The Scalable Checkpoint/Restart (SCR) library enables MPI applications to utilize distributed storage on Linux clusters to attain high file I/O bandwidth for checkpointing, restarting, and writing large datasets. The second 3.0 release candidate includes:

• added support for multiple outstanding asynchronous flushes
• restored watchdog support on SLURM systems
• updated SCR_Config behavior
• added support for static-only builds
• and much more!

Learn more:

• SCR v3.0rc2 release notes
• SCR GitHub repo
• SCR documentation
• SCR: Scalable Checkpoint/Restart for MPI