Reliable and energy-aware mapping of streaming series-parallel applications onto hierarchical platforms

Changjiang Gou (1, 2) , Anne Benoit (2) , Mingsong Chen (1) , Loris Marchal (2) , Tongquan Wei (1)

1 ECNU - East China Normal University [Shangaï]
2 ROMA - Optimisation des ressources : modèles, algorithmes et ordonnancement

Streaming applications come from various application fields such as physics, and many can be represented as a series-parallel dependence graph. We aim at minimizing the energy consumption of such applications when executed on a hierarchical platform, by proposing novel mapping strategies. Dynamic voltage and frequency scaling (DVFS) is used to reduce the energy consumption, and we ensure a reliable execution by either executing a task at maximum speed, or by triplicating it. In this paper, we propose a structure rule to partition the series-parallel applications, and we prove that the optimization problem is NP-complete. We are able to derive a dynamic- programming algorithm for the special case of linear chains, which provides an interesting heuristic and a building block for designing heuristics for the general case. The heuristics performance is compared to a baseline solution, where each task is executed at maximum speed. Simulations demonstrate that significant energy savings can be obtained.

Mots clés: Mapping; partitioning; scheduling; SPG; reliability; energy.

Domaines: Calcul parallèle, distribué et partagé [cs.DC]

Changjiang Gou, Anne Benoit, Mingsong Chen, Loris Marchal, Tongquan Wei. Reliable and energy-aware mapping of streaming series-parallel applications onto hierarchical platforms. [Research Report] RR-9346, INRIA. 2020. ⟨hal-02859980v2⟩