TY - GEN
T1 - Automatic feedback control of shared hybrid caches in 3D chip multiprocessors
AU - Sharifi, Akbar
AU - Kandemir, Mahmut
PY - 2011
Y1 - 2011
N2 - 3D integration enables building caches from different types of technologies such as SRAM, Magnetic RAM (MRAM), DRAM, and Phase-change RAM (PRAM). Hybrid cache architectures (HCAs) have been proposed to take advantage of the benefits offered by these types of technologies. Employing this novel cache architecture to build shared caches in chip multiprocessors (CMPs) can lead to significant performance and power consumption improvements. In this paper, we focus on a 3D CMP design in which the shared last level L2 cache is composed of an MRAM layer and an SRAM layer stacked upon the processing cores and present a control theory centric approach designed to partition this shared hybrid L2 cache space dynamically among concurrently running applications in order to satisfy the application-level performance QoS targets. At each time interval, the two layers of the hybrid L2 cache are partitioned, based on the cache demands made by the controllers of the applications, to satisfy the specified performance targets. We evaluate our feedback control based scheme using various workloads. Our experimental evaluation shows that the proposed scheme is able to satisfy the specified performance QoS in most of the tested cases, by partitioning the hybrid cache space of the 3D CMP among co-runner applications.
AB - 3D integration enables building caches from different types of technologies such as SRAM, Magnetic RAM (MRAM), DRAM, and Phase-change RAM (PRAM). Hybrid cache architectures (HCAs) have been proposed to take advantage of the benefits offered by these types of technologies. Employing this novel cache architecture to build shared caches in chip multiprocessors (CMPs) can lead to significant performance and power consumption improvements. In this paper, we focus on a 3D CMP design in which the shared last level L2 cache is composed of an MRAM layer and an SRAM layer stacked upon the processing cores and present a control theory centric approach designed to partition this shared hybrid L2 cache space dynamically among concurrently running applications in order to satisfy the application-level performance QoS targets. At each time interval, the two layers of the hybrid L2 cache are partitioned, based on the cache demands made by the controllers of the applications, to satisfy the specified performance targets. We evaluate our feedback control based scheme using various workloads. Our experimental evaluation shows that the proposed scheme is able to satisfy the specified performance QoS in most of the tested cases, by partitioning the hybrid cache space of the 3D CMP among co-runner applications.
UR - http://www.scopus.com/inward/record.url?scp=79955047392&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79955047392&partnerID=8YFLogxK
U2 - 10.1109/PDP.2011.83
DO - 10.1109/PDP.2011.83
M3 - Conference contribution
AN - SCOPUS:79955047392
SN - 9780769543284
T3 - Proceedings - 19th International Euromicro Conference on Parallel, Distributed, and Network-Based Processing, PDP 2011
SP - 393
EP - 400
BT - Proceedings - 19th International Euromicro Conference on Parallel, Distributed, and Network-Based Processing, PDP 2011
T2 - 19th International Euromicro Conference on Parallel, Distributed, and Network-Based Processing, PDP 2011
Y2 - 9 February 2011 through 11 February 2011
ER -