Parallel Simplex for Large Pure Network Problems: Computational Testing and Sources of Speedup

This paper reports on a new parallel implementation of the primal simplex method for minimum cost network flow problems that decomposes both the pivoting and pricing operations. The self-scheduling approach is flexible and efficient; its implementation is close in speed to the best serial code when using one processor, and is capable of substantial speedups as parallel computing units are added. An in-depth computational study of randomly generated transportation and transshipment problems verified the effectiveness of this approach, with results on a 20-processor 80386-based system that are competitive with, and occasionally superior to, massively parallel implementations using tens of thousands of processors. A microanalysis of the code's behavior identified unexpected sources of the occasionally superlinear speedup, including the evolutionary topology of the network basis.

[1]  Jay E. Aronson,et al.  A survey of dynamic network flows , 1989 .

[2]  Fred Glover,et al.  A Computation Study on Start Procedures, Basis Change Criteria, and Solution Algorithms for Transportation Problems , 1974 .

[3]  Egon Balas,et al.  A PARALLEL SHORTEST PATH ALGORITHM FOR THE ASSIGNMENT PROBLEM , 1989 .

[4]  Stavros A. Zenios,et al.  Proximal minimizations with D-functions and the massively parallel solution of linear network programs , 1993, Comput. Optim. Appl..

[5]  Donald L. Miller,et al.  Solution of large dense transportation problems using a parallel primal algorithm , 1990 .

[6]  Darwin Klingman,et al.  Enhancements of Spanning Tree Labeling Procedures for Network Optimization. , 1979 .

[7]  Anita Osterhaug.  Guide to parallel programming on Sequent computer systems , 1989 .

[8]  Michel Dubois,et al.  Synchronization, coherence, and event ordering in multiprocessors , 1988, Computer.

[9]  Fred W. Glover,et al.  Network models in optimization and their applications in practice , 1992 .

[10]  Katta G. Murty,et al.  Network programming , 1992 .

[11]  John M. Mulvey,et al.  Pivot Strategies for Primal-Simplex Network Codes , 1978, JACM.

[12]  Darwin Klingman,et al.  NETGEN: A Program for Generating Large Scale Capacitated Assignment, Transportation, and Minimum Cost Flow Network Problems , 1974 .

[13]  Dimitri P. Bertsekas,et al.  PARALLEL ASYNCHRONOUS PRIMAL-DUAL METHODS FOR THE MINIMUM COST FLOW PROBLEM' , 1990 .

[14]  Ravindra K. Ahuja,et al.  Network Flows: Theory, Algorithms, and Applications , 1993 .

[15]  Darwin Klingman,et al.  A modeling/solution approach for optimal deployment of a weapons arsenal , 1989 .

[16]  Jörg Peters,et al.  The network simplex method on a multiprocessor , 1990, Networks.

[17]  D. R. Fulkerson,et al.  A primal‐dual algorithm for the capacitated Hitchcock problem , 1957 .

[18]  R. V. Helgason,et al.  Algorithms for network programming , 1980 .

[19]  Xiaoye S. Li,et al.  Data-level parallel solution of min-cost network flow problems using ϵ-relaxations , 1994 .

[20]  Michael J. Flynn,et al.  Very high-speed computing systems , 1966 .

[21]  Dimitri P. Bertsekas,et al.  Linear network optimization - algorithms and codes , 1991 .