On Wed, 11 Aug 2004, Bogdan Costescu wrote:
> I don't know if you are talking about the maximum number of slots that
> can be allocated by SGE on the node or about the number of slots
> allocated for the job. Even if you have a maximum of 4 slots for a
> node, SGE might decide to allocate only one from this node because of
> it allocation policies (based on load, for example).
> To find out if this is the case, in the batch script before running
> lamboot add a line like:
>
> cat $pe_hostfile
>
> Then look in the .o file for the output. The second column specifies
> the number of slots allowed on each node.
The scheduler tends to take as many nodes as possible, if I specify 2
processors during the PE submission. ie, %qsub -pe lam 2 stuff.sh, it will
try to take 2 nodes with 1 slot each node if possible. In such case, the
number of allocated slots for the job is only 1.
So did you say the problem comes from here?
I have also tried specifying more required-processors than the number of
nodes, so that some nodes will have more than 1 slot allocated, but the
result is the same.
I am not sure how it causes the problem. Is it like the startup of lamd is
treated as part of the job and requires an extra slot for this startup?
So that the required-slots is allocated-slots + 1 ?
Regards,
Alan.
>
> --
> Bogdan Costescu
>
> IWR - Interdisziplinaeres Zentrum fuer Wissenschaftliches Rechnen
> Universitaet Heidelberg, INF 368, D-69120 Heidelberg, GERMANY
> Telephone: +49 6221 54 8869, Telefax: +49 6221 54 8868
> E-mail: Bogdan.Costescu_at_[hidden]
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
|
