Checkpointing and Localized Recovery for Nested Fork-Join Programs

TL;DR

This paper extends a low-overhead checkpointing and localized recovery technique from independent tasks to nested fork-join programs in distributed memory systems, enabling efficient fault recovery.

Contribution

It adapts an existing low-overhead checkpointing method to nested fork-join programs with work-stealing, maintaining minimal overheads.

Findings

Checkpointing overheads below 1% expected to be maintained

Localized recovery allows unaffected processes to continue

Algorithmic changes enable application to nested fork-join programs

Abstract

While checkpointing is typically combined with a restart of the whole application, localized recovery permits all but the affected processes to continue. In task-based cluster programming, for instance, the application can then be finished on the intact nodes, and the lost tasks be reassigned. This extended abstract suggests to adapt a checkpointing and localized recovery technique that has originally been developed for independent tasks to nested fork-join programs. We consider a Cilk-like work stealing scheme with work-first policy in a distributed memory setting, and describe the required algorithmic changes. The original technique has checkpointing overheads below 1% and neglectable costs for recovery, we expect the new algorithm to achieve a similar performance.