Efficiency and Quality of Solution of Parallel Simulated Annealing

The paper describes parallel computations for solving room assignment problem using simulated annealing on multicomputer platform. Parallel independent runs on all available processors are suggested providing both speedup of the calculations of the best final arrangement and high diversification of the independent moves in the state space. On the base of experiments with parallel MPI implementation both speedup and quality of solution of the suggested parallel simulated annealing algorithm are examined and analyzed. The impact of multicomputer size (number of parallel working workstations) and workload (number of persons) on the parallel performance are explored and performance profiling and analysis are presented.

[1]  Yves Crama,et al.  Local Search in Combinatorial Optimization , 2018, Artificial Neural Networks.

[2]  Ibrahim H. Osman,et al.  Heuristics for the generalised assignment problem: simulated annealing and tabu search approaches , 1995 .

[3]  Shian-Shyong Tseng,et al.  A Simulated Annealing Algorithm for Extended Cell Assignment Problem in a Wireless ATM Network , 2001, EvoWorkshops.

[4]  H. Martinez-Alfaro,et al.  Using simulated annealing to solve the classroom assignment problem , 2002, Proceedings Mexico-USA Collaboration in Intelligent Systems Technologies..

[5]  Chang-Sung Jeong,et al.  Fast parallel simulated annealing for traveling salesman problem on SIMD machines with linear interconnections , 1991, Parallel Comput..

[6]  Peter Salamon,et al.  Facts, Conjectures, and Improvements for Simulated Annealing , 1987 .

[7]  Slim Abdennadher,et al.  University course timetabling using constraint handling rules , 2000, Appl. Artif. Intell..

[8]  Norbert Oster,et al.  A Hybrid Genetic Algorithm for School Timetabling , 2002, Australian Joint Conference on Artificial Intelligence.

[9]  Geoffrey C. Fox,et al.  A Comparison of Annealing Techniques for Academic Course Scheduling , 1997, PATAT.

[10]  Cecilia R. Aragon,et al.  Optimization by Simulated Annealing: An Experimental Evaluation; Part II, Graph Coloring and Number Partitioning , 1991, Oper. Res..

[11]  Cecilia R. Aragon,et al.  Optimization by Simulated Annealing: An Experimental Evaluation; Part I, Graph Partitioning , 1989, Oper. Res..

[12]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.