Communication-Induced Checkpointing with Message Logging beyond the Piecewise Deterministic (PWD) Model for Distributed Systems

This paper introduces an effective communication-induced checkpointing protocol using message logging to enable the number of extra checkpoints be far lower than previous number. Even if a situation occurs in which it is decided that process receiving has perform forced checkpointing, our allows skip action recognizes state its sender right before receipt recoverable. Additionally, thus not required assume piecewise deterministic model, despite being combined with logging. can maintain these features by piggybacking one-bit variable and n-size vector on each sent. Our simulation results verify claim presented performs much better representative optimized respect frequency, regardless communication pattern.