The Community for Technology Leaders
RSS Icon
Issue No.05 - May (2010 vol.59)
pp: 694-706
Ghassem Jaberipur , Shahid Beheshti University and Institute for Research in Fundamental Sciences (IPM), Tehran
Behrooz Parhami , University of California, Santa Barbara
Saeid Gorgin , Shahid Beheshti University, Tehran
Due to the widespread use and inherent complexity of floating-point addition, much effort has been devoted to its speedup via algorithmic and circuit techniques. We propose a new redundant-digit representation for floating-point numbers that leads to computation speedup in two ways: 1) Reducing the per-operation latency when multiple floating-point additions are performed before result conversion to nonredundant format and 2) Removing the addition associated with rounding. While the first of these advantages is offered by other redundant representations, the second one is unique to our approach, which replaces the power- and area-intensive rounding addition by low-latency insertion of a rounding two-valued digit, or twit, in a position normally assigned to a redundant twit within the redundant-digit format. Instead of conventional sign-magnitude representation, we use a sign-embedded encoding that leads to lower hardware redundancy, and thus, reduced power dissipation. While our intermediate redundant representations remain incompatible with the IEEE 754-2008 standard, many application-specific systems, such as those in DSP and graphics domains, can benefit from our designs. Description of our radix-16 redundant representation and its addition algorithm is followed by the architecture of a floating-point adder based on this representation. Detailed circuit designs are provided for many of the adder's critical subfunctions. Simulation and synthesis based on a 0.13 \mu{\rm m} CMOS standard process show a latency reduction of 15 percent or better, and both area and power savings of around 58 percent, compared with the best designs reported in the literature.
Adder/subtractor, computer arithmetic, floating point, redundant format, rounding, signed-digit number system.
Ghassem Jaberipur, Behrooz Parhami, Saeid Gorgin, "Redundant-Digit Floating-Point Addition Scheme Based on a Stored Rounding Value", IEEE Transactions on Computers, vol.59, no. 5, pp. 694-706, May 2010, doi:10.1109/TC.2009.152
[1] B. Parhami, Computer Arithmetic: Algorithms and Hardware Designs, second ed. Oxford Univ. Press, 2010.
[2] Institute of Electrical and Electronics Engineers, IEEE Standard for Binary Floating-Point Arithmetic, ANSI/IEEE Std 754-1985, Aug. 1985.
[3] Institute of Electrical and Electronics Engineers, IEEE Standard for Floating-Point Arithmetic, IEEE Std 754-2008, Aug. 2008.
[4] T.-J. Lin, H.-Y. Lin, C.-M. Chao, and C.-W. Liu, "A Compact DSP Core with Static Floating-Point Arithmetic," J. VLSI Signal Processing, vol. 42, no. 2, pp. 127-138, Feb. 2006.
[5] R. Oslejsek and J. Sochor, "Generic Graphics Architecture," Proc. Conf. Theory and Practice of Computer Graphics, pp. 105-112, 2003.
[6] P.M. Farmwald, "On the Design of High Performance Digital Arithmetic Units," PhD thesis, Stanford Univ., Aug. 1981.
[7] A.M. Nielsen, D.W. Matula, C.N. Lyu, and G. Even, "An IEEE Compliant Floating-Point Adder that Conforms with the Pipelined Packet-Forwarding Paradigm," IEEE Trans. Computers, vol. 49, no. 1, pp. 33-47, Jan. 2000.
[8] S.F. Oberman and M.J. Flynn, "Reducing the Mean Latency of Floating-Point Addition," Theoretical Computer Science, vol. 196, pp. 201-214, 1998.
[9] P.-M. Seidel and G. Even, "On the Design of Fast IEEE Floating-Point Adders," Proc. 15th IEEE Symp. Computer Arithmetic, pp. 184-194, 2001.
[10] H.A.H. Fahmy and M.J. Flynn, "The Case for a Redundant Format in Floating-Point Arithmetic," Proc. 16th IEEE Symp. Computer Arithmetic, pp. 95-102, 2003.
[11] G. Jaberipur, B. Parhami, and M. Ghodsi, "Weighted Two-Valued Digit-Set Encodings: Unifying Efficient Hardware Representation Schemes for Redundant Number Systems," IEEE Trans. Circuits and Systems I, vol. 52, no. 7, pp. 1348-1357, July 2005.
[12] H.A.H. Fahmy, "A Redundant Digit Floating Point System," PhD thesis, Stanford Univ., June 2003.
[13] S.G. Campbell, "Floating-Point Operations," Planning a Computer System, W. Buchholz, ed., pp. 92-106, McGraw-Hill, 1962.
[14] S.F. Anderson, J.G. Earle, R.E. Goldschmidt, and D.M. Powers, "The IBM System/360 Model 91: Floating-Point Execution Unit," IBM J. Research and Development, vol. 11, no. 1, pp. 34-53, 1967.
[15] P.-M. Seidel and G. Even, "Delay-Optimized Implementation of IEEE Floating-Point Addition," IEEE Trans. Computers, vol. 53, no. 2, pp. 97-113, Feb. 2004.
[16] S. Vassiliadis, D.S. Lemon, and M. Putrino, "S/370 Sign-Magnitude Floating-Point Adder," IEEE J. Solid-State Circuits, vol. 24, no. 4, pp. 1062-1070, Aug. 1989.
[17] D.W. Matula and A.M. Nielsen, "Pipelined Packet-Forwarding Floating Point: I. Foundations and a Rounder," Proc. 13th IEEE Symp. Computer Arithmetic, pp. 140-147, 1997.
[18] G. Jaberipur, B. Parhami, and M. Ghodsi, "An Efficient Universal Addition Scheme for All Hybrid-Redundant Representations with Weighted Bit-Set Encoding," J. VLSI Signal Processing, vol. 42, no. 2, pp. 149-158, Feb. 2006.
[19] M.C. Mekhallalati and M.K. Ibrahim, "New High Radix Maximally Redundant Signed Digit Adder," Proc. Int'l Symp. Circuits and Systems, vol. 1, pp. 459-462, 1999.
[20] G. Jaberipur and S. Gorgin, "A Nonspeculative One-Step Maximally Redundant Signed Digit Adder," Comm. Computer and Information Science, vol. 6, pp. 235-242, 2008.
[21] G. Jaberipur and B. Parhami, "Constant-Time Addition with Hybrid-Redundant Numbers: Theory and Implementations," Integration: The VLSI J., vol. 41, no. 1, pp. 49-64, Jan. 2008.
[22] M. Daumas and D.W. Matula, "Further Reducing the Redundancy of a Notation over a Minimally Redundant Digit Set," J. VLSI Signal Processing, vol. 33, pp. 7-18, 2003.
[23] H.A.H. Fahmy and M.J. Flynn, "Rounding in Redundant Digit Floating Point Systems," Proc. SPIE Conf. Algorithms, Architectures, and Devices, Aug. 2003.
288 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool