Susumu Matsumae
ABSTRACT
A reconfigurable mesh (RM) is the two-dimensional mesh-connected computer enhanced with a reconfigurable bus system. The bus system is used to dynamically obtain various interconnection patterns among the processors during the execution of programs. This paper shows that the RM of size n × n can be simulated in O(n2/m2 log n) steps by the RM of smaller size m × m (m < n). The algorithm improves the previously known best results, O (n2/m2 log m log n / m) steps proposed by Fernández-Zepeda et al., under some conditions (e.g. m = nc for a constant c with 0 < c < 1). Furthermore, our simulation algorithm can run on the restricted model of the RM called linear RM, which is a simpler model than that used by Fernández-Zepeda et al.
- 1.Y. Ben-Asher, D. Gordon, and A. Schuster. Optimal Simulations in Reconfigurable Arrays. Technical Report ~p716, Technion - Israel Institute of Technology, Computer Science Department, February 1992.Google Scholar
- 2.Y. Ben-Asher, D. Gordon, and A. Schuster. Efficient Self-Simulation Algorithms for Reconfigurable Arrays. J. of Parallel and Distrib~tted Comporting, 30(1):1-22, October 1995. Google ScholarDigital Library
- 3.Y. Ben-Asher, K.-J. Lange, D. Peleg, and A. Schuster. The Complexity of Reconfiguring Network Models. Information and Computation, 121:41-58, 1995. Google ScholarDigital Library
- 4.Y. Ben-Asher, D. Peleg, R. Ramaswami, and A. Schuster. The Power of Reconfiguration. J. of Parallel and Distributed Computing, 13:139-153, 1991. Google ScholarDigital Library
- 5.B. Beresford-Smith, O. Diessel, and H. E1Gindy. Optimal Algorithms for Constrained Reconfigurable Meshes. J. of Parallel and Distributed Computing, 39:74-78, 1996. Google ScholarDigital Library
- 6.V. Bokka, H. Gurla, S. Olariu, and J. L. Schwing. Time-Optimal Domain-Specific Querying on Enhanced Meshes. IEEE Trans. on Parallel and Distributed Systerns, 8(1):13-24, 1997. Google ScholarDigital Library
- 7.K. Bondalapati and V. K. Prasanna. Reconfigurable Meshes: Theory and Practice. In Proc. of the 5th Reconfigurable Architectures Workshop (RAW'97), International Parallel Processing Symposium, 1997.Google Scholar
- 8.K. -L. Chung. Prefix Computations on a Generalized Mesh-Connected Computer with Multiple Buses. IEEE Trans. on Parallel and Distributed Systems, 6(2):196- 199, February 1995. Google ScholarDigital Library
- 9.J. A. Fern~ndez-Zepeda, R. Vaidyanathan, and J. L. Trahan. Scaling Simulation of the Fusing-Restricted Reconfigurable Mesh. IEEE Trans. on Parallel and Distributed Systems, 9(9):861-871, September 1998. Google ScholarDigital Library
- 10.T. Maeba, M. Sugaya, S. Tatsumi, and K. Abe. Semigroup Computations on a Processor Array with Partitioned Buses. IEICE Trans. A, J80-A(2):410-413, 1997.Google Scholar
- 11.M. Maresca. Polymorphic Processor Arrays. IEEE Trans. on Parallel and Distributed Systems, 4(5):490- 506, May 1993. Google ScholarDigital Library
- 12.Y. Matias and A. Schuster. Fast, Efficient Mutual and Self Simulations for Shared Memory and Reconfigurable Mesh. In Proc. of the 7th IEEE Syrup. on Parallel and Distributed Processing, pages 238-246, 1995. Google ScholarDigital Library
- 13.R. Miller, V. K. Prasanna-Kumar, D. I. Reisis, and Q. F. Stout. Parallel Computations on Reconfigurable Meshes. IEEE Trans. on Computers, 42(6):678-692, June 1993. Google ScholarDigital Library
- 14.M. M. Murshed and R. P. Brent. Algorithms for Optireal Self-Simulation of Some Restricted Reconfigurable Meshes. Technical Report TR-CS-97-16, Joint Computer Science Technical Report Series, The Australian National University, July 1997.Google Scholar
- 15.K. Nakano. A Bibliography of Published Papers on Dynamically Reconfigurable Architectures. Parallel Processing Letters, 5(1):111-124, 1995.Google ScholarCross Ref
- 16.V. K. Prasanna-Kumar and C. S. Raghavendra. Array Processor with Multiple Broadcasting. J. of Parallel and Distributed Computing, 4:173-190, 1987. Google ScholarDigital Library
- 17.M. J. Serrano and B. Parhami. Optimal Architectures and Algorithms for Mesh-Connected Parallel Computers with Separable Row/Column Buses. IEEE Trans. on Parallel and Distributed Systems, 4(10):1073-1080, October 1993. Google ScholarDigital Library
- 18.B. Wang and G. Chen. Constant Time Algorithms for the Transitive Closure and Some Related Graph Problems on Processor Arrays with Reconfigurable Bus Systerns. IEEE Trans. on Parallel and Distributed Systems, 1(4):500-507, October 1990. Google ScholarDigital Library
Index Terms
- An efficient self-simulation algorithm for reconfigurable meshes
Recommendations
An Efficient Algorithm for Row Minima Computations on Basic Reconfigurable Meshes
A matrix A of size m n containing items from a totally ordered universe is termed monotone if, for every i, j, 1 i < j m, the minimum value in row j lies below or to the right of the minimum in row i. Monotone matrices, and variations thereof, are known ...
An Efficient Scaling-Simulation Algorithm of Reconfigurable Meshes by Meshes with Partitioned Buses
IPDPS '03: Proceedings of the 17th International Symposium on Parallel and Distributed ProcessingThis paper presents an efficient scaling-simulation algorithm that simulates operations of the reconfigurable mesh (RM) of size n n using the mesh with partitioned buses (MPB) of size m m (m < n). The RM and the MPB are the two-dimensional mesh-...
Constant Time Sorting on Reconfigurable Meshes
We present a constant time sorting algorithm by adopting a 3D reconfigurable mesh with only O(n/sup 3/2/) processors. Our algorithm is developed on an n/sup 1/2//spl times/n/sup 1/2//spl times/n/sup 1/2/ 3-D reconfigurable mesh. Moreover, we further ...
Comments