Exploiting user activeness for data retention in HPC systems
Exploiting user activeness for data retention in HPC systems
复制标题
利用用户活跃度来保留 HPC 系统中的数据
DOI：
10.1145/3458817.3476201
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Chen, Yong
中科院分区：
文献类型：
--
作者：
Zhang, Wei;Byna, Suren;Sim, Hyogi;Lee, Sangkeun;Vazhkudai, Sudharshan;Chen, Yong
HPC系统通常依赖于固定生存期（FLT）的数据保留策略，它只考虑并行文件系统的数据访问的时间局部性。然而，我们基于领导级HPC系统跟踪的广泛分析表明，FLT方法通常无法捕获用户行为中的动态，并导致不期望的数据清除。在这项研究中，我们提出了一个基于活动的数据保留（ActiveDR）的解决方案，主张从整体的基于活动的角度考虑数据保留方法。通过评估用户活动的频率和影响，ActiveDR为非活动用户确定文件清除过程的优先级，并在并行存储上为活动用户提供延长的文件生命周期。我们基于先前Titan超级计算机的跟踪进行的广泛评估表明，与当前的FLT保留方法相比，当达到相同的清除目标时，ActiveDR实现了高达37%的文件未命中减少。
HPC systems typically rely on the fixed-lifetime (FLT) data retention strategy, which only considers temporal locality of data accesses to parallel file systems. However, our extensive analysis based on the leadership-class HPC system traces suggests that the FLT approach often fails to capture the dynamics in users' behavior and leads to undesired data purge. In this study, we propose an activeness-based data retention (ActiveDR) solution, which advocates considering the data retention approach from a holistic activeness-based perspective. By evaluating the frequency and impact of users' activities, ActiveDR prioritizes the file purge process for inactive users and rewards active users with extended file lifetime on parallel storage. Our extensive evaluations based on the traces of the prior Titan supercomputer show that, when reaching the same purge target, ActiveDR achieves up to 37% file miss reduction as compared to the current FLT retention methodology.