Shared Cache Allocation Based on Fairness in a Chip Multiprocessor Architecture

Cache, as the hub between the multi-core processor and the memory, is closely related to the performance of the CMP system. And with the number of processor core increasing, the contention of multi-core for shared cache becomes more intense. Fairness is important to affect the performance of CMP systems, so a Shared Cache allocation method based on fairness is proposed. According to the way of borrow-return, the method assigns Shared Cache to multiple to realize the dynamic balance. The experimental results show that the method significantly improves the system fairness and the system throughput.

[1]  Jichuan Chang,et al.  Cooperative cache partitioning for chip multiprocessors , 2007, ICS '07.

[2]  Shirshendu Das,et al.  Towards a Better Cache Utilization by Selective Data Storage for CMP Last Level Caches , 2016, 2016 29th International Conference on VLSI Design and 2016 15th International Conference on Embedded Systems (VLSID).

[3]  Vijay S. Pai,et al.  Runtime-driven shared last-level cache management for task-parallel programs , 2015, SC15: International Conference for High Performance Computing, Networking, Storage and Analysis.

[4]  Yingwei Luo,et al.  Optimal Cache Partition-Sharing , 2015, 2015 44th International Conference on Parallel Processing.

[5]  S. Kim,et al.  Fair cache sharing and partitioning in a chip multiprocessor architecture , 2004, Proceedings. 13th International Conference on Parallel Architecture and Compilation Techniques, 2004. PACT 2004..