Title :
Parallelization of scheduling algorithms
Author :
Wu, Min-You ; Shu, Wei
Author_Institution :
Dept. of Comput. Sci., State Univ. of New York, Buffalo, NY, USA
Abstract :
Most static scheduling algorithms that schedule parallel programs represented by directed acyclic graphs (DAGs) are sequential. Parallelization of scheduling algorithms is a new research area. This paper discusses the essential issues of parallel static scheduling and presents an efficient parallel scheduling algorithm, the HPMCP algorithm. The proposed algorithm has been implemented on an Intel Paragon machine. It produces high-quality scheduling and is much faster than existing sequential and parallel algorithms
Keywords :
data flow graphs; parallel programming; processor scheduling; HPMCP algorithm; Intel Paragon; parallel scheduling; parallel static scheduling; scheduling algorithms; Algorithm design and analysis; Computer science; Concurrent computing; Delay effects; Joining processes; NP-complete problem; Parallel algorithms; Processor scheduling; Scheduling algorithm;
Conference_Titel :
Parallel Architectures, Algorithms, and Networks, 1996. Proceedings., Second International Symposium on
Conference_Location :
Beijing
Print_ISBN :
0-8186-7460-1
DOI :
10.1109/ISPAN.1996.509009
