Mailing List Archives
Authenticated access
|
|
|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [HTCondor-users] request_cpus and slot weight
- Date: Mon, 01 Dec 2014 15:02:59 -0600
- From: Todd Tannenbaum <tannenba@xxxxxxxxxxx>
- Subject: Re: [HTCondor-users] request_cpus and slot weight
On 12/1/2014 10:07 AM, Uri Schneider wrote:
Hello,
I am trying to have partitionable dynamic slots (by CPU), and also have
each job count as its requested CPU count, for quota purposes.
The default is for HTCondor to use CPUs for its slot weight.
In v8.2.x, the relevant defaults are:
NEGOTIATOR_USE_SLOT_WEIGHTS = true
NEGOTIATOR_USE_WEIGHTED_DEMAND = true
SLOT_WEIGHT = Cpus
Perhaps you modified one of the defaults in your condor_config file(s) ?
So I tested your setup below on my laptop (running HTCondor v8.2.4)
which gives me a pool with one partitionable slot containing four CPUs
(since my laptop has four cores). I then submitted some jobs that have
request_cpus = 2
in the job submit file and things worked as expected according to
condor_status:
>condor_status -af:h Name State SlotType Cpus SlotWeight
Name State SlotType Cpus SlotWeight
slot1@ToddsThinkpad Unclaimed Partitionable 0 0
slot1_1@ToddsThinkpad Claimed Dynamic 2 2
slot1_2@ToddsThinkpad Claimed Dynamic 2 2
Note the SlotWeight in the newly created dynamic slots does equal the
number of Cpus in that dynamic slot (which in turn equals the number of
cpus requested in the job).
regards,
Todd
The relevant condor_config part is:
NUM_SLOTS = 1
NUM_SLOTS_TYPE_1 = 1
SLOT_TYPE_1 = cpus=100%
SLOT_TYPE_1_PARTITIONABLE = true
SLOT_TYPE_1_SLOT_WEIGHT = Cpus
My jobs' submit file includes:
request_cpus = 2
However, when such a job matches to a slot, I get the following, when
trying to figure the slot's weight:
# condor_status –long | grep Weight
SlotWeight = 1
SlotWeight = 1
(… and so on, one such line for each job matched)
# condor_q –long | grep RequestCpus
ReqeustCpus = 2
ReqeustCpus = 2
(… and so on, one such line for each job matched)
I would like each job's slot to have a weight of 2, as it requested 2 CPUs.
Any clue as to what I am doing wrong?
Thanks in advance,
Uri Schneider
_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users
The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/
--
Todd Tannenbaum <tannenba@xxxxxxxxxxx> University of Wisconsin-Madison
Center for High Throughput Computing Department of Computer Sciences
HTCondor Technical Lead 1210 W. Dayton St. Rm #4257
Phone: (608) 263-7132 Madison, WI 53706-1685