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:
John Joerger <[log in to unmask]>
Reply To:
John Joerger <[log in to unmask]>
Date:
Fri, 12 Jan 1996 15:03:26 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (132 lines)
On Fri, 12 Jan 1996, Mohan Das wrote:
> 1. We seem to have  almost a  consensus  on one thing.  A default job queue
>    based on some  membership  mechanism (it is not yet clear what it should
>    be; I'll come to that later) AND the ability to  override  this  default
>    through job initiation commands, both are needed.
>
>    The heirarchy will be:
>
>    1. ;JOBQ= specified in :STREAM command, if any
>    2. else ;JOBQ= specified in !JOB command, if any
>    3. else a default job Q based on some membership criteria, if any specified
>    4. else the system default Q, $SYSJOBQ.
>
>    Queue to which a job should belong to is decided by applying the above 4
>    criteria,  in that order.  (Some  people might argue for swapping 1 & 2,
>    though.  We need to decide on that).
 
I think this acurately reflects my desires.  I like the listed priorities
of 1 & 2 but could easily adapt to switching these two items.
 
> 2. There are two opinions on the mechanism for  determining the default job
>    queue (in cases when no job queue is specified in either !JOB or :STREAM
>    command).
>
>    1. Default job queue based on user.acct
>
>       Not sure whether it should be user.acct specified in !JOB or
>       that of the user who :STREAMed the job.
>
>    2. Some respondents  feel they need more  granularity in determining the
>       default  queue and they are of the opinion that it should be based on
>       "jobname,user.acct,group".
>
>    While option 1 can be achieved by adding a new parameter ;QUEUE= to
>    :NEWUSER and :ALTUSER commands, if we choose option 2, we need to
>    either have a file to decide membership or it should be part of
>    commands which create and alter job queues. Like,
>
>        :NEWJOBQ queuename;members="payroll,mohan.admin",@.sys
>        :ALTJOBQ usersq;members=+newbie.users,-oldie.users
>        etc.
>
>    We need to  decide if the extra  control  gained  by  option 2 is really
>    needed,  even  when  there is an option  of  explicitly  specifying  the
>    queuename in the job cards itself.
 
ABSOLUTELY!  There are many shops that run 3rd party software.  There is
no way that these vendors can assume queue management configuration for
all their customers.  Many shops don't have the resources to make the
!JOB card changes each time we get a new release.  In the case of our
apps, the :STREAM command is buried in an online driver program.  That
means that jobs are streamed via the application without presenting the
:STREAM command to the user.
 
> 3. Some people have asked for  properties  like jobfence and input priority
>    (default and  maximum) to be different  for each queue  instead of being
>    same across all the queues.  We would really appreciate feedback on this
>    matter.  Please note that our  objective is to meet basic needs  without
>    incurring too much cost, so we wouldn't want to try to provide  anything
>    which is not really essential.
 
Please refer to a previous post for my example.  Jobfence carries the
highest priority in my book.  The other properties may have value, but
seem to be more of a convenience.  But that's just from my chair.
 
> 4. Should  there be any  relative  priority  across  the queues in terms of
>    selcting the next job to run ?  Consider this scenario:
>
>    JOBFENCE=8, JLIMIT=10; 10 EXEC; 2 WAIT
>
>    JOBQ1: LIMIT = 5, 1 WAIT (#J11, INPRI=12)
>    JOBQ2: LIMIT = 3, 1 WAIT (#J12, INPRI=13)
>
>    Assume that #J12 was streamed before #J11.
>
>    Now one of the running jobs gets over and one slot is open.
>
>    Which job should be picked to run ?
>
>    1. Should  it be based on some  pre-defined  priority  between  queues ?
>       Say, jobs waiting in JOBQ1 have a higher  priority  over jobs waiting
>       in JOBQ2, irrespective of INPRI and order of streaming.
>
>       In that case, #J11 will be picked to run.
>
>    2. Should it be based on INPRI and temporal  order across all the queues?
>       (Within a queue, the jobs will be picked  based on INPRI and order
>       of streaming, of course)
>
>       In that case, #J12 will be picked to run.
 
I like option 2.  It looks less complicated than introducing queue
priorities and the extra management control that is inferred.  Queue
priorities and priorities of workgroups inside queues is more on the
side of full featured operations center application (IMHO).
 
> 5. We need feedback from  customers  who are using  Workload  Manager as to
>    what kind of  integration,  if any,  would  they want  between  Workload
>    Manager and Multiple Queues ?
 
Sorry, can't help here.
 
> 6. Do the majority of system  administrators  on this list want the ability
>    to *restrict* users to a queue or set of queues ?  That is, now that the
>    users  have the  option of  specifying  any  queue  they want in the job
>    command, do you want to have the ability to say, "No, you're not allowed
>    in this Queue"  through ACD or any other  mechanism  ?  I haven't  heard
>    anybody saying so far, that they really need this security feature.
 
I can certainly see how this would be very useful.  However, with the
inclusion of job/session name and logon group, our environment would
receive little benefit.
 
> 7. Is the ability to transfer jobs from one queue to another while the
>    job is waiting needed ? Like,
>
>    :ALTJOB #J12;QUEUE=JOBQ1
>
>    Some of you have mentioned it, but I just want to know if it is
>    really essential.
 
Yes!  The Computer Operator MUST be able to keep control of the
operating environment.  Some may argue that moving jobs in the
EXEC state isn't of value.  I disagree.  It allows the Operator
to easily "free up" a queue so the waiting jobs can continue.
The only other alternative would require each waiting job in the
blocked queue to be moved.
 
Regards,
 
--- John Joerger  ([log in to unmask])  ---  Press-Telegram (Long Beach)

ATOM RSS1 RSS2