Michael,

This is how our system is tuned:

                    ------QUANTUM-------
QUEUE  BASE  LIMIT  MIN    MAX    ACTUAL  BOOST 
TIMESLICE
-----  ----  -----  ---    ---    ------  ----- 
---------
 CQ    152    185   1      250    18      DECAY    500
 DQ    180    240   1000   1000   1000    DECAY   
1000
 EQ    241    255   1000   1000   1000    DECAY   
1000

There are a limited number of DQ jobs run during the
day and users haven't complained (at least about
this). And I do not change it for the batch jobs
during the night, seems to work fine. YMMV.

John Bawden
Wellpoint
--- Michael Caplin <[log in to unmask]> wrote:

> I've read thru some of the posted messages regarding
> this but I could use 
> some help.  Our N class box used to be dedicated at
> night for batch jobs 
> but we're now allowing session users.  The session
> users don't need too 
> much but it looks like their presence is degrading
> the jobs running in the 
> batch queue.  Last night (month-end) was a real joy.
>  
> 
> I'm planning to schedule a job to run at approx
> 10:00 pm that will alter 
> the DS queue and then put it back to where it was
> via a second scheuled job 
> at 5:00 am.
> 
> This is the 10:00 pm job.  Any suggestions ?
> 
> Mike
> 
> !COMMENT * THESE ARE THE DEFAULT VALUES
> !COMMENT
> !COMMENT            ------QUANTUM-------
> !COMMENT QUEUE BASE LIMIT MIN  MAX         BOOST 
> TIMESLICE
> !COMMENT ----- ---- ----- ---  ---         ----- 
> ---------
> !COMMENT  CQ   152   200  1    2000        DECAY   
> 200
> !COMMENT  DQ   202   238  2000 2000        DECAY   
> 200
> !COMMENT  EQ   240   253  2000 2000        DECAY   
> 200    
> !
> !TUNE DQ=152,200,1
> !EOJ
> 
> * To join/leave the list, search archives, change
> list settings, *
> * etc., please visit
> http://raven.utc.edu/archives/hp3000-l.html *
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *