Efficient techniques for adaptive independent checkpointing in distributedsystems

This work presents two novel algorithms to prevent rollback propagation for independent checkpointing: an efficient adaptive independent checkpointing algorithm and an optimized adaptive independent checkpointing algorithm. T he last opportunity strategy that yields a better performance than the cons ervation strategy is also employed to prevent useless checkpoints for both causal rewinding paths and non-causal rewinding paths. The two methods prop osed herein are domino effect-free and require only a limited amount of con trol information. They also take less unnecessary adaptive checkpoints than other algorithms. Furthermore, experimental results indicate Chat the chec kpoint overhead of our techniques is lower than that of the coordinated che ckpointing and domino effect-free algorithms fur service-providing applicat ions.