HP3000-L Archives

November 1995, 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:
Bruce Toback <[log in to unmask]>
Reply To:
Bruce Toback <[log in to unmask]>
Date:
Wed, 8 Nov 1995 12:43:33 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (30 lines)
Clint Schwartz writes:
>  1.  We changed our "queues" so that the job "queue" and the session "queue"
>did not overlap because of some known problem with sessions that wait on
>message files.  Evidently if a session drops down to the bottom of its
>"queue," it doesn't get bumped back up to the top of its queue like 4.0 did.
 
In dealing with a similar problem, our interrim solution was to shorten the
CS queue and set it to "oscillate", making the scheduling algorithm more
round-robin than priority queue. This prevented a lot of short transactions
from monopolizing the CPU at the expense of those processes that should
have been bumped up but weren't any longer.
 
   TUNE;CQ=152,160,100,1000,oscillate
 
Of course if you DO have a real CPU hog in the queue, it'll get a lot more
of the CPU than it would with the default parameters. But then you
shouldn't be running CPU hogs in the C queue in any case.
 
Our final solution was to make the application run faster, but you may not
have that option :-).
 
-- Bruce
 
---------------------------------------------------------------------------
Bruce Toback    Tel: (602) 996-8601| My candle burns at both ends;
OPT, Inc.            (800) 858-4507| It will not last the night;
11801 N. Tatum Blvd. Ste. 142      | But ah, my foes, and oh, my friends -
Phoenix AZ 85028                   | It gives a lovely light.
[log in to unmask]                 |     -- Edna St. Vincent Millay

ATOM RSS1 RSS2