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

Re: [HTCondor-users] CPU Usage reporting in log file



Hi Samik,

do you see CPU usage e.g. via condor_history?
If so, you should be looking for the ClassAd attributes RemoteUserCpu, RemoteSysCpu and perhaps RemoteWallClockTime and CumulativeSuspensionTime. You can get them into the log by using
	job_ad_information_attrs = RemoteUserCpu, RemoteSysCpu, RemoteWallClockTime, CumulativeSuspensionTime
in the job JDL.

Note that the Local*Cpu values reflect usage on the Schedd, and are usually not what one is looking for.

Cheers,
Max

> Am 06.07.2017 um 11:14 schrieb Samik Raychaudhuri <samikr@xxxxxxxxx>:
> 
> Group: any pointer on this?
> Thanks.
> -Samik
> 
> On 01-Jul-17 8:03 PM, Samik Raychaudhuri wrote:
>> Hello,
>> 
>> I am getting lost trying to follow the reporting of CPU usage for a job that I am running. Using 8.6.4 on a single debian VM for testing things out. 
>> 
>> When I launch the job, I clearly see using 'top' that the executable is at 99% CPU for ~3-4 seconds. However, when I examine the log file after the job completion, I do not see that reflected. I am assuming that I need to look at TotalRemoteUsage and TotalLocalUsage, but both are showing 0 values.
>> 
>> How do I get to the correct CPU usage figures?
>> 
>> Regards,
>> -Samik
>> 
>> --------------------------------
>> <c>
>>     <a n="Proc"><i>0</i></a>
>>     <a n="Cluster"><i>18</i></a>
>>     <a n="EventTime"><s>2017-07-01T19:55:44</s></a>
>>     <a n="MyType"><s>ExecuteEvent</s></a>
>>     <a n="ExecuteHost"><s>&lt;10.0.2.15:9618?addrs=10.0.2.15-9618+[--1]-9618&amp;noUDP&amp;sock=671_d423_5&gt;</s></a>
>>     <a n="Subproc"><i>0</i></a>
>>     <a n="EventTypeNumber"><i>1</i></a>
>> </c>
>> <c>
>>     <a n="ResidentSetSize"><i>3008</i></a>
>>     <a n="Size"><i>750</i></a>
>>     <a n="MemoryUsage"><i>3</i></a>
>>     <a n="Proc"><i>0</i></a>
>>     <a n="Cluster"><i>18</i></a>
>>     <a n="EventTime"><s>2017-07-01T19:55:53</s></a>
>>     <a n="MyType"><s>JobImageSizeEvent</s></a>
>>     <a n="Subproc"><i>0</i></a>
>>     <a n="EventTypeNumber"><i>6</i></a>
>> </c>
>> <c>
>>     <a n="SentBytes"><r>1.646900000000000E+04</r></a>
>>     <a n="TotalRemoteUsage"><s>Usr 0 00:00:00, Sys 0 00:00:00</s></a>
>>     <a n="TotalLocalUsage"><s>Usr 0 00:00:00, Sys 0 00:00:00</s></a>
>>     <a n="EventTypeNumber"><i>5</i></a>
>>     <a n="TotalSentBytes"><r>1.646900000000000E+04</r></a>
>>     <a n="Subproc"><i>0</i></a>
>>     <a n="MyType"><s>JobTerminatedEvent</s></a>
>>     <a n="RunRemoteUsage"><s>Usr 0 00:00:00, Sys 0 00:00:00</s></a>
>>     <a n="EventTime"><s>2017-07-01T19:55:53</s></a>
>>     <a n="Cluster"><i>18</i></a>
>>     <a n="Proc"><i>0</i></a>
>>     <a n="ReceivedBytes"><r>5.023000000000000E+03</r></a>
>>     <a n="TerminatedNormally"><b v="t"/></a>
>>     <a n="TotalReceivedBytes"><r>5.023000000000000E+03</r></a>
>>     <a n="ReturnValue"><i>0</i></a>
>>     <a n="RunLocalUsage"><s>Usr 0 00:00:00, Sys 0 00:00:00</s></a>
>> </c>
>> ---------------------------------------------------
> 
> _______________________________________________
> 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/

Attachment: smime.p7s
Description: S/MIME cryptographic signature