A GPU-based Parallel Ant Colony Algorithm for Scientific Workflow Scheduling

Scientific workflow scheduling problem is a combinatorial optimization problem. In the real application, the scientific workflow generally has thousands of task nodes. Scheduling large-scale workflow has huge computational overhead. In this paper, a parallel algorithm for scientific workflow scheduling is proposed so that the computing speed can be improved greatly. Our method used ant colony optimization approaches on the GPU. Thousands of GPU threads can parallel construct solutions. The parallel ant colony algorithm for workflow scheduling was implemented with CUDA C language. Scheduling problem instances with different scales were tested both in our parallel algorithm and CPU sequential algorithm. The experimental results on NVIDIA Tesla M2070 GPU show that our implementation for 1000 task nodes runs in 5 seconds, while a conventional sequential algorithm implementation runs in 104 seconds on Intel Xeon X5650 CPU. Thus, our GPU-based parallel algorithm implementation attains a speed-up factor of 20.7.

[1]  Radu Prodan,et al.  Scheduling Scientific Workflows to Meet Soft Deadlines in the Absence of Failure Models , 2010, Euro-Par.

[2]  Martyn Amos,et al.  Enhancing data parallelism for Ant Colony Optimization on GPUs , 2013, J. Parallel Distributed Comput..

[3]  Marco Dorigo,et al.  Optimization, Learning and Natural Algorithms , 1992 .

[4]  Marco Dorigo,et al.  Ant colony optimization , 2006, IEEE Computational Intelligence Magazine.

[5]  Marc Gravel,et al.  Parallel Ant Colony Optimization on Graphics Processing Units , 2013, J. Parallel Distributed Comput..

[6]  Koji Nakano,et al.  An Efficient GPU Implementation of Ant Colony Optimization for the Traveling Salesman Problem , 2012, 2012 Third International Conference on Networking and Computing.

[7]  Marco Dorigo,et al.  Ant system: optimization by a colony of cooperating agents , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[8]  Rajkumar Buyya,et al.  Scheduling scientific workflow applications with deadline and budget constraints using genetic algorithms , 2006, Sci. Program..

[9]  Yu Liu,et al.  Research on MPI-Based Parallel Max-Min Ant System , 2012 .

[10]  Radu Prodan,et al.  Bi-criteria Scheduling of Scientific Workflows for the Grid , 2008, 2008 Eighth IEEE International Symposium on Cluster Computing and the Grid (CCGRID).

[11]  V. Vasudevan,et al.  Scheduling of scientific workflows using Niched Pareto GA for Grids , 2006, 2006 IEEE International Conference on Service Operations and Logistics, and Informatics.

[12]  Teruo Tanaka,et al.  Acceleration of Ant Colony Optimization for the Traveling Salesman Problem on a GPU , 2011 .

[13]  Rajkumar Buyya,et al.  Cost-based scheduling of scientific workflow applications on utility grids , 2005, First International Conference on e-Science and Grid Computing (e-Science'05).