Uses and abuses of Amdahl's law

Authors:
S. Krishnaprasad
Affiliations:
Mathematical, Computing, and Information Sciences, Jacksonville State University, Jacksonville, AL
Venue:
Journal of Computing Sciences in Colleges
Year:
2001

Citing 2
Cited 3

Computer architecture (2nd ed.): a quantitative approach

Computer architecture (2nd ed.): a quantitative approach
Computer organization and design (2nd ed.): the hardware/software interface

Computer organization and design (2nd ed.): the hardware/software interface

Automatic Task Scheduling / Loop Unrolling using Dedicated RTR Controllers in Coarse Grain Reconfigurable Architectures

IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 3 - Volume 04
Faster and More Complete Extended Static Checking for the Java Modeling Language

Journal of Automated Reasoning
Using the Cowichan problems to investigate the programmability of X10 programming system

Proceedings of the 2011 ACM SIGPLAN X10 Workshop

Quantified Score

Hi-index	0.00

Visualization

Abstract

Amdahl's law has been widely used by designers and researchers to get a rough estimate of performance improvement when alternate designs and implementations are attempted. It gives a simple relationship between the nature of performance improvement and the problem characteristics. The negative way the original law was stated [Amd67] contributed to a good deal of pessimism about the nature of parallel processing. But, after observing remarkable speedups in some large-scale applications, researchers in parallel processing started wrongfully suspecting the validity and usefulness of Amdahl's law. In this paper we present the many uses of Amdahl's law as well as some of its abuses.