.
The riff started by me "complaining" that I didn't have a good intuitive
grasp of how fast CPUs were (tens of nanosecond cycle times at that point),
so "let's slow down our timeframe and say a CPU cycle is one second. Then a
page fault from Xstore is satisfied in [nn minutes], a DASD I/O satisfied
from cache takes [mm hours] and one that has to go to the real disk takes
[kk days]. An RPS miss adds [I think it was 16 hours] to that."
Alas, a quick search of my notes didn't turn up a copy of the discussion,
so I can't fill in the blanks. Perhaps someone who took one of the classes
and wrote it down....
Marty
____________________
Martin Zimelis
Principal
maz/Consultancy
-----Original Message-----
From: The IBM z/VM Operating System
[mailto:[EMAIL PROTECTED] On Behalf Of Phil Smith III
Sent: Thursday, November 08, 2007 1:48 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Poster of computer hardware events?
A buddy asked me:
"At a previous employer, someone had an article, poster or
something (I know - real specific - it was 15+ years ago)
that tried to put the time for computer events into
perspective. It started with the quickest instruction (RR)
having a baseline of 1 second. It the proceeded to go through
all of the instructions, RX, RS, SS etc. and then into I/O,
MIH and so on. Have you ever heard or seen anything like
this? I'm having trouble stressing the importance of poor I/O
response time and I thought this might be of use."
I had to tell him I hadn't ever seen such a thing, but would
like to. I figure if anyone else alive knows what this
is/was, they'll be on one of these two lists...!
Anyone?
--
...phsiii
Phil Smith III
Velocity Software
www.velocity-software.com
(703) 476-4511 (home office)
(703) 568-6662 (cell)
--
Jim Bohnsack
Cornell University
(607) 255-1760
[EMAIL PROTECTED]
