HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Reports

Dynamic Scheduling of Parallel Computations

Abstract : Structures of parallel programs are usually modeled by task graphs in the scheduling literature. Such graphs are sometimes obtained while compiling the parallel programs. In many other cases, however, they can be determined only at run time. In this paper, we consider the scheduling of parallel computations whose task graphs are generated at run time. We analyze the case where the task graph has a random out-tree structure. When the number of offspring of a task has geometric distribution with a parameter which is decreasing and convex in the level, or the generation, then the breadth first policy stochastically minimizes the makespan. If, however, this parameter is increasing and concave, then the depth first policy stochastically minimizes the makespan.
Document type :
Reports
Complete list of metadata

https://hal.inria.fr/inria-00073644
Contributor : Rapport de Recherche Inria Connect in order to contact the contributor
Submitted on : Wednesday, May 24, 2006 - 1:24:27 PM
Last modification on : Friday, February 4, 2022 - 3:09:24 AM
Long-term archiving on: : Sunday, April 4, 2010 - 9:37:41 PM

Identifiers

  • HAL Id : inria-00073644, version 1

Collections

Citation

Zhen Liu. Dynamic Scheduling of Parallel Computations. RR-3048, INRIA. 1996. ⟨inria-00073644⟩

Share

Metrics

Record views

60

Files downloads

78