Please use this identifier to cite or link to this item:
標題: 利用動態測錄的方式降低分支目標暫存器的功率消耗
Reducing BTB Energy Consumption Using Dynamic Profiling
作者: 羅孔男
Lo, Kung-Nan
關鍵字: branch prediction
branch target buffer
出版社: 資訊科學系所
引用: [1] Sung Woo Chung, Sung-Bae Park, “A Low Power Branch Predictor to Selectively Access the BTB”, Asia-Pacific Computer Systems Architecture Conference ,2004 [2] John L. Hennessy, and David A. patterson, “Computer Organization & Design for hardware/software interface”, second edition, Morgan Kaufmann Publishers, 2000 [3] Harold S. Stone, “High-Performance Computer Architecture second edition”, Addison-Wesley series in Electrical and Computer Engineering, 1992. [4] Doug Burger, Todd M. Austin, “SimpleScalar tool set,Version2.0”, available at [5] Standard Performance Evaluation Corp., SPEC CPU2000 Benchmarks, available at [6] Manne S., Klauser A., Grunwald D., “Pipeline gating: speculation control for energy reduction”, Computer Architecture. Proceedings. The 25th Annual International Symposium on, 1998 [7] Parikh D., Skadron K., Zhang Y. and Stan M, “Power-aware branch prediction: characterization and design” In Computers, IEEE Transactions on , Feb 2004 Page(s):168 - 186 [8] Petrov P. and Orailoglu A, “Low-power branch target buffer for application-specific embedded processors” In Digital System Design, 2003. Proceedings. Euromicro Symposium on, 1-6 Sept. 2003 Page(s):158 - 165 [9] S.J.E.Wilton, and Jouppi, N.P., “CACTI: an enhanced cache access and cycle time model”, Digital Object Identifier, May 1996 Page(s):677 - 688 [10] Veljko Milutinovic, “Surviving The Design Of Microprocessor Systems”, Wiley Inter-science Publication, John Willy and Sons, 2000. [11] Patrick H. and Stakem, “A Practitioner's Guide to RISC Microprocessor Architecture”, Wiley Inter-science Publication, John Willy and Sons, 1996. [12] D. Chaver, L. Pinuel, M. Prieto, F. Tirado, and Huang M.C, “Branch prediction on demand: an energy-efficient solution [microprocessor architecture]”. In Low Power Electronics and Design, 25-27 Aug. 2003 Page(s):390 - 395 [13] D. Parikh, K. Skadron, Yan Zhang, M. Barcella, and Stan M.R, “Power issues related to branch prediction.” In High-Performance Computer Architecture Proceedings. Eighth International Symposium on, 2-6 Feb. 2002 Page(s):233 - 244 [14] S. Vlaovic, E.S. Davidson, and G.S. Tyson, “Improving BTB performance in the presence of DLLs”, Microarchitecture, MICRO-33. Proceedings. 33rd Annual IEEE/ACM International Symposium on, 10-13 Dec. 2000 Page(s):77 - 86 [15] M.C. Huang, D. Chaver, L. Pinuel, M. Prieto, and F. Tirado, “Customizing the branch predictor to reduce complexity and energy consumption”, Micro, IEEE, Sept.-Oct. 2003 Page(s):12 - 25 [16] E.F. Weglarz, K.K. Saluja, and M.H. Lipasti, “Minimizing energy consumption for high-performance processing”, Design Automation Conference, 2002. Proceedings of ASP-DAC 2002. 7th Asia and South Pacific and the 15th International Conference on VLSI Design. Proceedings, Jan. 2002 Page(s):199 - 204 [17] C. H. Perleberg, and A. J. Smith, “Branch Target Buffer Design and Optimization,” IEEE Transaction on Computers, vol. 42, iss. 4, pages 396-412, April 1993 [18] I. Bate, and R. Reutemann, ” Efficient integration of bimodal branch prediction and pipeline analysis”, Embedded and Real-Time Computing Systems and Applications, 2005. Proceedings. 11th IEEE International Conference on,17-19 Aug. 2005 Page(s):39 - 44 [19] M.C. Chang, and Y.W. Chou, “Branch prediction using both global and local branch history information”, Computers and Digital Techniques, IEE Proceedings, 2, March 2002 Page(s):33 - 38 [20] M. Evers, and Tse-Yu Yeh, ”Understanding branches and designing branch predictors for high-performance microprocessors”, Proceedings of the IEEE, 11, Nov. 2001 Page(s):1610 - 1620 [21] D.R. Kaeli, and P.G. Emma, “Improving the accuracy of history-based branch prediction”, Computers, IEEE Transactions on April 1997 Page(s):469 - 472 [22] S. Jourdan, Tse-Hao Hsing, J. Stark, and Y.N. Patt, “The effects of mispredicted-path execution on branch prediction structures”, Parallel Architectures and Compilation Techniques, 1996., Proceedings of the 1996 Conference on,20-23 Oct. 1996 Page(s):58 - 67 [23] Jian Huang, and D.J. Lilja, “Exploiting basic block value locality with block reuse”, High-Performance Computer Architecture, 1999. Proceedings. Fifth International Symposium On,9-13 Jan. 1999 Page(s):106 - 114 [24] T. Austin, E. Larson, and D. Ernst. “Simplescalar: An infrastructure for computer system modelin” IEEE Transactions on Computers, Feb. 2002 35(2):59—67 [25] John L. Hennessy, and David A. patterson, “Computer Architecture A Quantitative Approach”, second edition, Morgan Kaufmann Publishers, 2003
摘要: 在現今處理器的設計上,為了提升效能,管線階層的分割越來越細,長度也跟著加長,當分支危障發生時所需付出的代價也將隨之上升,因此解決分支危障的問題變得越來越重要。用來改善此問題的分支預測的架構越顯重要,雖然需要付出額外的硬體代價,其所帶來的效能的提升是很值得的,但是其功率的消耗也是不可忽視,在一些處理器上所消耗的功率甚至佔整個處理器的功率消耗高達10%或者更多。 本篇論文利用dynamic profiling的方式追蹤發生的分支指令,加上簡單的累加器、estimator和暫存器,使得只有在可能為發生的分支指令時才去查詢分支目標暫存器,藉此減少大量且多餘的查詢來減少分支目標暫存器的消耗功率以達到低功率的設計,其效果平均達88.65%的功率節省在查詢分支預測暫存器上。
As the pipeline length increases, the accuracy in a branch- prediction gets critical to overall performance. The power consumption of branch prediction is larger to overall processor's power. In some case, it's more over 10% of all processor's power consumption. In this paper, we use dynamic profiling to trace the taken branch for our low power design. We use some accumulators, registers, estimator to decreasing the numbers of lookup BTB in our design. The power saving of our design on lookup BTB is 88.65% on average.
其他識別: U0005-1406200614065400
Appears in Collections:資訊科學與工程學系所



Show full item record
TAIR Related Article

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.