Optimizing checkpoint data placement with guaranteed burst buffer endurance in large-scale hierarchical storage systems

摘要

Non-volatile devices, such as SSDs, will be an integral part of the deepening storage hierarchy on large-scale HPC systems. These devices can be on the compute nodes as part of a distributed burst buffer service or they can be external. Wherever they are located in the hierarchy, one critical design issue is the SSD endurance under the write-heavy workloads, such as the checkpoint I/O for scientific applications. For these environments, it is widely assumed that checkpoint operations can occur once every 60 min and for each checkpoint step as much as half of the system memory can be written out. Unfortunately, for large-scale HPC applications, the burst buffer SSDs can be worn out much more quickly given the extensive amount of data written at every checkpoint step. One possible solution is to control the amount of data written by reducing the checkpoint frequency. However, a direct effect caused by reduced checkpoint frequency is the increased vulnerability window of system failures and therefore potentially wasted computation time, especially for large-scale compute jobs. In this paper, we propose a new checkpoint placement optimization model which collaboratively utilizes both the burst buffer and the parallel file system to store the checkpoints, with design goals of maximizing computation efficiency while guaranteeing the SSD endurance requirements. Moreover, we present an adaptive algorithm which can dynamically adjust the checkpoint placement based on the system's dynamic runtime characteristics and continuously optimize the burst buffer utilization. The evaluation results show that by using our adaptive checkpoint placement algorithm we can guarantee the burst buffer endurance with at most 5% performance degradation per application and less than 3% for the entire system.