Issue No. 09 - Sept. (2017 vol. 28)
ISSN: 1045-9219
pp: 2689-2702
Raphael Bleuse , Univ. Grenoble Alpes, CNRS, Inria, Grenoble INP, LIG, Grenoble, France
Sascha Hunold , Faculty of Informatics, TU Wien, Institute of Information Systems, Favoritenstraße 16/184-5, Vienna, Austria
Safia Kedad-Sidhoum , Sorbonne Universités, UPMC University Paris 06, UMR 7606, LIP6, Paris, France
Florence Monna , Sorbonne Universités, UPMC University Paris 06, UMR 7606, LIP6, Paris, France
Gregory Mounie , Univ. Grenoble Alpes, CNRS, Inria, Grenoble INP, LIG, Grenoble, France
Denis Trystram , Univ. Grenoble Alpes, CNRS, Inria, Grenoble INP, LIG, Grenoble, France
ABSTRACT
We present a new approach for scheduling independent tasks on multiple CPUs and multiple GPUs. The tasks are assumed to be parallelizable on CPUs using the moldable model: the final number of cores allotted to a task can be decided and set by the scheduler. More precisely, we design an algorithm aiming at minimizing the makespan—the maximum completion time of all tasks—for this scheduling problem. The proposed algorithm combines a dual approximation scheme with a fast integer linear program (ILP). It determines both the partitioning of the tasks, i.e., whether a task should be mapped to CPUs or a GPU, and the number of CPUs allotted to a moldable task if mapped to the CPUs. A worst-case analysis shows that the algorithm has an approximation ratio of $\frac{3}{2} + \epsilon$ . Since the time complexity of the ILP-based algorithm could be non-polynomial, we also present a polynomial-time algorithm with an approximation ratio of $2+\epsilon$ . We complement the theoretical analysis of our two novel algorithms with a simulation study. In these simulations, we compare our algorithms to a modified version of the classical HEFT algorithm, which we adapted to handle moldable tasks. The simulation results show that our algorithm with the $\left(\frac{3}{2} + \epsilon \right)$ -approximation ratio produces significantly shorter schedules than the modified HEFT for most of the instances. In addition, our results provide evidence that our ILP-based algorithm can solve larger problem instances in a reasonable amount of time.
INDEX TERMS
Approximation algorithms, Algorithm design and analysis, Scheduling, Graphics processing units, Scheduling algorithms
CITATION

R. Bleuse, S. Hunold, S. Kedad-Sidhoum, F. Monna, G. Mounie and D. Trystram, "Scheduling Independent Moldable Tasks on Multi-Cores with GPUs," in IEEE Transactions on Parallel & Distributed Systems, vol. 28, no. 9, pp. 2689-2702, 2017.
doi:10.1109/TPDS.2017.2675891