Re: [Condor-users] Condor Group Drive me Crazy.......

Mailing List Archives Authenticated access	UW Madison Computer Sciences Department Computer Systems Lab

On Mon, Nov 14, 2011 at 9:14 PM, Joe Boyd <boyd@xxxxxxxx> wrote:

Why can't you just set all the GROUP_ACCEPT_SURPLUS parameters to FALSE? What doesn't work the way you want then?

Well if I set it to false (based on the configuration) the GROUP_QUOTA_group_vcs.__verification_single will only have up to 5 jobs....

If the GROUP_ACCEPT_SURPLUS is set to true then if no one running a job from the other groups then it will have 13.

The all idea is to have some control on jobs that require FlexLM lic.

I know about Concurrency Limits but I found it not to be a good option in my case.

My users want to make sure that If the provide one of the VCS group it will start run right away. (this under the considerations that no other jobs from the same group is in the pool. So if 10 jobs already exists in the pool from the group design___single, and 10 more jobs are being submitted with the design___single group definition, then it is understood that the pool will be process in a FIFO way)

Here is the conf again:

GROUP_QUOTA_group_vcs = 13

GROUP_QUOTA_group_vcs.design___single = 4
GROUP_QUOTA_group_vcs.design___list = 1
GROUP_QUOTA_group_vcs.__verification_single = 5
GROUP_QUOTA_group_vcs.__verification_list = 3

joe

Sassy Natan wrote:

Thanks For the Help Man!

On Mon, Nov 14, 2011 at 8:01 PM, Joe Boyd <boyd@xxxxxxxx <mailto:boyd@xxxxxxxx>> wrote:

I may have missed a part of the thread or something as I'm not sure
what you're trying to have it do in the end.

If you have any of the GROUP_ACCEPT_SURPLUS_* parameters set to TRUE
that group will end up running more than the quota set if you have
free slots. You said it only gets 13 jobs running when you submit
that first job. Is that true even after several negotiation
cycles??? I'm surprised it wouldn't run more.
yes, even after several negotiation cycles running job not going up more then 13.

If I send the two submission file the same time (60 job totals of two group) I also don't get more then 13 jobs...

I see that you have

GROUP_ACCEPT_SURPLUS_group_vcs = FALSE

but I don't think that's going to make the top level group_vcs not
go above it's 13 if the subgroups have it TRUE. I'm not sure that
parameter is really enforced down the hierarchy (sounds like it's
not from your experience). Is that why you're saying it shouldn't
run more than 13? Because of the setting I quote above?

Yes this is what I'm saying.....

In that case I don't understand the preemption method.
what do u suggester?

joe

Sassy Natan wrote:

Hi Joe,

Well yes this is true... When setting the GROUP_ACCEPT_SURPLUS_*
to FALSE jobs doesn't leap outside the quota limit.

However, Since this configuration use Sub Groups I expect to
have dynamic allocation inside the group.
So in my configuration:

GROUP_QUOTA_group_vcs = 13
GROUP_QUOTA_group_vcs.design___single = 4
GROUP_QUOTA_group_vcs.design___list = 1
GROUP_QUOTA_group_vcs.__verification_single = 5
GROUP_QUOTA_group_vcs.__verification_list = 3
The VCS group has limit of 13 slots right?

So when someone from the vcs.verification_single send a Job
(queue 30) - and the pool is clean (no jobs at the moment) the
number of current running jobs should be 13 (17 in idle)
This in fact what happen when I submit the job. But once a send
a new job (queue 30) - from the verification_list I would expect
that at least 3 jobs will run right away, causing 3 jobs from
the vcs.verification_single group to be preempted or killed.
However what is happening is that the 13 jobs of the
vcs.verification_single group are keep running and 3 something
even 4 jobs being added to running state. Leaving me with total
of 16-17 running jobs which is not good.

Any Guess?

I working on this all day without any luck :-(

Thanks
Sassy On Mon, Nov 14, 2011 at 5:43 PM, Joe Boyd <boyd@xxxxxxxx
<mailto:boyd@xxxxxxxx> <mailto:boyd@xxxxxxxx

<mailto:boyd@xxxxxxxx>>> wrote:

If you want those groups to be limited to only what the quota has
you don't want to set these to TRUE do you?

GROUP_ACCEPT_SURPLUS_group_____vcs.verification_list = TRUE
GROUP_ACCEPT_SURPLUS_group_____vcs.verification_single = TRUE

That's telling it that those groups can use any "surplus"
slots in
the pool outside of the quota configuration if no one else is
using
them. If you set those to FALSE doesn't it do what you want?

joe

Sassy Natan wrote:

Hi Again....

I'm kind of lost here.
Enable debug mode and check the logs and still no good.

I attach the condor.local.conf file ....

Thanks for the help....

On Sun, Nov 13, 2011 at 6:00 PM, Sassy Natan
<sassyn@xxxxxxxxx <mailto:sassyn@xxxxxxxxx>
<mailto:sassyn@xxxxxxxxx <mailto:sassyn@xxxxxxxxx>>
<mailto:sassyn@xxxxxxxxx <mailto:sassyn@xxxxxxxxx>

<mailto:sassyn@xxxxxxxxx <mailto:sassyn@xxxxxxxxx>>>> wrote:

Hi All
Here is cut and paste from my condor configuration file:

GROUP_NAMES = GROUP_VCS, GROUP_VCS.DESIGN_SINGLE,
GROUP_VCS.DESIGN_LIST, GROUP_VCS.VERIFICATION_SINGLE,
GROUP_VCS.VERIFICATION_LIST

GROUP_QUOTA_group_vcs = 13
GROUP_QUOTA_group_vcs.design_____single = 4
GROUP_QUOTA_group_vcs.design_____list = 1
GROUP_QUOTA_group_vcs.____verification_single = 5
GROUP_QUOTA_group_vcs.____verification_list = 3

GROUP_AUTOREGROUP = FALSE
GROUP_ACCEPT_SURPLUS = FALSE

GROUP_AUTOREGROUP_group_vcs = FALSE
GROUP_ACCEPT_SURPLUS_group_vcs = FALSE

GROUP_AUTOREGROUP_group_vcs.____design_single = FALSE
GROUP_ACCEPT_SURPLUS_group_____vcs.design_single = TRUE

GROUP_AUTOREGROUP_group_vcs.____design_list = FALSE
GROUP_ACCEPT_SURPLUS_group_____vcs.design_list = TRUE

GROUP_AUTOREGROUP_group_vcs.____verification_single =
FALSE
GROUP_ACCEPT_SURPLUS_group_____vcs.verification_single
= TRUE

GROUP_AUTOREGROUP_group_vcs.____verification_list = FALSE
GROUP_ACCEPT_SURPLUS_group_____vcs.verification_list

= TRUE

I have now 2 submission files, each with 100 Jobs....
submit the first file name: verification_single.sub start
processing
13 jobs as expected (with the
group group_vcs.verification_single specified in the
submit file)

so far everything is good...
after 5 min I now submitting the next file
name verification_list.sub (with the
group group_vcs.verification_list specified in the
submit file)

Expected results are that at least 4 jobs from
verification_list.sub
will start run and total of 13 fobs will run in the
cluster.
All other 187 jobs should be idle consider none of
them as
finished
(Each submission include 100 jobs).

However the real results is that I get 18 jobs running
which
is not
good! Why? Why? Why? Why?
I just don't understand it.

I also enable NEGOTIATOR_CONSIDER_PREEMPTION since I would
like to
use PREEMPTION.
I would expect that from the 13 running process from
the verification_single.sub submission, once I submit
the verification_list.sub, 4 jobs will be PREEMPT...

Takes for any help....
Sassy

------------------------------____----------------------------__--__------------

___________________________________________________

Condor-users mailing list
To unsubscribe, send a message to
condor-users-request@xxxxxxxxx___edu
<mailto:condor-users-request@__cs.wisc.edu

<mailto:condor-users-request@cs.wisc.edu>> with a

subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/____mailman/listinfo/condor-users
<https://lists.cs.wisc.edu/__mailman/listinfo/condor-users>

<https://lists.cs.wisc.edu/__mailman/listinfo/condor-users
<https://lists.cs.wisc.edu/mailman/listinfo/condor-users>>

The archives can be found at:
https://lists.cs.wisc.edu/____archive/condor-users/
<https://lists.cs.wisc.edu/__archive/condor-users/>
<https://lists.cs.wisc.edu/__archive/condor-users/
<https://lists.cs.wisc.edu/archive/condor-users/>>

___________________________________________________

Condor-users mailing list
To unsubscribe, send a message to
condor-users-request@xxxxxxxxx___edu
<mailto:condor-users-request@__cs.wisc.edu

<mailto:condor-users-request@cs.wisc.edu>> with a

subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/____mailman/listinfo/condor-users
<https://lists.cs.wisc.edu/__mailman/listinfo/condor-users>

<https://lists.cs.wisc.edu/__mailman/listinfo/condor-users
<https://lists.cs.wisc.edu/mailman/listinfo/condor-users>>

The archives can be found at:
https://lists.cs.wisc.edu/____archive/condor-users/
<https://lists.cs.wisc.edu/__archive/condor-users/>
<https://lists.cs.wisc.edu/__archive/condor-users/
<https://lists.cs.wisc.edu/archive/condor-users/>>

------------------------------__------------------------------__------------

_________________________________________________
Condor-users mailing list
To unsubscribe, send a message to
condor-users-request@xxxxxxxxx_edu
<mailto:condor-users-request@cs.wisc.edu> with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/__mailman/listinfo/condor-users
<https://lists.cs.wisc.edu/mailman/listinfo/condor-users>

The archives can be found at:
https://lists.cs.wisc.edu/__archive/condor-users/
<https://lists.cs.wisc.edu/archive/condor-users/>

_________________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxx_edu
<mailto:condor-users-request@cs.wisc.edu> with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/__mailman/listinfo/condor-users
<https://lists.cs.wisc.edu/mailman/listinfo/condor-users>

The archives can be found at:
https://lists.cs.wisc.edu/__archive/condor-users/
<https://lists.cs.wisc.edu/archive/condor-users/>

------------------------------------------------------------------------

_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxedu with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/condor-users/

_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxedu with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/condor-users/

Mailing List Archives

Authenticated access

Re: [Condor-users] Condor Group Drive me Crazy.......