Title
Optimality of HLF for scheduling divide-and-conquer UET task graphs on identical parallel processors
Abstract
The problem of scheduling a set of n unit execution time (UET) tasks subject to precedence constraints on m identical parallel processors is known to be NP-hard in the strong sense. However, polynomial time algorithms exist for some classes of precedence graphs. In this paper, we consider a class of divide-and-conquer graphs that naturally models the execution of the recursive control abstraction of divide-and-conquer algorithms. We prove that the Highest Level First (HLF) strategy minimizes the schedule length for this class, thus settling a conjecture of Rayward-Smith and Clark.
Year
DOI
Venue
2009
10.1016/j.disopt.2008.09.001
Discrete Optimization
Keywords
Field
DocType
recursive control abstraction,hlf,precedence graph,schedule length,divide-and-conquer algorithm,divide-and-conquer graph,m identical parallel processor,highest level first,divide-and-conquer uet task graph,complexity,strong sense,makespan,n unit execution time,polynomial time algorithm,identical parallel processors,divide and conquer
Discrete mathematics,Graph,Job shop scheduling,Abstraction,Computer science,Scheduling (computing),Divide and conquer algorithms,Time complexity,Conjecture,Recursion
Journal
Volume
Issue
ISSN
6
1
Discrete Optimization
Citations 
PageRank 
References 
1
0.35
6
Authors
3
Name
Order
Citations
PageRank
Wieslaw Kubiak154062.61
Djamal Rebaine2197.34
Chris Potts310.35