Paper Info
Title
Overcoming Amdahl's Law by Distributing Workload Asymmetrically.
Abstract
Recently, many applications can be parallelized by using multicore platforms. In this paper, we propose a load balancing technique in parallelizing an application whose first module has data independency and its second module has data dependency. Instead of distributing the first module symmetrically over the multi-core platform, we distribute the workload asymmetrically. Based on the experimental results with compression/encryption application, we confirm that the asymmetric load balancing can provide a better performance than Amdahl's law computed with the typical symmetric load balancing.
Year
DOI
Venue
2012
10.1007/978-3-642-32692-9_58
Communications in Computer and Information Science
Keywords
Field
DocType
multicore,Amdahl's law,load balancing
Data dependency,Workload,Computer science,Load balancing (computing),Amdahl's law,Parallel computing,Encryption,Multi-core processor,Distributed computing
Conference
Volume
ISSN
Citations 
310
1865-0929
0
PageRank 
References 
Authors
0.34
6
3
Name
Order
Citations
PageRank
Heegon Kim1157.69
Sung-Ju Lee23511278.11
Yongwha Chung323347.32