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

Re: [Condor-users] memory usage info



On Tue, 5 Feb 2008, Leo Cristobal C. Ambolode II wrote:

Hi all,

This is not an answer to carl langlois's inquiry. This is an additional
inquiry about memory usage also. I want o know on how to get the average
memory used and average CPU used by a certain job. I can always get the
EXECUTION TIME (Real Time) of the job from the log (from job submission to
job execution to job termination), but can I include some attributes in my
submit file to also get/(or stream) the average CPU and MEMORY usage of my
program/job? How could I do this?

The current memory usage of your job is kept track of by ImageSize
and you can find it by using condor_q -l or condor_q -format.
This is a raw measure of memory that doesn't keep track of
multi-threaded applications very well.

The actual cpu usage of your job can be monitored by remoteusercpu
and remotesyscpu in your job classad, although it should be noted
that these values are not always right.

There is also a measure of current disk usage but it tends to lag
the real disk usage by 20-30 minutes.

No special attributes are required in the submit file to get any
of these.

Steve Timm


I'm running condor version 6.8.2 on Scientific Linux 4.3.

Thanks.

Leo

Actual memory usage is a bit difficult to define
and also very difficult to get from the OS.  Many
people use RSS (resident set size), but this
doesn't really reflect *actual* memory usage due
to page sharing (who should get the bill for
glibc?).  Additionally, RSS is not a good measure
of the memory a program needs, only of what it can
get  (it may need more, it may need less).

_joe

Ian Chesal wrote:
I was looking at condor_status -direct "host" -long ouput and was
wondering if the current usage in memory is represent somwhere?
I saw TotalMemory, TotalVirtualmemory... but that does not seem to
reflect the actual usage of memory.

Condor tracks the memory used by a job under its control in the
'ImageSize' classad variable. There's probably some fascinating ancedote
for why it's called ImageSize, but I don't know it.

Also, it's not always accurate depending on what you're running and on
what OS.

- Ian


Confidentiality Notice.  This message may contain information that is
confidential or otherwise protected from disclosure.
If you are not the intended recipient, you are hereby notified that any
use, disclosure, dissemination, distribution,
or copying of this message, or any attachments, is strictly prohibited.
If you have received this message in error,
please advise the sender by reply e-mail, and delete the message and any
attachments.  Thank you.




_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx 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@xxxxxxxxxxx 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@xxxxxxxxxxx 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/


--
------------------------------------------------------------------
Steven C. Timm, Ph.D  (630) 840-8525
timm@xxxxxxxx  http://home.fnal.gov/~timm/
Fermilab Computing Division, Scientific Computing Facilities,
Grid Facilities Department, FermiGrid Services Group, Assistant Group Leader.