[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [HTCondor-users] Cannot use multiple arguments to run MPI application in "parallel" universe



My collegue just explained to me that this isnât going to work the way you want it to because /public/openmpiscript

behaves differently on the head node than it does on all of the other nodes.  On the head node it pays attention to the

arguments passed to it by HTCondor (from the  queue statement).   On the other nodes, it ignores those arguments, instead getting the

arguments by talking to the head node.

 

This is why only the args passed to the head node are being used by all of the jobs.

 

You need to use the method Jason suggested.

 

If that doesnât work for you, you could have your job look in the .job.ad file to find the arguments that were passed to /public/openmpiscript

 

-tj

 

From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> On Behalf Of John M Knoeller
Sent: Tuesday, October 30, 2018 9:24 AM
To: HTCondor-Users Mail List <htcondor-users@xxxxxxxxxxx>
Subject: Re: [HTCondor-users] Cannot use multiple arguments to run MPI application in "parallel" universe

 

Iâm sorry.  the thing I asked you to try has the wrong type of braces.  It should be (), not {}.  like this

 

 

queue arguments from (

mympiapp args1

mympiapp args2

)

 

Did you notice my mistake and correct it?  because if you tried what I suggested you would have gotten an error message from condor_submit.  If you did not, then the problem is that you are not submitting the file that you think you are.

 

try adding the arguments

 

  -dump test.out

 

to your condor_submit command line and then look at the contents of test.out.  You should see something like this at the end

 

Args="mympiapp args1"

ProcId=0

JobStatus=1

.

ProcId=1

JobStatus=1

Args="mympiapp args2"

.

 

 

-tj

 

 

From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> On Behalf Of Tiger Hu
Sent: Monday, October 29, 2018 9:35 PM
To: HTCondor-Users Mail List <htcondor-users@xxxxxxxxxxx>
Subject: Re: [HTCondor-users] Cannot use multiple arguments to run MPI application in "parallel" universe

 

John,

 

I tried your method but still got the same result. âArgs2â wasnât be used as before. Any ideas?

Sent from my iPhone


å 2018å10æ30æïäå4:37ïJohn M Knoeller <johnkn@xxxxxxxxxxx> åéï

try this

 

universe=parallel

executable=/public/openmpiscript

machine_count=2

queue arguments from {

mympiapp args1

mympiapp args2

}

 

-tj

 

 

 

From: HTCondor-users <htcondor-users-bounces@xxxxxxxxxxx> On Behalf Of hufh
Sent: Friday, October 26, 2018 10:57 AM
To: htcondor-users@xxxxxxxxxxx
Subject: [HTCondor-users] Cannot use multiple arguments to run MPI application in "parallel" universe

 

Dear,

 

I tried to run a OpenMPI application with different arguments by writing two "queue" section in"parallel" universe, 

 

Here is my submission file:

universe=parallel

executable=/public/openmpiscript

machine_count=2

arguments="mympiapp args1"

queue

 

machine_count=2

arguments="mympiapp args2"

queue

 

I am expecting that two "mympiapp" instances running with "args1" and "args2" respectively, each running on two machines, but the reality is that only the first one("mympiapp args1") is running, but on four machines, not two machines, looks like it is using the resource claimed in two queues).

 

Who can give me hand? Thanks a lot.

 

hufh

 

 

_______________________________________________
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/