HP3000-L Archives

January 1996, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Steve Dirickson b894 WestWin <[log in to unmask]>
Reply To:
Steve Dirickson b894 WestWin <[log in to unmask]>
Date:
Wed, 10 Jan 1996 18:29:00 P
Content-Type:
text/plain
Parts/Attachments:
text/plain (25 lines)
<<Now that we have covered many of the syntactical issues about the
Multiple Job Queues, I thought that I would ask the list what they
thought about the idea of being able to segregate a particular processor
in a multiple processor environment to a particular queue/workgroup?>>
 
Trying to micro-manage CPU usage at the chip level sounds like a losing
proposition to me: it seems extremely unlikely that any individual is
going to be as successful at scheduling jobs for execution as the OS, so
the net result is almost guaranteed to be an overall loss of throughput.
Selected jobs/queues/whatever would benefit from having a dedicated CPU,
but the system-wide effect would be to exact a disproportionately large
CPU-usage tax from everyone else to support these chosen few.
 
I'd rather just tell the OS things like
 "These jobs are important; give them prority"
 "These jobs are not important; run more important stuff first"
 "These users routinely submit large CPU-hog jobs that don't really need
to run at high priority; watch them, and lower the priority of their jobs
if it's too high"
 
and have the OS figure out how to make it happen.
 
Steve Dirickson         WestWin Consulting
(360) 598-6111  [log in to unmask]

ATOM RSS1 RSS2