Did you take a look at the “claim” time? Sometimes, machines remain allocated even after job completes assuming the same user may fire another job. This time is configurable. Just check this out, in case if this is your problem. Oh Sorry, I missed the interval part. Can you tell me which section in the manual? (I want to understand because it is relevant to us as well) From: condor-users-bounces@xxxxxxxxxxx [mailto:condor-users-bounces@xxxxxxxxxxx] On Behalf Of Guy Schwartz Sent: Tuesday, July 17, 2012 6:11 PM To: Condor-Users Mail List Subject: Re: [Condor-users] Fast job execution and matchmaking Hi,
Using the vanilla universe. As i understand from the manuals and config files the smaller the intervals, the faster the job will find the proper home.
if i schedule 100 of the same job (queue 100), they do run faster (not 100*8 sec) but its not the case in our lab. we run individual jobs....
Thanks,
Guy.
On Tue, Jul 17, 2012 at 2:36 PM, Sarnath K - ERS, HCLTech <k_sarnath@xxxxxxx> wrote: Hi Guy,
What is the job universe?
I don't think Intervals are related to job scheduling latency though I may be wrong. I think it is better you keep the intervals back to 300.
Did you ever try with the default intervals? Do you observe 9 second latency for each and every job that you fire? Did you try firing jobs continuously in a shell script (say some 100 jobs) and see what happen?
Best Regards, Sarnath -----Original Message----- From: condor-users-bounces@xxxxxxxxxxx [mailto:condor-users-bounces@xxxxxxxxxxx] On Behalf Of Guy Sent: Tuesday, July 17, 2012 5:47 PM To: condor-users@xxxxxxxxxxx Subject: [Condor-users] Fast job execution and matchmaking
Hi!
We have a local condor pool in our university lab. a condor master and several quad core computers (all with the same hardware).
We want to run many small condor jobs, and we want very fast matchmaking/execution.
I did set all the intervals in the config file to 1sec (instead of 300). we don't care about the network traffic between them because it is just 20 cores and not 2000.
The weird thing is it still takes 8 or 9 seconds until a job executes...
Any way to make it faster? maybe a config parameter that i am missing?
Could i set an interval of under 1 second?
Maybe i can change something in the source codes?
Sorry about all the questions..... hoping that someone can help.
Thanks,
Guy
_______________________________________________ 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/
::DISCLAIMER:: ----------------------------------------------------------------------------------------------------------------------------------------------------
The contents of this e-mail and any attachment(s) are confidential and intended for the named recipient(s) only. E-mail transmission is not guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or may contain viruses in transmission. The e mail and its contents (with or without referred errors) shall therefore not attach any liability on the originator or HCL or its affiliates. Views or opinions, if any, presented in this email are solely those of the author and may not necessarily reflect the views or opinions of HCL or its affiliates. Any form of reproduction, dissemination, copying, disclosure, modification, distribution and / or publication of this message without the prior written consent of authorized representative of HCL is strictly prohibited. If you have received this email in error please delete it and notify the sender immediately. Before opening any email and/or attachments, please check them for viruses and other defects.
---------------------------------------------------------------------------------------------------------------------------------------------------- |