Science Atlas

How We Know What We Know
Sign In
Text size
100%
Theme
Theory

Amdahl's Law

Computer Science

Amdahl's law is a formula describing the theoretical limit on how much a computing task can be sped up by adding more parallel processors, given the proportion of the task that must still be carried out sequentially. Computer architect Gene Amdahl introduced the argument in a 1967 conference paper, showing that if some fraction of a program cannot be parallelized, that fraction alone sets a hard ceiling on the overall speedup achievable no matter how many additional processors are applied to the remaining parallel portion. The law is commonly illustrated with the example that a program only ten percent sequential can never run more than ten times faster, however many processors are added, since the sequential ten percent alone always takes a fixed minimum amount of time. Amdahl's law remains a foundational concept in computer architecture and parallel computing for reasoning about the practical returns of adding parallel hardware, informing the design of both multiprocessor systems and, more recently, multi-core consumer processors.

Facts
Proposed Year
1967 1
Connections

Belongs To

Sources
1. Wikidata: Amdahl's Law
lead
Quote, lead
It is named after computer scientist Gene Amdahl, and was presented at the American Federation of Information Processing Societies (AFIPS) Spring Joint Computer Conference in 1967.
View the Source
Comments (0)
No comments yet. Be the first to share a thought.
Reader Challenges (0)
No disputes yet. Spotted an error or a better source? Open the first one.