Title :
Massively parallel simulated annealing embedded with downhill-a SPMD algorithm for cluster computing
Author :
Du, Zhihui ; Li, Sanli ; Li, Shuyou ; Wu, Mengyue ; Zhu, Jing
Author_Institution :
Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing, China
Abstract :
Simulated Annealing (SA) is a frequently used stochastic algorithm to deal with combinatorial optimization problems and it converges with probability infinitely close to 1. SA is an NP algorithm and the long executive time prevents it from being accepted for many real-time applications. This paper presents a SPMD (Single Program Multiple Data) algorithm which combines SA with local searching algorithm-downhill. The hybrid method not only keeps the convergence of SA but also improves the convergence speed of SA. Approximate solutions can be found quickly for complex optimization problems and more precise solutions can also be found by employing the same algorithm to fine-tune the approximate solutions. SA is an essential serial algorithm, but the SPMD algorithm breaks up the serial bottleneck of SA and its performance scales up linearly with the increase of processors, at the same time, the SPMD algorithm does not require careful choice of control parameters. Application cases show that the algorithm is robust and it can find high quality solution with high speed
Keywords :
parallel algorithms; simulated annealing; stochastic programming; workstation clusters; SPMD algorithm; cluster computing; combinatorial optimization problems; complex optimization problems; local searching algorithm; massively parallel simulated annealing; single program multiple data algorithm; stochastic algorithm; Clustering algorithms; Computational modeling; Computer science; Concurrent computing; Convergence; Embedded computing; Materials science and technology; Optimization methods; Simulated annealing; Stochastic processes;
Conference_Titel :
Cluster Computing, 1999. Proceedings. 1st IEEE Computer Society International Workshop on
Conference_Location :
Melbourne, Vic.
Print_ISBN :
0-7695-0343-8
DOI :
10.1109/IWCC.1999.810899