Checkpointing Jobs

Checkpointing is a technique to take a long running program and break it up into smaller pieces. Checkpointing has many benefits especially in the context of running jobs on the HPCC. You can also use checkpointing to take advantage of the Scavenger Queue.

For details, please see:

• The Checkpointing Overview
• Our tutorials for:
  • Checkpointing using DMTCP
  • Checkpointing using DMTCP in batch jobs
  • Checkpointing using the longjob powertool
• A template script to checkpoint your own code