[gridengine users] "Packing" jobs on nodes v2
iqtcub at gmail.com
Mon May 21 06:00:32 UTC 2012
On 05/15/2012 03:02 PM, Mark Dixon wrote:
> On Tue, 15 May 2012, iqtcub wrote:
>>> I don't know if I've missed anything myself but, in the past, I've
>>> found that this (queue_sort_method seqno, job_load_adjustments NONE,
>>> etc.) doesn't always work with PEs unless you specify a different
>>> seq_no for each queue instance. You might want to give that a go.
>> Regarding the option you mentioned, we want to do it using only one
>> queue, which is what we have at the moment(they don't like to limit the
>> runtime of the jobs).
> Yes, one queue - but each host in that queue has what is termed a
> queue instance.
> (e.g. v20z.q = queue, v20z.q at hostname = queue instance).
> You can assign them different seq_no's with something like:
> qconf -mattr queue seq_no 100 v20z.q at host1
> qconf -mattr queue seq_no 101 v20z.q at host2
> Take a look at the output of "qconf -sq v20z.q" to see what effect
> this had.
> PS v20z? I liked them, but for some reason I always ended up bleeding
> slightly from my right hand index finger whenever I took the lids off.
Thanks for the explanation, i read about these options in older mails
but didn't seem to find how to do this exactly.
I've tested it and it works the way we want.
More information about the users