Performance
The speedup of a program as a result of parallelization is given by Amdahl's law. Amdahl's Law states that potential program speedup is defined by the fraction of code (P) that can be parallelized: 1/(1-P)
If the code cannot be broken up to run over multiple processors, P = 0 and the speedup = 1 (no speedup). If it is possible to break up the code to be perfectly parallel then P = 1 and the speedup is infinite (in theory, although other factors such as scalability and complexity limit this possibility). Thus, there is an upper bound on the usefulness of adding more parallel execution units.
Gustafson's law is a law closely related to Amdahl's law but doesn’t make as many assumptions and tries to model these factors in the representation of performance. The equation can be modelled by S(P) = P − α * (P − 1) where P is the number of processors, S is the speedup, and α the non-parallelizable part of the process.
Read more about this topic: Distributed GIS
Famous quotes containing the word performance:
“To vote is like the payment of a debt—a duty never to be neglected, if its performance is possible.”
—Rutherford Birchard Hayes (1822–1893)
“They say all lovers swear more performance than they are able, and yet reserve an ability that they never perform; vowing more than the perfection of ten, and discharging less than the tenth part of one.”
—William Shakespeare (1564–1616)
“So long as the source of our identity is external—vested in how others judge our performance at work, or how others judge our children’s performance, or how much money we make—we will find ourselves hopelessly flawed, forever short of the ideal.”
—Melinda M. Marshall (20th century)