Abstract
In an asymmetric multi-core architecture, multiple heterogeneous cores share the last-level cache (LLC). Due to the different memory access requirements among heterogeneous cores, the LLC competition is more intense. In the current work, we propose a heterogeneity-aware replacement policy for the partitioned cache (HAPC), which reduces the mutual interference between cores through cache partitioning, and tracks the shared reuse state of each cache block within the partition at runtime to guide the replacement policy to keep cache blocks shared by multiple cores in multithreaded programs. In the process of updating the reuse state, considering the difference of memory accesses to LLC by heterogeneous cores, the cache replacement policy tends to keep cache blocks required by big cores, to better improve the LLC access efficiency of big cores. Compared with LRU and the SRCP, which are the state-of-the-art cache replacement algorithms, the performance of big cores can be significantly improved by HAPC when running multithreaded programs, while the impact on little cores is almost negligible, thus improving the overall performance of the system.
Funder
National Natural Science Foundation of China
Beijing Natural Science Foundation
Subject
Electrical and Electronic Engineering,Mechanical Engineering,Control and Systems Engineering
Reference36 articles.
1. There’s plenty of room at the Top: What will drive computer performance after Moore’s law?;Leiserson;Science,2020
2. Kim, C., and Huh, J. (2016, January 1–3). Fairness-Oriented OS Scheduling Support for Multi-core Systems. Proceedings of the 2016 International Conference on Supercomputing, Istanbul, Turkey.
3. A Survey of Techniques for Architecting and Managing Asymmetric Multi-core Processors;Mittal;ACM Comput. Surv.,2016
4. Improving multitask performance and energy consumption with partial-ISA multi-cores;Souza;J. Parallel Distrib. Comput.,2021
5. Karaoui, M.L., Carno, A., Lyerly, R., Kim, S.H., Olivier, P., Min, C., and Ravindran, B. (2019, January 16–20). POSTER: Scheduling HPCWorkloads on Heterogeneous-ISA Architectures. Proceedings of the 24th Symposium on Principles and Practice of Parallel Programming, New York, NY, USA.
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献