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

Re: [HTCondor-users] Checkpoints in HTCondor



 1.  Could you, please, let me know if it is because the script (and the
 DMTCP / HTCondor integration in general) is abandoned, or because it
 did not need any changes?

As far as I know, the script and the integration were abandonded several years ago.

 2.  Also, the link to the latest version does not seem to work:
 ftp://ftp.cs.wisc.edu/condor/contrib/dmtcp_condor_integration-0.6-Any-Any.tar.gz
 What is the new location of the file?

	I don't know; it's certainly not maintained anymore.

3. If DMTCP / HTCondor integration is not supported anymore, is it still true (http://dmtcp.sourceforge.net/condor.html) that HTCondor vanilla universe does not support checkpoint-restart?

	It only supports application-level checkpoint/restart.

 1.  What is the best way for me to check-point such jobs in HTCondor?

I don't know what the best way for you to checkpoint (your) Python 3 jobs is; application-level checkpointing can be advantageous in some situations.

- ToddM