HP3000-L Archives

January 2006, Week 1

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:
"Reynolds, James" <[log in to unmask]>
Reply To:
Reynolds, James
Date:
Thu, 5 Jan 2006 12:41:50 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (155 lines)
And to sacrifice the small animals! 

-----Original Message-----
From: HP-3000 Systems Discussion [mailto:[log in to unmask]] On
Behalf Of Johnson, Tracy
Sent: Thursday, January 05, 2006 12:40 PM
To: [log in to unmask]
Subject: Re: [HP3000-L]


Don't forget your Witch-doctor mask and rattle.

Tracy Johnson
Measurement Specialties, Inc. 

BT







NNNN

> -----Original Message-----
> From: HP-3000 Systems Discussion
> [mailto:[log in to unmask]] On Behalf Of David N Lukenbill
> Sent: Thursday, January 05, 2006 12:33 PM
> To: [log in to unmask]
> Subject: Re: [HP3000-L]
> 
> 
> Kent,
> 
> Tuning an HP-3000 is as much art as science.  A lot of it
> depends upon the characteristics of the jobs/sessions running 
> on the system.  If you have Glance or SOS, take a look at the 
> processes running to see if you can get an idea on how the 
> CPU(s) are being used.
> 
> On my N-4000-200-500, I'm using the following:
> 
> showq;status
> 
>                     ------QUANTUM-------
> QUEUE  BASE  LIMIT  MIN    MAX    ACTUAL  BOOST  TIMESLICE
> -----  ----  -----  ---    ---    ------  -----  ---------
>  CQ    152    204   1      2000   1       OSC      200
>  DQ    202    242   2000   2000   2000    OSC      200
>  EQ    240    253   2000   2000   2000    OSC      200
> 
> I think by using a CQ timeslice of 2000, you may be letting a
> single process grab the CPU for too long.  I also prefer 
> setting the MAX QUANTUM to a larger value so processes decay 
> in priority a little slower. Sometimes processes keep 
> "jumping" to the top of the CQ.  I prefer to have a lower 
> timeslice so they don't dominate the queue.  I also overlap 
> the queues a little so I don't completely starve the job queues.
> 
> David N. Lukenbill
> Computer Sciences Corporation
> 
> 
> 
> 
> |---------+---------------------------->
> |         |           Kent Wallace     |
> |         |           <kwallace        |
> |         |           @primaryhealth.co|
> |         |           m>               |
> |         |           Sent by: HP-3000 |
> |         |           Systems          |
> |         |           Discussion       |
> |         |           <[log in to unmask]
> |         |           TC.EDU>          |
> |         |                            |
> |         |                            |
> |         |           2006/01/05 12:08 |
> |         |           PM               |
> |         |           Please respond to|
> |         |           Kent Wallace     |
> |---------+---------------------------->
>   
> >-------------------------------------------------------------
> -----------------------------------------------------------------|
>   |                                                           
>                                                                    |
>   |       To:       [log in to unmask]                    
>                                                                    |
>   |       cc:                                                 
>                                                                    |
>   |       Subject:  [HP3000-L]                                
>                                                                    |
>   
> >-------------------------------------------------------------
> -----------------------------------------------------------------|
> 
> 
> 
> 
> Listers
> 
> 
> 
> When we run large LINK's in SUPRTOOL none of the session
> users seem to get resources.  Does anyone know how to get the 
> session users reasonable resources while LINK is running?
> 
> 
> 
> Here is a SHOWQ on my N-Class HP3000.
> 
> 
> 
>                     ------QUANTUM-------
> 
> QUEUE  BASE  LIMIT  MIN    MAX    ACTUAL  BOOST  TIMESLICE
> 
> -----  ----  -----  ---    ---    ------  -----  ---------
> 
>  CQ    152    200   1      200    11      DECAY    2000
> 
>  DQ    202    238   2000   2000   2000    DECAY    200
> 
>  EQ    248    253   2000   2000   2000    DECAY    100
> 
> 
> 
> 
> 
> Thanks
> 
> Kent Wallace
> 
> 
> 
> 
> 
> 
> 
> 
> * 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 *
> 

* 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 *

ATOM RSS1 RSS2