[torquedev] [torqueusers] Problem with ppn and routing

"Mgr. Šimon Tóth" SimonT at mail.muni.cz
Mon Dec 6 16:27:55 MST 2010


> Can we avoid the "should the server be able to schedule" debate for a
> minute?

Uh? The server does schedule. The only thing is that it doesn't schedule
according to available resources only ppn and properties.

And even if I ignore scheduling the server is the only place where
resource requests can be checked effectively. The scheduler can do this,
but there can be no concurrent run requests (or other schedulers
connected to the server), and using nodes to check for resource request
violations is extremely ineffective.

> I dont think the first of these has any meaning right now - I think the
> ":mem=100GB" is treated as a node property.   But I like the idea, it
> would allow us to have non-uniform memory requests across nodes. It seems
> to me that -lnodes=2:mem=100MB should be equivalent to -lmem=200MB,nodes=2.
> Possibly the first should have a syntax like -lnodes=2:mempn=100MB to make
> it less ambiguous.

It has no meaning inside torque (apart from mem=100GB being treated as a
property), but it is well supported in PBSPro (and should also be
supported in external schedulers).

There is pmem for memory per process, therefore mempn would be very
confusing.

-- 
Mgr. Šimon Tóth

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3366 bytes
Desc: S/MIME Cryptographic Signature
Url : http://www.supercluster.org/pipermail/torquedev/attachments/20101207/4fd79bbc/attachment-0001.bin 


More information about the torquedev mailing list