TY - GEN
T1 - High-endurance hybrid cache design in CMP architecture with cache partitioning and access-aware policy
AU - Syu, Shun Ming
AU - Shao, Yu Hui
AU - Lin, Ing Chao
PY - 2013
Y1 - 2013
N2 - In recent years, NVM (non-volatile memory) technologies, such as STT-RAM (spin transfer torque RAM) and PRAM (phase change RAM), have drawn a lot of attention due to their low leakage and high density. However, both NVMs suffer from high write latency and limited endurance problems. To overcome these problems, the SRAM/NVM hybrid cache architecture has been proposed, and the write pressure on NVM can be mitigated with appropriate write management policy. Moreover, many wear leveling techniques have been proposed to extend the lifetime of NVM in the hybrid cache. In this paper, we proposed a hybrid cache design that includes SRAM cache, STT-RAM cache, and STT-RAM/SRAM hybrid cache banks for CMP (chip multi-processors) architecture. We also propose a partition-level wear leveling scheme and access-aware policies to mitigate unbalanced wear-out of STT-RAM lines within a partition and among different cache partitions. Experimental results show that, our proposed scheme and policies can achieve an average of 89 times improvement in cache lifetime and are able to save 58% power consumption compared to SRAM cache.
AB - In recent years, NVM (non-volatile memory) technologies, such as STT-RAM (spin transfer torque RAM) and PRAM (phase change RAM), have drawn a lot of attention due to their low leakage and high density. However, both NVMs suffer from high write latency and limited endurance problems. To overcome these problems, the SRAM/NVM hybrid cache architecture has been proposed, and the write pressure on NVM can be mitigated with appropriate write management policy. Moreover, many wear leveling techniques have been proposed to extend the lifetime of NVM in the hybrid cache. In this paper, we proposed a hybrid cache design that includes SRAM cache, STT-RAM cache, and STT-RAM/SRAM hybrid cache banks for CMP (chip multi-processors) architecture. We also propose a partition-level wear leveling scheme and access-aware policies to mitigate unbalanced wear-out of STT-RAM lines within a partition and among different cache partitions. Experimental results show that, our proposed scheme and policies can achieve an average of 89 times improvement in cache lifetime and are able to save 58% power consumption compared to SRAM cache.
UR - http://www.scopus.com/inward/record.url?scp=84878206095&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84878206095&partnerID=8YFLogxK
U2 - 10.1145/2483028.2483052
DO - 10.1145/2483028.2483052
M3 - Conference contribution
AN - SCOPUS:84878206095
SN - 9781450319027
T3 - Proceedings of the ACM Great Lakes Symposium on VLSI, GLSVLSI
SP - 19
EP - 24
BT - GLSVLSI 2013 - Proceedings of the ACM International Conference of the Great Lakes Symposium on VLSI
T2 - 23rd ACM International Conference of the Great Lakes Symposium on VLSI, GLSVLSI 2013
Y2 - 2 May 2013 through 3 May 2013
ER -