Checkpointing strategies with prediction windows

TL;DR

This paper introduces a novel checkpointing strategy that optimally manages fault prediction windows, improving system efficiency through analytical modeling and simulation validation.

Contribution

It proposes a new approach using two periodic modes for checkpointing based on prediction window sizes, optimizing platform waste minimization.

Findings

Optimal checkpointing periods derived for various window sizes

Model validated through extensive simulations

Strategy reduces system waste effectively

Abstract

This paper deals with the impact of fault prediction techniques on checkpointing strategies. We suppose that the fault-prediction system provides prediction windows instead of exact predictions, which dramatically complicates the analysis of the checkpointing strategies. We propose a new approach based upon two periodic modes, a regular mode outside prediction windows, and a proactive mode inside prediction windows, whenever the size of these windows is large enough. We are able to compute the best period for any size of the prediction windows, thereby deriving the scheduling strategy that minimizes platform waste. In addition, the results of this analytical evaluation are nicely corroborated by a comprehensive set of simulations, which demonstrate the validity of the model and the accuracy of the approach.