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

[HTCondor-users] condor batch job in flocking cluster

Hi Jay,

Need your suggestion if you have any better idea?

I am interpreting my batch jobs package which I use to run on my cluster at FresnoState(A single one) to the same if I run through "AtlasConnect" (A group of clusters joined together like flocking)

Let me try to explain how my batch job structure is:

I have submitparallel.sh as my main file. Here I split my inputfile into n parts. But the thing is that to use split.sh I have to choose some directory.

In my fresnoState since all the nodes have my home or /tmp as common. so I can use that!!!

But in ATLASCONNECT, I cannot use since there is no common dir. Because I am not sure which cluster my job goes and I may not be having home area or any account over there.

I am attaching the submitparallel.sh and parallel.sh where one can see how I split my input in submitparallel.sh and then use in parallel.sh

Any idea or suggestion is appreciated


getenv         = false
should_transfer_files = YES
transfer_input_files = lists, dijet.tgz,/tmp/x509up_u55261
+ProjectName = "atlas-org-fresnostate"

Attachment: submitParallel.sh
Description: Bourne shell script

Attachment: parallel.sh
Description: Bourne shell script

Attachment: input.list
Description: Binary data

Attachment: command_to_submit
Description: Binary data