Checkpointing algorithms and fault prediction
Journal of Parallel and Distributed Computing
Hi-index | 0.00 |
This paper presents a new checkpoint scheme that utilizes the memory usage profile and time series analysis for low-overheadcheckpoint.The proposed checkpoint scheme checks current and future checkpoint overhead based on the on the changes of the memory size and the expected check-point overhead usingmemory profile and adaptive time series analysis when it decideswhether or not to take a check-point.Unlike the previous works that do not utilize the memory usage profile, it is possible to reduce the total over-head of the execution time.We also presentexperimental results which show that the checkpoint overhead of the pro-posed scheme is reduced compared with the previously developed checkpoint scheme.