Abstract
Most work on the problem of synthesizing a systolic array from a system of recurrence equations is restricted to systems of uniform recurrence equations. Recently, researchers have begun to relax this restriction to include systems of affine recurrence equations. A system of uniform recurrence equations typically can be embedded in spacetime so that the distance between a variable and a dependent variable does not depend on the problem size. Systems of affine recurrence equations that are not uniform do not enjoy this property. A method is presented for converting a system of affine recurrence equations to an equivalent system of recurrence equations that is uniform, except for points near the boundaries of its index sets. Necessary and sufficient conditions are given for an affine system to be amenable to such a conversion, along with an algorithm that checks for these conditions, and a procedure that converts those affine systems which can be converted.
The characterization of convertible systems brings together classical ideas in algebraic geometry, number theory, and matrix representations of groups. While the proof of this characterization is complex, the characterization itself is simple, suggesting that the mathematical ideas are well chosen for this difficult problem in array design.
Similar content being viewed by others
References
H.-T. Kung and Charles E. Leiserson, “Algorithms for VLSI processor arrays,”Introduction to VLSI Systems, Menlo Park, CA: Addison-Wesley Publishing Co., 1980, pp. 271–292.
H.-T. Kung, “Why systolic architectures?,”Computer, Vol. 15, No. 1, pp. 37–45, January 1982.
S.-Y. Kung, “On supercomputing with systolic/wavefront array processors,”Proceedings of the IEEE, July 1984.
Charles E. Leiserson, and James B. Saxe, “Optimizing synchronous systems,”Proc. IEEE 22nd Annual Symp. Foundations of Computer Science, Oct. 1981.
Charles E. Leiserson, Flavio M. Rose, and James B. Saxe, “Optimizing synchronous circuitry by retiming,”Proc. Third Caltech Conf. on VLSI, Rockville, MD: Computer Science Press, 1983.
Jeffrey D. Ullman,Computational Aspects of VLSI, Rockville, MD: Computer Science Press, Inc., 1984.
R.G. Melhem and W.C. Rheinboldt, “A mathematical model for the verification of systolic networks,”SIAM J. Computing, Vol. 13, No. 3, pp. 541–565, 1984.
Richard M. Karp, Richard E. Miller, and Shmuel Winograd, “The organization of computations for uniform recurrence equations,”J. ACM, Vol. 14, pp. 563–590, 1967.
Richard M. Karp, Richard E. Miller, and Shmuel Winograd, “Properties of a model for parallel computations: Determinacy, termination, queueing,”SIAM J. Appl. Math, Vol. 14, pp. 1390–1411, 1966.
Marina C. Chen and Carver Mead, “Concurrent algorithms as space-time recursion equations,” S.-Y. Kung, H. J. Whitehouse, and Thomas Kailath (Eds.),VLSI & Modern Signal Processing, Englewood Cliffs: Prentice-Hall, 1985.
Patrice Quinton, “Automatic synthesis of systolic arrays from uniform recurrent equations,”Proc. 11th Ann. Symp. on Computer Architecture, pp. 208–214, 1984.
Peter R. Cappello, “VLSI Architectures for Digital Signal Processing,” Ph.D. Thesis, Princeton University, Princeton, NJ, October 1982.
Peter R. Cappello and Kenneth Steiglitz, “Unifying VLSI array design with linear transformations of space-time,” Franco P. Preparata (Ed.),Advances in Computing Research, Vol. 2: VLSI Theory, Greenwich, CT: JAI Press, Inc., 1984, pp. 23–65.
José A. B. Fortes, King-Sun Fu, and Benjamin W. Wah, “Systematic approaches to the design of algorithmically specified systolic arrays,”Proc. Int. Conf. on Acoustics, Speech, and Signal Processing, Tampa, 1985, pp. 300–303.
Dan I. Moldovan, “On the design of algorithms for VLSI systolic arrays,”Proc. IEEE, Vol. 71, pp. 113–120, 1983.
Dan I. Moldovan, “On the analysis and synthesis of VLSI algorithms,”IEEE Trans. Comput., Vol. C-31, pp. 1121–1126, 1982.
Willard L. Miranker and Andrew Winkler, “Spacetime representations of computational structures,”Computing, Vol. 32, pp. 93–114, 1984.
Jean-Marc Delosme and Ilse Ipsen, “Efficient systolic arrays for the solution of toeplitz systems: An illustration of a methodology for the construction of systolic architectures in VLSI,” W. Moore et al. (Ed.),Systolic Arrays, Bristol: Adam Hilger, 1987, pp. 37–46.
Dan I. Moldovan and José A. B. Fortes, “Partitioning and mapping algorithms into fixed systolic arrays,”IEEE Trans. Comput., Vol. C-35, pp. 1–12, 1986.
Sailesh K. Rao, “Regular Iterative Algorithms and Their Implementation on Processor Arrays,” Ph.D. Thesis, Stanford University, October 1985.
Jean-Marc Delosme and Ilse C. F. Ipsen, “Systolic array synthesis: Computability and time cones,” Technical Report Yale/DCS/RR-474, Yale, May 1986.
C. Choffrut and K. Culik II, “Folding of the plane and the design of systolic arrays,”Information Processing Letters, Vol. 17, pp. 149–153, 1983.
Sanjay Rajopadhye, Lap Mui, and Sayfe Kiaei, “Piecewise linear schedules for recurrence equations,” Kung Yao et al. (Eds.),VLSI Signal Processing V, New York: IEEE Press, October 1992, pp. 375–384.
Alain Darte, Leonid Khachiyan, and Yves Robert, “Linear scheduling is close to optimal,” José Fortes et al. (Eds.),Application Specific Array Processors, IEEE Computer Society Press, August 1992, pp. 37–46.
Xiaoxiong Zhong and Sanjay Rajopadhye, “Optimal parallel schedules for uniform recurrence equations,” Computer science, University of Oregon, Eugene, 1992.
Weijia Shang and José A. B. Fortes, “Time optimal linear schedules for algorithms with uniform dependencies,”IEEE Trans. Comput., Vol. 40, pp. 723–742, 1991.
Yoav Yaacoby, “Computing Systems of Affine Recurrence Equations on Processor Arrays,” Ph.D. Thesis, University of California, Santa Barbara, Santa Barbara, CA, May 1988.
Sun-Yuan Kung and Y. H. Hu, A highly concurrent algorithm and pipelined architecture for solving toeplitz systems,IEEE Trans. on Acoustics, Speech, and Signal Processing, Vol. 31, pp. 66–74, 1983.
J.-M. Delosme and I. C. F. Ipsen, “Parallel solution of symmetric positive definite systems with hyperbolic rotations,”Linear Algebra Appl., Vol. 77, pp.75–111, 1986.
U. Banerjee, S.-C. Chen, D. J. Kuck, and R. A. Towle, “Time and parallel processor bounds for Fortran-like loops,”IEEE Trans. Computers, Vol. C-28, pp. 660–670, 1979.
F. Harary,Graph Theory, Reading, MA: Addison-Wesley, 1972.
E.R. Berlekamp,Algebraic Coding Theory, New York: McGraw-Hill Book Company, 1968.
M. Newman, “Matrix representation of groups,”Applied Mathematics Series -60. Institute for Basic Standards, National Bureau of Standards, Washington D.C. 20234, July 1968.
C.W. Curtis and I. Reiner,Representation Theory of Finite Groups and Associative Algebras, New York: John Wiley & Sons, 1962.
Yoav Yaacoby, Peter R. Cappello, D. Witt, and K.C. Millett, “Computing a fundamental region for a finite matrix group acting on a Euclidean space,” Technical Report 4, Dept. Computer Science, UCSB, Santa Barbara, CA 93106, Feb. 1988.
Yoav Yaacoby, Peter R. Cappello, D. Witt, and K.C. Millett, Computing a fundamental region for a finite matrix group acting on a Euclidean space, Submitted toSIAM J. Computing, 1988.
H.S.M. Coxeter,Regular Polytopes, New York: Dover Publications, Inc., third edition, 1973.
Shimon Even,Graph Algorithms, Rockville, MD: Computer Science Press, Inc., 1979.
Yuepeng Zheng and Sayfe Kiaei, “Multi-rate transformation of directional affine recurrence equations,” Luigi Dadda and Benjamin Wah (Eds.),Proc. Int. Conf. Application-Specific Array Processors, Venice, Italy, October 1993, pp. 392–403.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Yaacoby, Y., Cappello, P. Converting affine recurrence equations to quasi-uniform recurrence equations. Journal of VLSI Signal Processing 11, 113–131 (1995). https://doi.org/10.1007/BF02106826
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF02106826