Gustafson's law states the following:

Here, as we indicated in the equation the following applies:

• P is the number of processors.
• S is the speedup factor.
• α is the non-parallelizable fraction of any parallel process.

Gustafson's law is in contrast to Amdahl's law, which, as we described, assumes that the overall workload of a program does not change with respect to the number of processors.

In fact, Gustafson's law suggests that programmers first set the time allowed for solving a problem in parallel and then based on that (that is time) to size the problem. Therefore, the faster the parallel system is, the greater the problems that can be solved over the same period of time. 

The effect of Gustafson's law was to direct the objectives of computer research towards the selection or reformulation of problems in such a way that the solution of a larger problem would still be possible in the same amount of time. Furthermore, this law redefines the concept of efficiency as a need to reduce at least the sequential part of a program, despite the increase in workload.