Mapping tasks onto nodes: a parallel local neighborhood approach

Abstract We consider the problem of mapping processes onto computing nodes so as to reduce the execution time by minimizing communication delays. Our approach relies on a genetic algorithm implementation of the local neighborhood search (LNS) approach and is called Genetic-LNS or GLNS. We also present our parallel version of the GLNS algorithm, called parallel genetic local neighborhood search (P-GLNS). LNS, GNLS, and P-GLNS were implemented and compared. Simulations demonstrate that the GLNS algorithm has better performance than LNS, and that, when the workload is sufficiently high, the P-GLNS algorithm achieved near linear scalability.

[1]  Shahid H. Bokhari,et al.  On the Mapping Problem , 1981, IEEE Transactions on Computers.

[2]  Ali R. Hurson,et al.  Scheduling and Load Balancing in Parallel and Distributed Systems , 1995 .

[3]  El-Ghazali Talbi,et al.  A parallel genetic algorithm for the graph partitioning problem , 1991, ICS '91.

[4]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[5]  Ophir Frieder,et al.  A Parallel Solution for the Multiprocessor Document Allocation Problem , 1994, 1994 International Conference on Parallel Processing Vol. 3.

[6]  Benjamin W. Wah,et al.  Load Balancing: An Automated Learning Approach , 1995 .

[7]  R. F. Freund,et al.  Scheduling resources in multi-user, heterogeneous, computing environments with SmartNet , 1998, Proceedings Seventh Heterogeneous Computing Workshop (HCW'98).

[8]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[9]  David Mark Levine,et al.  A parallel genetic algorithm for the set partitioning problem , 1995 .

[10]  Virginia Mary Lo,et al.  Heuristic Algorithms for Task Assignment in Distributed Systems , 1988, IEEE Trans. Computers.

[11]  Hava T. Siegelmann,et al.  On the allocation of documents in multiprocessor information retrieval systems , 1991, SIGIR '91.

[12]  Mihalis Yannakakis,et al.  The Analysis of Local Search Problems and Their Heuristics , 1990, STACS.

[13]  Nashat Mansour,et al.  Physical optimization algorithms for mapping data to distributed-memory multiprocessors , 1992 .

[14]  Francisco Sandoval Hernández,et al.  Genetic Algorithms on LAN-message Passing Architectures using PVM: Application to the Routing Problem , 1994, PPSN.

[15]  Fikret Ercal,et al.  Heuristic approaches to task allocation for parallel computing , 1988 .

[16]  Jake K. Aggarwal,et al.  A Mapping Strategy for Parallel Processing , 1987, IEEE Transactions on Computers.

[17]  Massoud Pedram,et al.  MERLIN: semi-order-independent hierarchical buffered routing tree generation using local neighborhood search , 1999, DAC '99.