Speedup - Definition and Overview

Speedup, in parallel computing, refers on how much a parallel algorithm is faster than a corresponding sequential algorithm.

It is defined by the following formula:

<math>S_p = \frac{T_1}{T_p}<math>

where:

• p is the number of processors
• <math>T_1<math> is the execution time of the sequential algorithm
• <math>T_p<math> is the execution time of the parallel algorithm (with p processors)

Ideal speedup is obtained when <math>S_p = p<math>.

See also

• Amdahl's law