(say).
Is this possible with SLURM?
I could turn off one partition with a cron job, but that would allow jobs
to schedule at 7:59am and run into the closed window.
Any other thoughts on how this might be accomplished?
Thanks in advance,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@
l, but it works for us for now.
Any thoughts on the matter would be helpful.
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
Yes, that's going to be a big problem for us to. It's good that 2.6 is
working well for us.
Any chance we can get the old behavior back?
Best
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512
Thanks. Very much appreciated.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 5/21/14 1:49 PM, "David Bigagli" wrote:
>
>
>By popular acclaim we keep supporting th
the
reservation, and if I stop SLURM on the node (/etc/init.d/slurm stop) it
doesn't get replaced either. I would have sworn up and down that at least
the latter worked.
Can anyone provide some feedback?
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Pho
No thoughts on this from the list? I wouldn't have thought we were the
only ones encountering this issue.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 6/5/14 3:
this summer.
git clone g...@github.com:rtevans/tacc_stats.git (this will eventually move
the the main TACC GitHub, but that's a work in progress)
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-
load code, so I think you'd be hard pressed
to schedule around the situation where one user wants to run native and
another says they're running host-only but actually runs an offload code.
Maybe you have something else in mind?
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.ut
Why not make one partition and use fairshare to balance the usage over
time? That way both institutes can run large jobs that span the whole
machine when others are not using it.
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435
on
in the ratio of their contribution, but is allowed to use the whole
machine. If both institutes have periods of down time, then the machine
will be less likely to sit idle and more work will get done.
I'll get off my soapbox now.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.
t just goes into the
bit bucket.
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 9/10/14, 8:39 PM, "Christopher Samuel" wrote:
>
>On 10/09/14 03:10, Erica Riello wr
probably use cp. I haven't tested this, but you should probably escape the
$ and use cp:
srun cp /scratch/Hello /scratch/\$HOSTNAME
might work.
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475
At TACC we solve this problem by disabling HyperThreads/SMT/whatever your
CPU vendor calls it. Up to Intel SNB we haven't additional hardware
threads to be useful for most of our codes.
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office
John,
We're doing some work on tracking what codes get used at TACC which I'd be
happy to share on-list or off-list if advertising is here is frowned upon.
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435
That's what I sent John off-list. Wasn't sure self-promotion was OK here.
You could add an srun wrapper (called srun) to XALT and call it a day.
That's how it supports everything else!
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7
up for the
slack caused by the bad one. One can also make the reservation larger by a
few nodes to account for bad luck.
I'm really wondering if there are any better options or any automated
options.
What do others do?
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@
No such luck, Aaron. I specified a node count in my original
specification. I would have thought what you suggested was the behavior,
but it appears not to be.
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435
I don't know about it's use for cloud services, but our supercomputers at
TACC that run SLURM are made entirely of commodity parts. In most cases
these days, there's very little difference between commodity clusters and
supercomputers.
Best,
Bill.
--
Bill Barth, Ph.D., Di
ition priorities override job priorities completely.
This is consistent with our experience since 7/2 when we upgraded. Was
this a recent change? Is there anything we can do to go back to the old
behavior?
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232
nses=(null) State=INACTIVE
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
nt=36 CoreCnt=576 Features=(null)
PartitionName=normal-mic Flags=
Users=bbarth,foo,bar,baz Accounts=(null
) Licenses=(null) State=INACTIVE
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On
efore the upgrade. I wonder if something went weird
during the upgrade process.
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 6/24/15, 12:00 PM, "Jacqueline Scoggins"
lap in the future due to recurrence? It
doesn't seem to to me.
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 6/24/15, 1:35 PM, "Bill Barth" wrote:
>
was going to solve my problem, but no luck. Any
advice would be much appreciated.
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
#x27;t optimal.
Any thoughts on preventing this?
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 7/6/15, 11:19 AM, "John Desantis" wrote:
>
>Bill,
>
&
h wrong with the upgrade. No luck.
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 7/7/15, 7:31 AM, "John Desantis" wrote:
>
>Are you using SlurmDBD or the MySQL accounting plugin? Since you've
>re-created all reservations and there is still unintentional
>overlapping occurring, I'd recommend looking at the reservation
>table(s) within the DB and possibly truncating it(the
On 7/7/15, 4:40 PM, "Bruce Roberts" wrote:
>
>The reservations in the database are only for historical purposes, they
>don't get read in from the slurmctld. The DBD should be purging them as
>configured you shouldn't have the manually do anything.
I would have thought so. Any thoughts on my s
launch independent serial tasks in parallel.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 7/14/15, 11:17 AM, "John Desantis" wrote:
>
>Charles,
>
>Have y
nyone else seen this?
Thanks,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
Thanks, Lyn. That's very interesting. I'm glad that this is reproducible.
I was driving myself crazy over here.
Thanks again,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
e had more than 10k nodes under management across a handful of clusters
at once with this system.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 3/15/16, 7:39 AM, "Bjørn-H
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435 | Fax: (512) 475-9445
On 4/8/16, 2:44 PM, "Craig Yoshioka" wrote:
>Personally, I don¹t enjoy the overhead of creating batch files, so I use
>srun
I was thinking of MPI codes primarily since that kind of work is my
background. If you are running a serial/threaded program on one node, then
you don't need srun, just set OMP_NUM_THREADS and go!
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232
modules, and libraries
used by a job.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435| Fax: (512) 475-9445
On 3/22/17, 10:29 AM, "E.M. Dragowsky" wrote:
Greetings,
.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435| Fax: (512) 475-9445
On 3/30/17, 10:52 AM, "Chad Cropper" wrote:
We would like to clean our memory buffers/cache on a regular bas
mounted. It took a little while to figure out
where in the PAM stack to insert the pam_systemd.so configuration line to
guarantee that it was working for all our SLURM jobs, but the above method
seems to solve the problem.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu
Fortunately, once we figured out what systemd was doing, we didn’t need to
interact with it besides adding its PAM module configuration line to slurm’s
PAM config file.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435
We don’t use cgroups with our SLURM at this time, though we have some ongoing
investigations in that direction. There’s probably a way to get both plugins to
cooperate.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435
way. We also do
*everything else* via our own RPMs, so it fits with our local methodology.
Best,
Bill.
--
Bill Barth, Ph.D., Director, HPC
bba...@tacc.utexas.edu| Phone: (512) 232-7069
Office: ROC 1.435| Fax: (512) 475-9445
On 10/5/17, 2:07 AM, "Ole Holm Ni
40 matches
Mail list logo