期刊文献+

一种128位高性能全流水浮点乘加部件 被引量:4

A High Performance Pipeline Architecture of 128bit Floating-point Fused Multiply-add Unit
下载PDF
导出
摘要 高精度的浮点乘加融合(FMA)部件一直是高性能微处理器设计追求的目标。提出了一种128位精度全流水FMA体系结构,采用10级平衡流水线,重点对超宽位的乘法器、加法器、前导零预测和规格化进行了流水优化。设计综合的结果表明,基于SMIC0.13μm工艺,该结构频率可以达到465MHz,比现有128位FMA性能提高了130%;在TSMC65nm工艺下,该结构的频率可达到1.075GHz,基本满足高性能计算的要求。 FMA(Fused Multiply-Add) with high precision is required in high performance microprocessors. A new 10 stages pipelined architecture of 128bit FMA is propesed. In this architecture, multiply, adder, IZA(leading Zero Anticipator) and normalization with large width data-paths were partitioned and optimized carefully to balance the latency at every pipeline stage. After designed and synthesized with SMIC 0.13um technology, the frequency of the FMA can reach 465MHz, which is about 130% better than previous 128hit FMA. Furthermore, its frequency can reach 1.075GHz with TSMC 65nm technology, which basically meets the requirements of the high performance computation.
出处 《国防科技大学学报》 EI CAS CSCD 北大核心 2010年第2期56-60,共5页 Journal of National University of Defense Technology
基金 国家自然科学基金重点资助项目(90707003)
关键词 浮点乘加融合 前导零预测 高性能微处理器 Fused Multlply-Add( FMA ) Leading Zero Anticipator(IZA) high performance microprocessor
  • 相关文献

参考文献1

二级参考文献10

  • 1梅小露.浮点乘加部件中三操作数前导1预测算法的设计[J].微电子学与计算机,2005,22(12):16-20. 被引量:8
  • 2Rabaey J M, Chandrakasan A, Nikolic B. Digital Integrated Circuits: A Design Perspective[M]. 2nd ed. 北京:清华大学出版社,2004:586-594.
  • 3Bruguera J D, Lang T. Leading-One Prediction with Concurrent Position Correction [J]. IEEE Trans on Computers, 1994,48(10) : 1083-1097.
  • 4Tomas L, Javier B D. Floating-Point Multiply-Add Fused with Reduced Latency[J]. IEEE Trans on Computers, 2004, 53(8):988-1003.
  • 5张峰,黎铁军,徐炜遐.一种浮点乘加部件的测试方法[C]//第十一届计算机工程与工艺学术年会论文集,2007:322-324.
  • 6凌志强.支持并行整数乘的双通路浮点融合乘加结构的研究与实现:[工学硕士学位论文][D].长沙:国防科技大学,2005.
  • 7ANSI/IEEE Std 754-1985, IEEE Standard for Binary Floating-Point Arithmetic[S].
  • 8Sun Mierosystems Inc. OpenSpareTM T1 Microaxchiteeture Specification. Part No. 919-6650-10[S].
  • 9Montoye R K, Hokenek E, Runyon S L. Design of the IBM RISC System/6000 Floating-Point Execution Unit[J]. IBM Journal of Research and Development, 1990, 34 (1), 59-70.
  • 10Gerwig G, Wetter H, Schwarz E M, et al. The IBM eServer z990 Floating-Point Unit[J]. IBM Journal of Research and Development, 2004,48(3-4):311-322.

共引文献3

同被引文献43

  • 1AMIRA A, BENSAAI.I F. An FPGA based parameterizable system for matrix product inq,lemelltation[ C ]//IEEE Work- shop on Signal Processing Systems (SPIS'02). San Diego, 2002 : 75-79.
  • 2JANG J, CHOI S, PRASANNA V K K. Area and lime effi- cient implementations of matrix multiplication on FPGAs [C ]//2002 IEEE International Conference on Field Pro- grammabIe Technology. Seoul, Korea, 2002: 93-100.
  • 3CAMPBELL S J, KHATRI S P. Resource and delay effi- cient matrix multiplication using newer FPGA devices[ C ]// Proceedings of the 16th ACM Great Lakes Symposium on VLSI. Philadelphia, USA, 2006: 308-311.
  • 4LEISERSON C, ROSE F, SAXE J. Optimizing synchronouscircuitry by retiming [ C ]//Proceedings of the 3rd Cahech Conference On VLS1. Rockville, Maryland, 1983 : 87-116.
  • 5SU Ming, ZHOU Lili. Maximizing the throughput-area effi- ciency of fully-parallel low-density parity-eheek decoding with C-slow retiming and asynchronous deep pipellning [ C]//The 25th International Conference on Computer De- sign. Washington, DC, USA, 2007: 93-100.
  • 6肖字,王建业,张伟.基于IP核的数选式浮点矩阵相乘设计[J].集成电路应用,2011,37(6):52-55.
  • 7Bailey D H. High-precision floating-point arithmetic in scientific computation [J]. Computing in Science and Engineering, 2005,7 (3) :54-61.
  • 8IEEE Computer Society. IEEE Standard for Floating-Point A- rithrnetic[S3. IEEE Standard 754-2008, 3 Park Avenue New York, NY 10016-5997, USA, August 2008.
  • 9Akkas A,Sehulte M J. Dual-Mode Floating-Point Multiplier Ar- chitectures with Parallel Operations [J]. Journal of Systems Ar- chitecture, 2006,52 : 549-562.
  • 10Akkas A. Dual-Mode Quadruple Precision Floating Point Adder [C] 9th Euromiero Conference on Digital System Desigr. 2006:211-220.

引证文献4

二级引证文献9

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部