Speedup and Amdahl's law¶
Category: Foundations · Status: stub · Lessons: chapter 07, Parallelism
One line: How much faster more processors make a program is capped by the part that must still run sequentially: if a tenth of the work is serial, no number of cores gives more than ten times the speed.
Also called: Amdahl's law, Gustafson's law, speedup.
How it connects¶
Where to read more¶
- In the books: The Art of Concurrency, Clay Breshears — ch. 1, 'Want to Go Faster? Raise Your Hands if You Want to Go Faster!'
- In the books: Learn Concurrent Programming with Go, James Cutajar — ch. 1, 'Stepping into concurrent programming' → 'Increasing throughput'
- In the books: Hands-On Concurrency with Rust, Brian L. Troutwine — ch. 2, 'Sequential Rust Performance and Testing' → 'Diminishing returns'
- In the books: Java Concurrency in Practice, Brian Goetz, Tim Peierls, Joshua Bloch, Joseph Bowbeer, David Holmes, Doug Lea — ch. 11, 'Performance and Scalability' → 'Amdahl’s Law'
- In the books: Concurrent Programming on Windows, Joe Duffy — ch. 14, 'Performance and Scalability' → 'Speedup: Parallel vs. Sequential Code'
- In the books: Programming Concurrency on the JVM, Venkat Subramaniam — ch. 2, 'Division of Labor' → 'Speedup for the IO-Intensive App'
- Reference: Wikipedia: Amdahl's law ↗