Publication 2003 Issue No. 12 - December Abstract - Improved Methods for Divisible Load Distribution on k-Dimensional Meshes Using Pipelined Communications
Improved Methods for Divisible Load Distribution on k-Dimensional Meshes Using Pipelined Communications
December 2003 (vol. 14 no. 12)
pp. 1250-1261
 ASCII Text x Keqin Li, "Improved Methods for Divisible Load Distribution on k-Dimensional Meshes Using Pipelined Communications," IEEE Transactions on Parallel and Distributed Systems, vol. 14, no. 12, pp. 1250-1261, December, 2003.
 BibTex x @article{ 10.1109/TPDS.2003.1255637,author = {Keqin Li},title = {Improved Methods for Divisible Load Distribution on k-Dimensional Meshes Using Pipelined Communications},journal ={IEEE Transactions on Parallel and Distributed Systems},volume = {14},number = {12},issn = {1045-9219},year = {2003},pages = {1250-1261},doi = {http://doi.ieeecomputersociety.org/10.1109/TPDS.2003.1255637},publisher = {IEEE Computer Society},address = {Los Alamitos, CA, USA},}
 RefWorks Procite/RefMan/Endnote x TY - JOURJO - IEEE Transactions on Parallel and Distributed SystemsTI - Improved Methods for Divisible Load Distribution on k-Dimensional Meshes Using Pipelined CommunicationsIS - 12SN - 1045-9219SP1250EP1261EPD - 1250-1261A1 - Keqin Li, PY - 2003KW - Divisible loadKW - linear arrayKW - load distributionKW - performance analysisKW - pipelined communicationKW - speedup.VL - 14JA - IEEE Transactions on Parallel and Distributed SystemsER -
Keqin Li, IEEE

Abstract—We give the closed form solutions to the parallel time and speedup of the classic method for processing divisible loads on linear arrays as functions of N, the network size. We propose two methods which employ pipelined communications to distribute divisible loads on linear arrays. We derive the closed form solutions to the parallel time and speedup for both methods and show that the asymptotic speedup of both methods is \beta+1, where \beta is the ratio of the time for computing a unit load to the time for communicating a unit load. Such performance is even better than that of the known methods on k{\hbox{-}}{\rm{dimensional}} meshes with k>1. The two new algorithms which use pipelined communications are generalized to distribute divisible loads on k{\hbox{-}}{\rm{dimensional}} meshes, and we show that the asymptotic speedup of both algorithms is k\beta+1, where k\ge1. We also prove that, on k{\hbox{-}}{\rm{dimensional}} meshes where k\ge1, as the network size becomes large, the asymptotic speedup of 2k\beta+1 can be achieved for processing divisible loads by using interior initial processors.

[1] G.M. Amdahl, Validity of the Single Processor Approach to Achieving Large Scale Computing Capabilities Proc. AFIPS Spring Joint Computer Conf., vol. 30, pp. 483-485, 1967.
[2] G.D. Barlas,, Collection Aware Optimum Sequencing of Operations and Closed Form Solutions for the Distribution of a Divisible Load on Arbitrary Processor Trees IEEE Trans. Parallel and Distributed Systems, vol. 9, no. 5, pp. 429-441, May 1998.
[3] S. Bataineh, T. Hsiung, and T.G. Robertazzi, Closed Form Solutions for Bus and Tree Networks of Processors Load Sharing a Divisible Job IEEE Trans. Computers, vol. 43, no. 10, pp. 1184-1196, Oct. 1994.
[4] S. Bataineh and T.G. Robertazzi, Ultimate Performance Limits for Networks of Load Sharing Processors Proc. Conf. Information Sciences and Systems, pp. 794-799, 1992.
[5] V. Bharadwaj, D. Ghose, and V. Mani, An Efficient Load Distribution Strategy for a Distributed Linear Network of Processors with Communication Delays Computers and Math. with Applications, vol. 29, no. 9, pp. 95-112, 1995.
[6] V. Bharadwaj, D. Ghose, V. Mani, and T.G. Robertazzi, Scheduling Divisible Loads in Parallel and Distributed Systems. IEEE Computer Soc. Press, Los Alamitos, Calif., 1996.
[7] J. Blazewicz and M. Drozdowski, Scheduling Divisible Jobs on Hypercubes Parallel Computing, vol. 21, pp. 1945-1956, 1995.
[8] J. Blazewicz and M. Drozdowski, The Performance Limits of a Two-Dimensional Network of Load Sharing Processors Foundations of Computing and Decision Sciences, vol. 21, no. 1, pp. 3-15, 1996.
[9] J. Blazewicz and M. Drozdowski, Distributed Processing of Divisible Jobs with Communication Startup Costs Discrete Applied Math., vol. 76, nos. 1-3, pp. 21-41, 1997.
[10] J. Blazewicz, M. Drozdowski, F. Guinard, and D. Trystram, Scheduling a Divisible Task in a Two-Dimensional Toroidal Mesh Discrete Applied Math., vol. 94, nos. 1-3, pp. 35-50, 1999.
[11] J. Blazewicz, M. Drozdowski, and M. Markiewicz, Divisible Task Scheduling Concept and Verification Parallel Computing, vol. 25, no. 1, pp. 87-98, 1999.
[12] Y.C. Cheng and T.G. Robertazzi, Distributed Computation with Communication Delays IEEE Trans. Aerospace and Electronic Systems, vol. 24, no. 6, pp. 700-712, Nov. 1988.
[13] Y.C. Cheng and T.G. Robertazzi, "Distributed Computation for a Tree Network with Communication Delays," IEEE Trans. Aerospace and Electronic Systems, vol. 26, no. 3, pp. 511-516, May 1990.
[14] M. Drozdowski and W. Glazek, Scheduling Divisible Loads in a Three-Dimensional Mesh of Processors Parallel Computing, vol. 25, no. 4, pp. 381-404, 1999.
[15] D. Ghose and V. Mani, Distributed Computation with Communication Delays: Asymptotic Performance Analysis J. Parallel and Distributed Computing, vol. 23, no. 3, pp. 293-305, 1994.
[16] K. Ko, Scheduling Data Intensive Parallel Processing in Distributed and Networked Environments PhD dissertation, Dept. of Electrical and Computer Eng., State Univ. of New York, Stony Brook, 2000.
[17] K. Li, Managing Divisible Load on Partitionable Networks High Performance Computing Systems and Applications, J. Schaeffer ed., Kluwer Academic Publishers, pp. 217-228, 1998.
[18] K. Li, Parallel Processing of Divisible Loads on Partitionable Static Interconnection Networks Cluster Computing, vol. 6, no. 1, pp. 47-55, 2003.
[19] K. Li, Speedup of Parallel Processing of Divisible Loads on$k{\hbox{-}}{\rm{Dimensional}}$Meshes and Tori Proc. Int'l Conf. Parallel and Distributed Processing Techniques and Applications, pp. 171-177, June 2002.
[20] D. Ghose and V. Mani, "Distributed Computation in a Linear Network: Closed-form Solutions and Computational Techniques," IEEE Trans. Aerospace and Electronic Systems, vol. 30, no. 2, pp. 471-483, Apr. 1994.
[21] D.A.L. Piriyakumar and C.S.R. Murthy, Distributed Computation for a Hypercube Network of Sensor-Driven Processors with Communication Delays Including Setup Time IEEE Trans. Systems, Man, and Cybernetics Part A: Systems and Humans, vol. 28, no. 2, pp. 245-251, Mar. 1998.
[22] T.G. Robertazzi, "Processor Equivalence for a Linear Daisy Chain of Load Sharing Processors," IEEE Trans. Aerospace and Electronic Systems, vol. 29, no. 4, pp. 1,216-1,221, Oct. 1993.
[23] T.G. Robertazzi, Ten Reasons to Use Divisible Load Theory Computer, vol, 36, no. 5, pp. 63-68, May 2003.
[24] J. Sohn and T.G. Robertazzi, Optimal Load Sharing for a Divisible Job on a Bus Network IEEE Trans. Aerospace and Electronic Systems, vol. 32, no. 1, pp. 34-40, Jan. 1996.
[25] J. Sohn, T.G. Robertazzi, and S. Luryi, Optimizing Computing Costs Using Divisible Load Analysis IEEE Trans. Parallel and Distributed Systems, vol. 9, no. 3, pp. 225-234, Mar. 1998.
[26] V. Bharadwaj, X. Li, and K.C. Chung, On the Influence of Start-Up Costs in Scheduling Divisible Loads on Bus Networks IEEE Trans. Parallel and Distributed Systems, vol. 11, no. 12, pp. 1288-1305, Dec. 2000.

Index Terms: