A reindexing based approach towards mapping of DAG with affine schedules onto parallel embedded systems

Clementin Tayou Djamegni (1) , Patrice Quinton (2) , Sanjay Rajopadhye (3) , Tanguy Risset (4, 5, 6) , Maurice Tchuente (7)

1 Faculté des Sciences - Université de Dschang [Cameroun]
2 CAIRN - Energy Efficient Computing ArchItectures with Embedded Reconfigurable Resources
3 Department of Computer Science [Colorado]
4 ARES - Architectures of networks of services
5 COMPSYS - Compilation and embedded computing systems
6 CITI - CITI Centre of Innovation in Telecommunications and Integration of services
7 Laboratoire de Calcul Parallèle

We address the problem of optimally mapping uniform DAGs to systolic arrays, given an ane timing function. We introduce an automatic allocation method based on a preprocessing by reindexing that transforms the initial DAG into a new one that enables the well known projection method to minimize the number of processors along a number of directions. We demonstrate its superiority to other methods, and establish the space-optimality of the proposed method. We also show an upper bound on the number of processors that corresponds to the best space complexity that both the projection method and the so-called grouping method can give for the initial DAG. We also describe how the new allocation method can be implemented in tools.

Domaines: Informatique

Clementin Tayou Djamegni, Patrice Quinton, Sanjay Rajopadhye, Tanguy Risset, Maurice Tchuente. A reindexing based approach towards mapping of DAG with affine schedules onto parallel embedded systems. Journal of Parallel and Distributed Computing, 2009, 69 (1), pp.1-11. ⟨hal-00410708⟩