Cache Energy Management through Dynamic Reconfiguration Approach in Opto-Electrical NoC

Multi/Many-core architectures will be the popular platform for future system design. Recent investigations show that the hybrid optical-electrical interconnection network can be an appropriate alternative to the traditional electrical NoC. Undoubtedly, memory wall is one of the most important challenges of multi/many-core systems which can somehow be alleviated thanks to hierarchical memory structure. Cache subsystem plays an essential role in increasing the efficiency of the memory structure. In this paper, after exploring the effect of cache subsystem's parameters in a many-core platform with opto-electrical interconnect, we propose a mechanism to increase the cache energy efficiency. The simulation results of the proposed approach show that after applying our method, the energy efficiency parameter improves by 17% and 23% in SPLASH2 and PARSEC benchmarks, respectively.

[1]  Zhe Wang,et al.  Memory Access Analysis of Many-core System with Abundant Bandwidth , 2015, 2015 IEEE 9th International Symposium on Embedded Multicore/Many-core Systems-on-Chip.

[2]  A. Zain,et al.  Optical Routers based on Microring Resonator for Optical Networks-on-chip , 2016 .

[3]  Jaume Abella,et al.  Power efficient data cache designs , 2003, Proceedings 21st International Conference on Computer Design.

[4]  Luigi Carro,et al.  Exploring Cache Size and Core Count Tradeoffs in Systems with Reduced Memory Access Latency , 2016, 2016 24th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP).

[5]  Jürgen Becker,et al.  Adaptive Cache Structures , 2016, ARCS.

[6]  Siamak Mohammadi,et al.  Clustering Effects on the Design of Opto-Electrical Network-on-Chip , 2016, 2016 24th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP).

[7]  Prabhat Mishra,et al.  Reliability and energy-aware cache reconfiguration for embedded systems , 2016, 2016 17th International Symposium on Quality Electronic Design (ISQED).

[8]  Sudhakar M. Reddy,et al.  Cache size selection for performance, energy and reliability of time-constrained systems , 2006, Asia and South Pacific Conference on Design Automation, 2006..

[9]  André C. Nácul,et al.  Dynamic voltage and cache reconfiguration for low power , 2004, Proceedings Design, Automation and Test in Europe Conference and Exhibition.

[10]  Doug Burger,et al.  An Adaptive Cache Structure for Future High-Performance Systems , 2002 .

[11]  Huaxi Gu,et al.  A hierarchical cluster-based optical network-on-chip , 2010, 2010 2nd International Conference on Future Computer and Communication.

[12]  Björn Franke,et al.  Cooperative partitioning: Energy-efficient cache partitioning for high-performance CMPs , 2012, IEEE International Symposium on High-Performance Comp Architecture.

[13]  Shaahin Hessabi,et al.  ONC3: All-Optical NoC Based on Cube-Connected Cycles with Quasi-DOR Algorithm , 2012, 2012 15th Euromicro Conference on Digital System Design.

[14]  Frank Vahid,et al.  A highly configurable cache for low energy embedded systems , 2005, TECS.

[15]  Qinfen Hao,et al.  LioeSim: A Network Simulator for Hybrid Opto-Electronic Networks-on-Chip Analysis , 2014, Journal of Lightwave Technology.

[16]  George Kurian,et al.  ATAC: A 1000-core cache-coherent processor with on-chip optical network , 2010, 2010 19th International Conference on Parallel Architectures and Compilation Techniques (PACT).

[17]  M. Channoufi,et al.  Toward all optical interconnections in chip multiprocessor , 2011, ICM 2011 Proceeding.

[18]  Michal Lipson,et al.  WDM-compatible mode-division multiplexing on a silicon chip , 2014, Nature Communications.

[19]  Zhao Zhang,et al.  MASTER: A Multicore Cache Energy-Saving Technique Using Dynamic Cache Reconfiguration , 2014, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[20]  George Kurian,et al.  Graphite: A distributed parallel simulator for multicores , 2010, HPCA - 16 2010 The Sixteenth International Symposium on High-Performance Computer Architecture.

[21]  Tianzhou Chen,et al.  Energy-efficient Hybrid Optical-Electronic Network-on-Chip for Future Many-core Processors , 2014 .