Is Acyclic Directed Graph Partitioning Effective for Locality-Aware Scheduling?

Yusuf M. Özkaya (1) , Anne Benoit (2, 3, 4) , Ümit V. Çatalyürek (4)

1 Georgia Institute of Technology [Atlanta]
2 ROMA - Optimisation des ressources : modèles, algorithmes et ordonnancement
3 LIP - Laboratoire de l'Informatique du Parallélisme
4 GATECH - College of Computing

We investigate efficient execution of computations, modeled as Directed Acyclic Graphs (DAGs), on a single processor with a two-level memory hierarchy, where there is a limited fast memory and a larger slower memory. Our goal is to minimize execution time by minimizing redundant data movement between fast and slow memory. We utilize a DAG partitioner that finds localized, acyclic parts of the whole computation that can fit into fast memory, and minimizes the edge cut among the parts. We propose a new scheduler that executes each part one-by-one, obeying the dependency among parts, aiming at reducing redundant data movement needed by cut-edges. Extensive experimental evaluation shows that the proposed DAG-based scheduler significantly reduces redundant data movement.

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

Yusuf M. Özkaya, Anne Benoit, Ümit V. Çatalyürek. Is Acyclic Directed Graph Partitioning Effective for Locality-Aware Scheduling?. PPAM 2019 - 13th International Conference on Parallel Processing and Applied Mathematics, Sep 2019, Bialystok, Poland. ⟨hal-02273122⟩