Training large-scale machine learning (ML) models is expensive and time-intensive, consuming many hardware accelerators for days or weeks. As the scale of hardware deployments and training time continue to grow, the probability of failures also increases. The desire to use cheaper cloud resources, such as spot VMs, to lower costs also dramatically increases the frequency of failures. The standard approach to deal with failures is to periodically pause training and checkpoint model parameters to persistent storage. Unfortunately, today's checkpointing mechanisms introduce high overhead when applied at high frequencies, yet frequent checkpointing is necessary to avoid long recovery times. We present a concurrent checkpointing mechanism, PCcheck, that allows frequent checkpointing with minimal overhead. Our framework supports persisting checkpoints to SSD and persistent main memory (PMEM) for both single-machine and distributed settings. PCcheck enables checkpointing as frequently as every 10 iterations for detailed monitoring and fast recovery times in case of failures, while maintaining minimal (3%) overhead on training throughput.

@inproceedings{Strati0K25,
  title = {PCcheck: Persistent Concurrent Checkpointing for ML},
  author = {Foteini Strati and Michal Friedman and Ana Klimovic},
  booktitle = {ASPLOS},
  year = {2025},
  doi = {10.1145/3669940.3707255}
}