Data Parallelism
The functional and list-oriented characteristics of the Wolfram Language allow it to provide immediate built-in data parallelism, automatically distributing computations across available computers and processor cores.
Parallelize — perform a computation using automatic parallelization
ParallelMap — map a function in parallel over the elements of a list
ParallelTable — generate a table in parallel
ParallelDo ▪ ParallelSum ▪ ParallelProduct ▪ ParallelArray ▪ ParallelCases ▪ ParallelSelect
ParallelCombine — evaluate expressions in parallel and combine their results
Setup »
DistributeDefinitions — distribute definitions needed for parallel computations
ParallelNeeds — load a package into all parallel subkernels
ClearDistributedDefinitions — clears previously distributed definitions
Tuning
Method — how computations are distributed ("CoarsestGrained", "FinestGrained", ...)