Parallel Computing: Amdahl's Law

The formula of Amdahl's Law is:

$$ {Speedup = \frac{1}{(1-p)+\frac{p}{n}}.} $$

In the above equation, $ {p} $ is the portion of the program that runs in parallel (a ratio from 0 to 1,) and $ {n} $ is the number of available CPU cores in the device (integer greater than or equal to 1.)

That is, to calculate the maximum possible speedup, you need to know (1) what chunk of the program will be run by threads on multiple CPU cores, and (2) how many CPU cores the computer has. Then, just plug-in the quantities to find the actual speedup.