DocumentCode :
1606551
Title :
The impact of synchronization and granularity on parallel systems
Author :
Chen, Ding-Kai ; Su, Hong-Men ; Yew, Pen-Chung
Author_Institution :
Center for Supercomput. Res. & Dev., Illinois Univ., Urbana, IL, USA
fYear :
1990
Firstpage :
239
Lastpage :
248
Abstract :
A study is made of the impact of synchronization and granularity on the performance of parallel systems using an execution-driven simulation technique. It is found that, even though there can be a lot of parallelism at the fine-grain level, synchronization and scheduling strategies determine the ultimate performance of the system. Loop-iteration-level parallelism seems to be a more appropriate level when those factors are considered. Barrier synchronization and data synchronization at the loop-iteration level are also studied. It is found that both schemes are needed for a better performance
Keywords :
parallel architectures; performance evaluation; synchronisation; barrier synchronisation; data synchronization; execution-driven simulation; granularity; loop-iteration level; parallel systems; parallelism; performance; scheduling; synchronization; Hardware; NASA; Parallel processing; Program processors; Research and development; Scheduling; Statistics; System performance; US Department of Energy; VLIW;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Architecture, 1990. Proceedings., 17th Annual International Symposium on
Conference_Location :
Seattle, WA
Print_ISBN :
0-8186-2047-1
Type :
conf
DOI :
10.1109/ISCA.1990.134531
Filename :
134531
Link To Document :
بازگشت