Abstract
The purpose of this work is to present two novel architectures for inner product computation. The proposed architectures incorporate shift switching into the reconfigurable buses. Given two arrays of N elements, each consisting of m bits, our first architecture achieves a latency of O((logN + logM)ta + (logN)tb), using Nm2 basic shift switches and m2 adders assuming that broadcasting on a bus takes tb time and an addition takes ta time. The second architecture extends the first one by pipelining and achieves a throughput of one inner product per lime unit, while keeping the latency unchanged. Replacing large number of adders with simple shift switches, both architectures improve the state of the art by reducing the amount of hardware for the computation.
This author was supported by National Science Foundation under grant CCR-8909996
Preview
Unable to display preview. Download preview PDF.
References
Y. Ben-Asher, D. Peleg, R. Ramaswam, and A. Schuster, The Power of Reconfiguration, Journal of Parallel and Distributed Computing.vol.13 (1991) 139–151.
J. J. F. Cavanaugh, Digital Computer Arithmetic Design and Implementation, New York: McGraw-Hill Book Co. 1984.
H. T. Kung and C. E. Leiserson, Algorithms for VLSI Processor Arrays, Introduction to VLSI Systems, C. Mead and L. Conway, Reading MA: Addison-Wesley, 1980.
H. Li and M Maresca, Polymorphic-torus network, IEEE Trans. Comput.,38 (9), 1989.
R. Lin, Fast algorithms for the lowest common ancestor problem on a processor array with reconfigurable buses, Information Processing Letters, 40, Nov. 1991.
R. Lin, Reconfigurable Buses with Shift Switching — VLSI Radix sort, to appear in Proc. of International Conference on Parallel Processing, St. Charles, IL, August 1992.
R. Lin, and S. Olariu, A new method of inner product computation on reconfigurable buses with shift switching, TR 8-92, Dept. of Comp., Sci., Old Dominion Univ. 1992.
R. Lin, S. Olariu, J. Schwing, and J. Zhang, Sorting in O(1) time on an n × n reconfigurable mesh, to appear in Proc. of 9-th European Workshop on Parallel Computing, Barcelona, Spain, March 1992.
R. Miller, V. K. Prasanna Kumar, D. Reisis and Q.F. Stout, Mesh with reconfigurable buses, in Proc. 5th MIT Conference on Advanced Research in VLSI (1988) 163–178.
J. Rothstein, Bus automata, brains, and mental models, IEEE Trans. on Systems Man Cybernetics 18, (1988).
S. P. Smith and H. C. Torng, Design of a Fast Inner Product Processor, in Proc. IEEE 7th Symposium on Computer Arithmetic, 1985.
E. E. Swartzlander, Jr., Barry K. Gilbert, and Irving S. Reed, Inner Product Computers, IEEE Trans. Computers, C-27, 1, Jan. 1978, pp. 21–31.
E. E. Swartzlander, Jr., Computer Arithmetic Vol. 1, (IEEE CSP, CA, 1990).
J. D. Ullman, Computational Aspect of VLSI, (Computer Science Press,MD, 1983).
B. F. Wang, G.H. Chen, and F.C. Lin, Constant Time Sorting on a processing array with a rcconfigurablc bus system, Information Processing Letters,vol. 34, no. 4.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1992 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lin, R., Olariu, S. (1992). Computing the inner product on reconfigurable buses with shift switching. In: Bougé, L., Cosnard, M., Robert, Y., Trystram, D. (eds) Parallel Processing: CONPAR 92—VAPP V. VAPP CONPAR 1992 1992. Lecture Notes in Computer Science, vol 634. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-55895-0_413
Download citation
DOI: https://doi.org/10.1007/3-540-55895-0_413
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-55895-8
Online ISBN: 978-3-540-47306-0
eBook Packages: Springer Book Archive