parfor on a supercomputer

8 Ansichten (letzte 30 Tage)
Michelle
Michelle am 29 Sep. 2014
Beantwortet: Raymond Norris am 30 Sep. 2014
Hello. I'm trying to run a code that uses matlabpool and parfor on a supercomputer. The code works on my desktop, but I'm getting weird error messages when I run it on a supercomputer. Please help The output I get is
Destroying 2 pre-existing communicating jobs created by matlabpool.
[^HWarning: Cannot cancel or destroy a job that was not created by this Local
cluster.]^H
[^H> In Local.cancelOrDestroyJob at 35
In Local.Local>Local.hDestroyJob at 207
In CJSJobMethods>CJSJobMethods.destroyOneJob at 70
In CJSCommunicatingJob>CJSCommunicatingJob.destroyOneJob at 85
In Job.Job>Job.delete at 735
In distcomp.interactiveclient.pGetClusterAdaptor>@(jobs)delete(jobs) at 10
In distcomp.interactiveclient.cleanup at 31
In MatlabpoolHelper>MatlabpoolHelper.doClose at 328
In MatlabpoolHelper>MatlabpoolHelper.doMatlabpool at 139
In matlabpool at 139
In tj2Janus at 1
In run at 64]^H
Starting matlabpool using the 'local' profile ... connected to 4 workers.
{^HError using parallel_function (line 589)
Index exceeds matrix dimensions.
Error in tj2Janus (line 10)
parfor ci=1:5%length(c_s)
Error in run (line 64)
evalin('caller', [script ';']);
}^H
^G>> [^HWarning: Objects of distcomp.abstractstorage class exist - not clearing this
class
or any of its superclasses]^H
[^HWarning: Objects of distcomp.filestorage class exist - not clearing this class
or any of its superclasses]^H
[^HWarning: Objects of distcomp.serializer class exist - not clearing this class
or any of its superclasses]^H
[^HWarning: Objects of distcomp.fileserializer class exist - not clearing this
class
or any of its superclasses]^H
  1 Kommentar
José-Luis
José-Luis am 29 Sep. 2014
Strange. Is anybody else running parallel matlab jobs there?

Melden Sie sich an, um zu kommentieren.

Antworten (1)

Raymond Norris
Raymond Norris am 30 Sep. 2014
I would suggest contacting MathWorks Technical Support (<mailto:support@mathworks.com support@mathworks.com>)

Kategorien

Mehr zu Parallel Computing Toolbox finden Sie in Help Center und File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by