> #$ -l nodes=2,ppn=12 (--> here: 12 slots on 2 nodes = 24 in sum) .
>
> when you restrict "nodes=1" (with resource quotas "qconf -mrqs" for example)
> then nobody > should be able to use more than 1 node.
sorry, correction: "to use more than 1 node PER JOB" of course.
___
Hi Ansger,
do you have experiences with Torque/PBS?
There is a useful gridengine plugin which works quite similar:
https://github.com/brlindblom/gepetools
I have it in use on one of my clusters. The advantage: users must order not
just slots but also nodes in the form:
#$ -l nodes=2,ppn=12 (
Hello Arnau,
As far as I know, there are no Centos 7 packages – but you can install the
CentOS 6 rpm packages without problems.
Regards,
Ursula
Von: users-boun...@gridengine.org [mailto:users-boun...@gridengine.org] Im
Auftrag von Arnau
Gesendet: Freitag, 01. Juli 2016 15:29
An: Gridengine Use
>> Am 16.11.2015 um 14:14 schrieb Winkler, Ursula (ursula.wink...@uni-graz.at)
>> :
>>
>>>> Am 16.11.2015 um 11:41 schrieb Winkler, Ursula
>>>> (ursula.wink...@uni-graz.at) :
>>>>
>>>> Dear gridengine members,
>>
>> Am 16.11.2015 um 11:41 schrieb Winkler, Ursula (ursula.wink...@uni-graz.at)
>> :
>>
>> Dear gridengine members,
>>
>> when there are two queues and one is a subordinate of the other, then
>> the release/shift of slots to the superordinate jobs
Dear gridengine members,
when there are two queues and one is a subordinate of the other, then the
release/shift
of slots to the superordinate jobs is no problem, but (other) consumeable
resources
like h_vmem are. Does somebody have an idea to bypass that?
Thanks,
Ursula
___
> So it sounds like the $pe_slots=serial behavior isn't true for your grid
> engine.
> An alternative would be to convert single node jobs to serial jobs that
> requests
> a consumable(other than slots) representing the number of cores on the node.
>
> However this will change the behavior
>> complex_valuesoccupied_slots=12,slots=12,num_proc=12,h_vmem=45.01G, \
> > exclusive=true
>
> To streamline this (but it won't solve your problem): I usually recommend not
> to set num_proc,
> as it's a feature of the host which is detected by SGE already. Limitatio
> Have you got slots defined per host or just per queue?
> It may need to be defined at a host level via complex_values
> in order to work.
Both:
# qconf -sc | grep "#name\|slots"
#name shortcut typerelop requestable consumable
default urgency
occupied_slots o_s
> What happens with serial jobs (ie no PE at all)? If they exhibit the same
> behavior then we
> just need to figure out how to tweak the scheduler config to fill up nodes.
> If serial jobs are
> already clumping onto a few nodes then the claim that $pe_slots behaves like
> serial on the
>
>> Well, I created an additional PE with alloacation_rule "$pe_slots",
>> and built in an if condition in "pe.jsv" for all jobs which request
>> just a single node to be assigned to this new PE. But the annoying
>> situation didn't change. The scheduler configuration is set to
>> "queue_sort_met
> Sorry to step in the discussion: `qstat -j ...` shows the requested one, the
> granted one is in `qstat -r`.
>
> $ qsub -pe "*" 2 test.sh
> Your job 44329 ("test.sh") has been submitted $ qstat -j 44329 ...
> parallel environment: * range: 2
> ...
My jobs:
qstat -j ...
...
Am 30.07.2015 um 18:29 schrieb "Reuti" :
>
>> Am 30.07.2015 um 18:14 schrieb Winkler, Ursula (ursula.wink...@uni-graz.at)
>> :
>>
>>
>>
>>> On Thu, 30 Jul 2015 12:57:13 +0000
>>> "Winkler, Ursula (ursula.wink...@uni-gra
> On Thu, 30 Jul 2015 12:57:13 +
> "Winkler, Ursula (ursula.wink...@uni-graz.at)"
> wrote:
>
>>> My suggestion was to modify your jsv/gepetools to force single node
>>> parallel jobs into PEs with $pe_slots allocation rules (which gives
>>&
> I believe it should be a load_formula of -slots so the more slots are
> available(fewest used) the lower the load and the more attractive the
> node. The page Reuti pointed to manages to write this both ways around.
I'll try it out tomorrow - I'm not at the office now and it's a little bit
di
>> Well, I created an additional PE with alloacation_rule "$pe_slots",
>> and built in an if condition in "pe.jsv" for all jobs which request
>> just a single node to be assigned to this new PE. But the annoying
>> situation didn't change. The scheduler configuration is set to
>> "queue_sort_metho
> My suggestion was to modify your jsv/gepetools to force single node
> parallel jobs into PEs with $pe_slots allocation rules (which gives
> you control over where they are scheduled via queue_sort_method and
> load_formula) while sending the others to PEs with other (appropriate)
>allocation rul
> My comment was meant to be read in the context of:
> i)Reuti's earlier suggestion that jobs in PEs with $pe_slots allocation_rule
> will be scheduled according to the queue_sort_method and load_formula like
> serial jobs
> ii)Your response that settting the allocation rule to $pe_slots stops
> $pe_slots restricts you to a single node so I'm guessing the jobs that don't
> start are jobs that need more than one node.
Yes, that should be right.
> While we don't use gepetools we do have a JSV that rewrites people's
> requested PE based on the number
> What you need I think is someth
-Ursprüngliche Nachricht-
Von: Reuti [mailto:re...@staff.uni-marburg.de]
Gesendet: Mittwoch, 29. Juli 2015 15:10
An: Winkler, Ursula (ursula.wink...@uni-graz.at)
Cc: users@gridengine.org
Betreff: Re: [gridengine users] Filling up nodes when using gepetools
Hi,
>> Am 29.07.2015
>> Node1 has 12 Cores/Slots and 1 MPI-Job with 2 Slots is running on it.
>> A user submits job2 which require maximal 10 slots.
>> Independently from schedule_interval, job_load_adjustments,
>> load_formula and/or load_adjustment_decay_time parameters-settings
>> job2 usually won't start on Node1
Hi Gridengine users group members,
I have the following problem - let me explain it with following example:
Node1 has 12 Cores/Slots and 1 MPI-Job with 2 Slots is running on it.
A user submits job2 which require maximal 10 slots.
Independently from schedule_interval, job_load_adjustments, load_fo
Hi gridengine mailinglist members,
I'll trying to find a solution for an environment running serial jobs as well
as mpi jobs on
6 hosts where each host has 32 cores/slots. Due to the small number of nodes,
assigning
each sort of jobs to separate nodes (e.g. nodes 1-2 for serial, nodes 3-6 for
m
> What is the setting of "slots" in the PE itself?
492(the cluster has 35 hosts with 12 slots and 3 hosts with 24 slots; but
all hosts with 12 slots are disabled in this testqueue).
Ursula
___
users mailing list
users@gridengine.org
https://gri
obs now with
less slots...
Ursula
-Ursprüngliche Nachricht-
Von: Reuti [mailto:re...@staff.uni-marburg.de]
Gesendet: Donnerstag, 23. Oktober 2014 11:26
An: Winkler, Ursula (ursula.wink...@uni-graz.at)
Cc: William Hay; Gridengine Users Group
Betreff: Re: AW: [gridengine users] Weird scheduler cal
> "schedd_job_info" is switched on then? But even if switched off it should
> show up in `qalter -w p`.
Yes, it is on.
> And 24 slots per machine then - `qstat -g c ` reveals the slots as being free
> too?
A good question: it reveals that from 72 are now (!) 70 are free and no one is
used. Wh
> The qsub man page states that -w p and -w v don't take into account load
> values. Possibly the job is requesting a complex whose value is determined
> by a load sensor and the returned value is not suitable but not causing an
> alarm.
Should not "qstat -j " list the shortage of a complex?
n 64 slots available?
-Original Message-
From: users-boun...@gridengine.org [mailto:users-boun...@gridengine.org] On
Behalf Of Winkler, Ursula (ursula.wink...@uni-graz.at)
Sent: 21 October 2014 11:46
To: Gridengine Users Group
Subject: Re: [gridengine users] Weird scheduler calculation error?
(ursula.wink...@uni-graz.at)
Betreff: Re: [gridengine users] Weird scheduler calculation error?
Hi,
Am 21.10.2014 um 11:21 schrieb Ursula Winkler:
> Hi gridengine members,
>
> For now I ran out of ideas with an annoying problem:
>
> A job with 72 slots does not start because of &q
Hi Rafa,
for such purposes we have configured a separate Parallel Environment with
"allocation_rule" "$pe_slots" (instead of "$fill_up"). Jobs scheduled with this
rule can run ONLY on one host.
Regards,
Usula
-Ursprüngliche Nachricht-
Von: users-boun...@gridengine.org [mailto:users-bou
Hi, try
qsub -q @,... (e.g. qsub -q mpi@host1,mpi@host2, )
Ursula
Von: users-boun...@gridengine.org [mailto:users-boun...@gridengine.org] Im
Auftrag von Henrichs, Juryk
Gesendet: Freitag, 21. März 2014 11:24
An: users@gridengine.org
Betreff: [gridengine users] submitting to a specif
>> I've a question concerning "qacct" and resource requests. "qacct" prints out
>> a lot of information - but little about resource requests that jobs asked
>> for by "qsub -l ...".
>> Of course it is possible to find out via "qacct -l" the jobs with a
>> *particular* value (as one has to specif
Dear gridengine list members,
I've a question concerning "qacct" and resource requests. "qacct" prints out a
lot of information - but little about resource requests that jobs asked for by
"qsub -l ...".
Of course it is possible to find out via "qacct -l" the jobs with a
*particular* value (as o
Hi everyone,
I have a problem with SGE: it looks like (though I'm really not sure) that
running jobs are killed when other jobs are submitted which otherwise should
wait in the queue because no free cores are available. It happens repeatedly so
this raises suspicion. Unfortunately the error of
The problem is solved. Was my fault, sorry!
-Ursprüngliche Nachricht-
Von: users-boun...@gridengine.org [mailto:users-boun...@gridengine.org] Im
Auftrag von Ursula Winkler
Gesendet: Donnerstag, 05. April 2012 10:52
An: users@gridengine.org
Betreff: [gridengine users] Job not scheduled
I
Thank you!
Ursula
Von: Ron Chen [mailto:ron_chen_...@yahoo.com]
Gesendet: Donnerstag, 29. März 2012 09:34
An: Winkler, Ursula (ursula.wink...@uni-graz.at); 'Tim Wright'
Cc: 'users@gridengine.org'
Betreff: Re: [gridengine users] qtcsh building error
The install script copie
on: Tim Wright [mailto:tim.dormo...@gmail.com]
Gesendet: Mittwoch, 28. März 2012 17:22
An: Winkler, Ursula (ursula.wink...@uni-graz.at)
Betreff: Re: [gridengine users] qtcsh building error
Have you tried adding -lcurses or -lncurses to your compile string?
On 28 March 2012 15:23, Ursula Winkler
mailto:
-Ursprüngliche Nachricht-
Von: Reuti [mailto:re...@staff.uni-marburg.de]
Gesendet: Mittwoch, 28. März 2012 18:22
An: Winkler, Ursula (ursula.wink...@uni-graz.at)
Cc: users@gridengine.org
Betreff: Re: AW: [gridengine users] qtcsh building error
>>>> I know it's poss
>> I know it's possible to build the binaries without qtcsh but I'm not sure if
>> it is not needed later? Can I disclaim it?
>qtcsh is a version of tcsh where certain commands (which you can specify) are
>send to a node by SGE instead of being executed on the machine where you
>issued the comm
39 matches
Mail list logo