Amdahl's Law Calculator
Test cases replayed : 2 of 2 match · Methodology · No outside source listed
Find how much faster a program runs on more processors when only part of its work can run in parallel.
Result
Speedup (×)
Maximum speedup with unlimited processors (×)
- Parallel efficiency (speedup ÷ processors)
Estimate for planning only, not tax, legal, financial or medical advice. This calculator uses the method explained on this page.
This browser did not allow copying. The sentence is in the box below, already selected: copy it with your keyboard or the menu.
How to use this calculator
Fill in the fields in this order; each step takes you to its field.
- Parallel share of the work, in percent: enter a number from 0 to 100.
- Number of processors or cores: enter 1 or more.
- Read the result: Speedup (×); the panel also shows Maximum speedup with unlimited processors (×) and Parallel efficiency (speedup ÷ processors).
How this calculator works
Amdahl's law (Gene Amdahl, 1967): speedup S = 1 ÷ ((1 − p) + p ÷ N), where p is the share of the work that can run in parallel and N the number of processors.
As N grows, S approaches 1 ÷ (1 − p): the serial part sets a hard limit. With 100% parallel work the maximum is unlimited and shows as a dash.
The formula
| Symbol | Meaning | Unit |
|---|---|---|
| Parallel share of the work | % | |
| Number of processors or cores | ||
| Speedup (×) |
Worked example
| Parallel share of the work | 90% |
|---|---|
| Number of processors or cores | 8 |
| Maximum speedup with unlimited processors (×) | 10 |
|---|---|
| Parallel efficiency (speedup ÷ processors) | 58.82% |
Speedup (×) 4.7059
Each step uses the unrounded value of the step before it; figures on this page are rounded for display.
Frequently asked questions
What speedup do 8 cores give with 90% parallel code?
1 ÷ (0.1 + 0.9 ÷ 8) = 4.7059 times, far from 8.
What is the maximum speedup under Amdahl's law?
1 ÷ (1 − p): with 90% parallel work, never more than 10 times however many processors you add.