Power and Performance Trade-Offs in Contemporary DRAM System Designs for Multicore Processors

被引:24
作者
Zheng, Hongzhong [1 ]
Zhu, Zhichun [1 ]
机构
[1] Univ Illinois, Dept Elect & Comp Engn, Chicago, IL 60607 USA
基金
美国国家科学基金会;
关键词
Multicore processors; DRAM systems; power; performance;
D O I
10.1109/TC.2010.108
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
DRAM memory is playing an increasingly important role in the overall power profile of latest-generation servers with multicore processors. With many power saving techniques adopted into processor design, memory power consumption can now exceed processor power consumption when a system runs memory-intensive workloads. There is an urgent need to fully evaluate the memory power profile of contemporary DRAM memories and to re-investigate DRAM memory designs, configurations, and optimizations from both power and performance perspectives. This study fills the gap by studying the performance and power consumption of multicore systems with DDR3 memory under different configurations. It includes comprehensive results regarding memory power breakdown, including background, operation, read/write, and I/O power, as well as performance. Comparisons with DDR2 and FB-DIMM are also included. The results show clearly that DRAM system configurations, including page policy, power mode, device configuration, burst length, channel organization, and the selection of DRAM technology, affects the memory power consumption significantly besides the performance. The optimal choice of some configurations is application-dependent, suggesting that reconfigurable or hybrid configurations are worth further studies.
引用
收藏
页码:1033 / 1046
页数:14
相关论文
共 39 条
[31]   DMA-aware memory energy management [J].
Pandey, Vivek ;
Jiang, Weihang ;
Zhou, Yuanyuan ;
Bianchini, Ricardo .
TWELFTH INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE, PROCEEDINGS, 2006, :134-+
[32]  
Rixner S, 2000, PROCEEDING OF THE 27TH INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE, P128, DOI [10.1145/342001.339668, 10.1109/ISCA.2000.854384]
[33]   Automatically characterizing large scale program behavior [J].
Sherwood, T ;
Perelman, E ;
Hamerly, G ;
Calder, B .
ACM SIGPLAN NOTICES, 2002, 37 (10) :45-57
[34]  
SNAVELY A, 2002, P 2002 ACM SIGMETRIC, P66, DOI [10.1145/511334, DOI 10.1145/511334]
[35]  
Wang D., 2005, Ph.D. Thesis
[36]   A permutation-based page interleaving scheme to reduce row-buffer conflicts and exploit data locality [J].
Zhang, Z ;
Zhu, ZC ;
Zhang, XD .
33RD ANNUAL IEEE/ACM INTERNATIONAL SYMPOSIUM ON MICROARCHITECTURE: MICRO-33 2000, PROCEEDINGS, 2000, :32-41
[37]   Mini-Rank: Adaptive DRAM Architecture for Improving Memory Power Efficiency [J].
Zheng, Hongzhong ;
Lin, Jiang ;
Zhang, Zhao ;
Gorbatov, Eugene ;
David, Howard ;
Zhu, Zhichun .
2008 PROCEEDINGS OF THE 41ST ANNUAL IEEE/ACM INTERNATIONAL SYMPOSIUM ON MICROARCHITECTURE: MICRO-41, 2008, :210-+
[38]   Dynamic tracking of page miss ratio curve for memory management [J].
Zhou, P ;
Pandey, V ;
Sundaresan, J ;
Raghuraman, A ;
Zhou, YY ;
Kumar, S .
ACM SIGPLAN NOTICES, 2004, 39 (11) :177-188
[39]  
Zhu ZC, 2005, INT S HIGH PERF COMP, P213