On Characterizing the Data Movement Complexity of Computational DAGs for Parallel Execution

Abstract : Technology trends are making the cost of data movement increasingly dominant, both in terms of energy and time, over the cost of performing arithmetic operations in computer systems. The fundamental ratio of aggregate data movement bandwidth to the total computational power (also referred to the \emph{machine balance parameter}) in parallel computer systems is decreasing. It is therefore of considerable importance to characterize the inherent data movement requirements of parallel algorithms, so that the minimal architectural balance parameters required to support it on future systems can be well understood. In this paper, we develop an extension of the well-known red-blue pebble game to develop lower bounds on the data movement complexity for the parallel execution of computational directed acyclic graphs (CDAGs) on parallel systems. We model multi-node multi-core parallel systems, with the total physical memory distributed across the nodes (that are connected through some interconnection network) and in a multi-level shared cache hierarchy for processors within a node. We also develop new techniques for lower bound characterization of non-homogeneous CDAGs. We demonstrate the use of the methodology by analyzing the CDAGs of several numerical algorithms, to develop lower bounds on data movement for their parallel execution.
Type de document :
Communication dans un congrès
Symposium on Parallelism in Algorithms and Architectures (SPAA '14), 2014, Prague, Poland. ACM, pp.296-306, 2014, 〈10.1145/2612669.2612694〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01016090
Contributeur : Fabrice Rastello <>
Soumis le : vendredi 27 juin 2014 - 16:57:32
Dernière modification le : mercredi 11 avril 2018 - 01:52:13

Identifiants

Citation

Venmugil Elango, Fabrice Rastello, Louis-Noël Pouchet, Jagannathan Ramanujam, Ponnuswamy Sadayappan. On Characterizing the Data Movement Complexity of Computational DAGs for Parallel Execution. Symposium on Parallelism in Algorithms and Architectures (SPAA '14), 2014, Prague, Poland. ACM, pp.296-306, 2014, 〈10.1145/2612669.2612694〉. 〈hal-01016090〉

Partager

Métriques

Consultations de la notice

350