What is Amdahl's law?

The law that when only part of a program is parallelized, overall speedup is limited by the part that cannot be. For example, parallelizing 90% gives at most a 10x speedup.

Field: CPUs and architecture

Amdahl's law in the explanations

Questions that mention Amdahl's law (1)

  1. If 90% of a program can be parallelized, what is the theoretical maximum speedup when it runs on an unlimited number of processors?

Related terms

Branch prediction Cache coherency Cache memory DMA (direct memory access) DVFS (dynamic voltage and frequency scaling) GPU Instruction set architecture (ISA) Interrupt IPC Microcontroller

Back to the glossary