[GE users] SGE-6.0: How to achieve increasing and decreasing seq_no?

Ralf Sigmund sigmund at ipk-gatersleben.de
Wed Oct 27 15:14:00 BST 2004

    [ The following text is in the "ISO-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Reuti wrote:

> Hi,
> Erik Soyez wrote:
>> Good Day,
>> with SGE-5.3 I used to configure clusters with e.g. increasing seq_no 
>> for
>> a "normal" queue and decreasing seq_no for an "express" queue in 
>> order to
>> keep parallel jobs stick together and keep express and normal jobs as 
>> far
>> away from each other as long as possible.  That was pretty easy with a
>> script that just added up seq_no when modifying queues.
> there will hardly be a difference, where the jobs are running in the 
> cluster. If you want a cleaner output with qstat, you could use a 
> script to make a qstat two times, and select only each type of queue.
> Are you using dual CPU machines? In this case having this setup would 
> fill one machine after the other. Because there is always an overhead 
> running 2 jobs on one machine, it will a little bit faster to fill 
> first one slot on each machine.

Hello Reuti,
I am exactly trying to do achieve this.  How can use only one processor 
as long as there are idle nodes left?

is this done with slots (do I need 2 queues for this)?

---- from my queue configuration ----
processors            UNDEFINED
qtype                 BATCH INTERACTIVE
ckpt_list             NONE
pe_list               make
rerun                 FALSE
slots                 1,[compute-0-1.local=2],[compute-0-2.local=2], \
                      [compute-0-6.local=2],[compute-0-8.local=2], \
                      [compute-0-4.local=2],[compute-0-3.local=2], \

Cheers - Ralf

To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net

More information about the gridengine-users mailing list