Title :
A message segmentation technique to minimize task completion time
Author :
Kim, Sukil ; Pande, Santoshkumar S. ; Agrawal, Dharma P. ; Mauney, Jon
Author_Institution :
North Carolina State Univ., Raleigh, NC, USA
fDate :
30 Apr-2 May 1991
Abstract :
Optimal partitioning of multiprocessor programs is a trade-off: as the granularity of subtasks of a parallel task increases, the communication overhead decreases but so does the total parallelism. The authors propose a new technique to determine the optimal segment size of messages between a producer and a consumer to minimize the overall execution time, and apply it to allocation of DOACROSS loops
Keywords :
parallel programming; programming theory; DOACROSS loops; communication overhead; consumer; execution time; message segmentation; multiprocessor programs; parallel task; producer; program partitioning; subtasks granularity; task completion time; Character generation; Communication channels; Concurrent computing; Context; Costs; Electronic mail; Hypercubes; Magnetic heads; Partitioning algorithms; Yarn;
Conference_Titel :
Parallel Processing Symposium, 1991. Proceedings., Fifth International
Conference_Location :
Anaheim, CA
Print_ISBN :
0-8186-9167-0
DOI :
10.1109/IPPS.1991.153830
