parallelApply -- apply a function to each element in parallel

Synopsis

Usage:

parallelApply(L,f)
Inputs:
- L, a basic list,
- f, a function,
Optional inputs:
- Strategy => ..., default value null, null or a string,
Outputs:
- a list or a basic list, If the Strategy option is null, then this behaves like toList apply(L,f), with the result computed in parallel in chunks using all cores. If it is the string "raw", then this behaves like apply(apply(L, e -> schedule(f, e)), taskResult).

Description

If the option Strategy is given the string "raw", then a separate task is created for each element of L. L is not split into chunks, allowableThreads is used unchanged, and the result has the same class as L. Normally the default strategy (Strategy => null) is more efficient.

See parallel programming with threads and tasks for more information and an important warning about thread safety.

Ways to use parallelApply :

parallelApply(BasicList,Function)

For the programmer

The object parallelApply is a method function with options.