[torqueusers] Need to partition or allocate nodes to queues?
aquarijen at gmail.com
Mon Dec 22 01:58:08 MST 2008
I am not sure if this is a torque or Maui question, so please forgive the
I used to do this blissfully boringly using acl_hosts. Something has
changed and this is not working anymore between torque 2.1.x and now. I
have the following:
I have a large cluster with several departments and these departments want
to use the nodes they purchased. In the past, I set up a queue for each of
these departments and I assigned nodes to the queues in torque. Has the way
to do this changed? I have seen several messages lately from folks no
longer able to get it to work...
I am not married to this method and will happily implement any method that
will keep my (hundreds) of users separated and on their own nodes. I do not
want them to have to change their scripts. These are heterogeneous, too, so
it is not just a policy thing - the infiniband jobs won't run on the non IB
nodes, and the some are 8proc, some are 2, etc.
I have tried the "neednodes" method. It did nothing.
I've been trying to do partitions in Maui and that seems to be ignored as
well. This is a 64 bit system, by the way.
Can anyone point me to modern documentation on how to partition the nodes?
I tried defining attributes in "nodes" under server_priv:
node74 np=2 noib
node75 np=2 noib
node76 np=2 altix
node77 np=8 altix
node78 np=8 hiio
And then I have in my torque config, something like:
# Create and define esmg08q
create queue esmg08q
set queue esmg08q queue_type = Execution
set queue esmg08q resources_max.cput = 10000:00:00
set queue esmg08q resources_max.ncpus = 179
set queue esmg08q resources_max.nodect = 22
set queue esmg08q resources_max.walltime = 10000:00:00
set queue esmg08q resources_min.cput = 00:00:01
set queue esmg08q resources_min.ncpus = 1
set queue esmg08q resources_min.nodect = 1
set queue esmg08q resources_default.neednodes=altix
set queue esmg08q resources_min.walltime = 00:00:01
set queue esmg08q resources_default.cput = 10000:00:00
set queue esmg08q resources_default.ncpus = 8
set queue esmg08q resources_default.walltime = 100:00:00
set queue esmg08q acl_logic_or = true
set queue esmg08q acl_groups = esmg
set queue esmg08q acl_groups += admin
set queue esmg08q acl_group_sloppy = true
set queue esmg08q acl_group_enable = true
set queue esmg08q enabled = True
set queue esmg08q started = True
- From this I see in maui a line in showconfig:
maui]# /usr/local/maui/bin/showconfig | grep esmg
So I was hopeful! But it ignores it and doesn't assign the nodes with the
altix attribute? No matter what I do, I always get the first nodes and in
order. Like it is reading from a machine file that I don't know about... I
do get 8 proc nodes if I ask for nodes with 8 processors, I do get whatever
node I ask for in my submit script. I literally lets me request any node I
want, no matter which user I am or which queue I submit to. Does anyone
have any ideas?
Thanks all for your help.
Really Sleepy Admin, ORNL Institutional Clusters
Use it up; wear it out; make it do or do without! - L Reid
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the torqueusers