canUseParallelPool

Verify that parallel functions can use a parallel pool

Since R2020b

Syntax

tf = canUseParallelPool()

Description

tf = canUseParallelPool() returns a logical value indicating if parallel functions can create and use a parallel pool.

The function returns logical 1 (true) if Parallel Computing Toolbox™ is installed and licensed for use, a default parallel pool is configured and supported, and automatic creation of parallel pools is enabled. Otherwise, the function returns logical 0 (false). This function does not create a parallel pool.

Use canUseParallelPool to check if parallel functions such as parfor (Parallel Computing Toolbox) and parfeval (Parallel Computing Toolbox) can use a parallel pool. This enables you to avoid executing code that requires a parallel pool if one is not available.

Even if this function returns true, you can still receive an error when you try to create a parallel pool if your parallel cluster is not properly configured or cannot be contacted.

Examples

Check Parallel Availability

Verify that you can use a parallel pool before executing code that runs in parallel. If a pool is not available, run the code in serial.

N=100;

if canUseParallelPool
    f(1:N) = parallel.FevalFuture;
    for i = 1:N
        f(i) = parfeval(@rank,1,magic(i));
    end
    results = fetchOutputs(f);
else
    results = zeros(1,N);
    for i = 1:N
        results(i) = rank(magic(i));
    end
end

If canUseParallelPool returns true, then parfeval creates a parallel pool and the code runs in parallel. Otherwise, the code runs in a normal for-loop.

Version History

Introduced in R2020b