A multi-population based parallel genetic algorithm for multiprocessor task scheduling with Communication Costs

Multiprocessor task scheduling is one of the hardest combinatorial optimization problems in parallel and distributed systems. It is known as NP-hard and therefore, scanning the whole search space using an exact algorithm to find the optimal solution is not practical. Instead, metaheuristics are mostly employed to find a near-optimal solution in a reasonable amount of time. In this paper, a multi-population based parallel genetic algorithm is presented for the optimization of multiprocessor task scheduling in the presence of communication costs. To the best of our knowledge, this parallel genetic algorithm approach is applied to the problem at hand for the first time using a benchmark set that includes well-known task graphs from different sources. Our experiments conducted on several task graphs with different sizes from the benchmark set showed the superiority of the approach over a conventional genetic algorithm and the related works in the literature in terms of two different performance metrics. Our parallel implementation not only decreased the execution time but also increased the quality of the scheduling solutions considerably.

[1]  Jan Janeček,et al.  Static vs. Dynamic List-Scheduling Performance Comparison , 2003 .

[2]  Heinz Mühlenbein,et al.  The parallel genetic algorithm as function optimizer , 1991, Parallel Comput..

[3]  Gurvinder Singh,et al.  Heuristics Based Genetic Algorithm for Scheduling Static Tasks in Homogeneous Parallel System , 2022 .

[4]  Fatma A. Omara,et al.  Genetic algorithms for task scheduling problem , 2010, J. Parallel Distributed Comput..

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

[6]  Mitsuo Gen,et al.  A comparison of multiprocessor task scheduling algorithms with communication costs , 2008, Comput. Oper. Res..

[7]  Hermann Gehring,et al.  A Parallel Genetic Algorithm for Solving the Container Loading Problem , 2002 .

[8]  Tatsuhiro Tsuchiya,et al.  Fault-secure scheduling of arbitrary task graphs to multiprocessor systems , 2000, Proceeding International Conference on Dependable Systems and Networks. DSN 2000.

[9]  Ishfaq Ahmad,et al.  Efficient Scheduling of Arbitrary TAsk Graphs to Multiprocessors Using a Parallel Genetic Algorithm , 1997, J. Parallel Distributed Comput..

[10]  Tomasz Kalinowski,et al.  Task Scheduling using Parallel Genetic Algorithm implemented with GRADE , 1998 .

[11]  Goran Lj. Djordjevic,et al.  A Heuristic for Scheduling Task Graphs with Communication Delays Onto Multiprocessors , 1996, Parallel Comput..

[12]  Ravreet Kaur,et al.  Task Graph Scheduling on Multiprocessor System using Genetic Algorithm , 2012 .

[13]  Kirat Pal Singh,et al.  Design of High Performance MIPS Cryptography Processor Based on T-DES Algorithm , 2015, ArXiv.

[14]  Erick Cantú-Paz,et al.  Migration Policies, Selection Pressure, and Parallel Evolutionary Algorithms , 2001, J. Heuristics.

[15]  Mohammad Reza Bonyadi,et al.  A Bipartite Genetic Algorithm for Multi-processor Task Scheduling , 2009, International Journal of Parallel Programming.

[16]  Habib Motee Ghader,et al.  A hybrid method for task scheduling , 2010, 2010 2nd International Conference on Education Technology and Computer.

[17]  Y.-K. Kwok,et al.  Static scheduling algorithms for allocating directed task graphs to multiprocessors , 1999, CSUR.