计算机科学 ›› 2014, Vol. 41 ›› Issue (7): 36-39.doi: 10.11896/j.issn.1002-137X.2014.07.006
方娟,王帅,于璐
FANG Juan,WANG Shuai and YU Lu
摘要: 如何提高多核处理器的性能和降低多核处理器中Cache的功耗已经成为下一代多核处理器的研究热点。为了降低片上多核处理器的功耗,基于路适应算法可以采用一种新的动态划分机制,该机制主要由路分配模块和动态功耗控制模块组成。路分配模块在程序运行过程中根据处理器核所运行线程的工作集的大小调整处理器核所分配的Cache路。动态功耗控制模块利用程序运行的局部性原理,将处理器核所运行线程的工作空间控制在少数Cache路中。关闭剩余的Cache路,从而达到降低Cache功耗的目的。该机制使用Simics全系统模拟平台模拟多核处理器,并用SpecOMP测试集测试了系统的性能和功耗。与传统的 Cache(Conventional L2Cache,C-L2)相比,其IPC提高了9.27%,功耗降低了10.95%。
[1] Herrero E,Gonz′alez J,Canal R.Distributed cooperative caching[C]∥Proceedings of the 17th International Conference on Pa-rallel Architectures and Compilation Techniques.2008:134-143 [2] Kotera I.Power-Aware Dynamic Cache Partitioning for CMPs H[C]∥Transactions on High-Performance Embedded Architectures and Compilers III.2011:135-153 [3] Sinharoy B,KallaR N,Tendler J M.Power 5system micro-architecture[M].IBM J.Res Dev,2005,9:505-521 [4] Kim S,Chandra D,Solihin Y.Fair Cache sharing and partitioning in a chip multiprocessor architecture[C]∥Proc.of PACT 2004.Antibes,Juanles-Pins,France,2004:111-122 [5] Sui Xiu-feng,Wu Jun-min,Chen Guo-liang,et al.AugmentingCache Partitioning with Thread-Aware InsertionIPromotion Policies to Manage Shared Caches[C]∥Proceedings of the 7th ACM International Conference on Computing Frontiers.2004:79-80 [6] Meng Y,Sherwood T,Kastner.Exploring the limits of leakage power reduction in Caches[J].ACM Trans.Archit.Code Optim.,2005,2:221-246 [7] Janapsatya A,Parameswaran S,Ignjatovic A.HitME:Low power Hit MEmory buffer for embedded systems[C]∥Asia and South Pacific Design Automation Conference.2009:335-340 [8] Tsai Y-Y,Chen C-H.Energy-Efficient Trace Reuse Cache forEmbedded Processors[J].IEEE Transaction On Very Large Scale Integration(VLSI)System,2010,19:1681-1694 [9] Datta K A M A S.Energy efficient i-Cache using multiple line buffers with prediction[J].IET Comput.Digit.Tech.,2008,2(5):355-362 [10] Ali K,Aboelaze M.Energy efficient I-Cache using multiple line buffers with prediction[J].Computers and Digital Techniques,IET,2008(2):355-362 [11] Inoue K.Way-Predicting Set-Associative Cache for High Per-formance and Low Energy Consumption[C]∥Proceedings.1999International Symposium on Low Power Electronics and Design 1999(Cat.No.99TH8477).1999:273-275 [12] Suo Guang,Yang Xue-jun,Liu Guang-hui,et al.IPC-BasedCache Partitioning:An IPC-Orient Dynamic Shared Cache [C]∥International Conference on Convergence and Hybrid Information Technology.Washington DC:IEEE Computer Society,2008:399-406 [13] Kim S,Chandra D,Solin Y.Fair Cache sharing and partitioning in a chip multiprocessor architecture[C]∥Proc.of PDCT 2004.Antibes,Juanles-PINS,France IEEE,2004:111-122 [14] Zhang C,Yang J,Vahid F.Low static-power frequent-value data Caches[C]∥The Design,Automation and Test in Europe Conference and Exhibition.Paris,France,2004,1:214-219 [15] Qureshi M K,Patt Y N.Utility based Cache partitioning:a low overhead,high performance,runtime mechanism to partition shared Caches[C]∥Proc.of the 39th Annual IEEE/ACM Int Symp on Microarchitecture.Orlando.Florida,USA:IEEE,2006:423-432 [16] Kobayashi H,Kotera I,Takizawa H.Locality analysis to control dynamicallyway-adaptable Caches[J].SIGARCH Comput.Archit.News,2005,33(3):25-32 [17] Kobayashi H,Kotera I,Takizawa H.Locality analysis to control dynamically way-adaptable Caches[J].SIGARCH Comput.Archit.News,2005,3:25-32 [18] Christensson M,Eskilson.Simics:a full system simulation platform[J].IEEE Computer,2002(3):50-58 |
No related articles found! |
|