A Floating-Point Fused Dot-Product Unit

被引：26
|
作者：
Saleh, Hani H. [1 ]
Swartzlander, Earl E., Jr. [2 ]
机构：
[1] Intel Corp, 5000 Plaza Lake Blvd, Austin, TX 78746 USA
[2] Univ Texas Austin, ECE Dept, Austin, TX 78712 USA
来源：
2008 IEEE INTERNATIONAL CONFERENCE ON COMPUTER DESIGN | 2008年
关键词：
D O I：
10.1109/ICCD.2008.4751896
中图分类号：
TP3 [计算技术、计算机技术];
学科分类号：
0812 ;
摘要：
A floating-point fused dot-product unit is presented that performs single-precision floating-point multiplication and addition operations on two pairs of data in a time that is only 150% the time required for a conventional floating-point multiplication. When placed and routed in a 45nm process, the fused dot-product unit occupied about 70% of the area needed to implement a parallel dot-product unit using conventional floating-point adders and multipliers. The speed of the fused dot-product is 27% faster than the speed of the conventional parallel approach. The numerical result of the fused unit is more accurate because one rounding operation is needed versus at least three for other approaches.
引用
收藏
页码：427 / +
页数：2
相关论文
共 50 条
  • [21] A High Speed Floating Point Dot Product Unit
    Gupta, Akash Kumar
    Biswal, Birendra
    [J]. PROCEEDINGS OF THE 2014 INTERNATIONAL CONFERENCE ON ISSUES AND CHALLENGES IN INTELLIGENT COMPUTING TECHNIQUES (ICICT), 2014, : 314 - 318
  • [22] Fused Floating-Point Arithmetic for DSP
    Swartzlander, Earl E., Jr.
    Saleh, Hani H.
    [J]. 2008 42ND ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS AND COMPUTERS, VOLS 1-4, 2008, : 767 - +
  • [23] Dual-Path Architecture of Floating-Point Dot Product Computation
    Yao Tao
    An Jianfeng
    Gao Deyuan
    Fan Xiaoya
    [J]. 2011 INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND NETWORK TECHNOLOGY (ICCSNT), VOLS 1-4, 2012, : 2272 - 2276
  • [24] Floating-point fused multiply-add: Reduced latency for floating-point addition
    Bruguera, JD
    Lang, T
    [J]. 17TH IEEE SYMPOSIUM ON COMPUTER ARITHMETIC, PROCEEDINGS, 2005, : 42 - 51
  • [25] Implementation of Fused Floating Point Four Term Dot Product Unit using Modified Booth Algorithm
    Srilakshmi, Pinapothu
    Latha, P. Pushpa
    [J]. 2017 IEEE INTERNATIONAL CONFERENCE ON POWER, CONTROL, SIGNALS AND INSTRUMENTATION ENGINEERING (ICPCSI), 2017, : 2181 - 2186
  • [26] An efficient multiple precision floating-point Multiply-Add Fused unit
    Manolopoulos, K.
    Reisis, D.
    Chouliaras, V. A.
    [J]. MICROELECTRONICS JOURNAL, 2016, 49 : 10 - 18
  • [27] Fused Floating-Point Two-term Sum-of-Squares Unit
    Min, Jae Hong
    Swartzlander, Earl E., Jr.
    [J]. PROCEEDINGS OF THE 2013 IEEE 24TH INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS (ASAP 13), 2013, : 147 - 152
  • [28] Peak Performance Model for a Custom Precision Floating-Point Dot Product on FPGAs
    Muecke, Manfred
    Lesser, Bernd
    Gansterer, Wilfried N.
    [J]. EURO-PAR 2010 PARALLEL PROCESSING WORKSHOPS, 2011, 6586 : 399 - 406
  • [29] Multi-functional floating-point MAF designs with dot product support
    Gok, Mustafa
    Ozbilen, Metin Mete
    [J]. MICROELECTRONICS JOURNAL, 2008, 39 (01) : 30 - 43
  • [30] FFT Implementation with Fused Floating-Point Operations
    Swartzlander, Earl E., Jr.
    Saleh, Hani H. M.
    [J]. IEEE TRANSACTIONS ON COMPUTERS, 2012, 61 (02) : 284 - 288
12345