I’ve previously written about parallelizing C/C++ for loops with OpenMP. Many of the concepts explained in that article also apply to C#. The main difference is that C# does not use OpenMP. For loops are instead parallelized using the Parallel.For function:
Parallel.For(0,n, i =>
{
result[i]=func(data[i]);
});
Parallel.ForEach is also available.
It might be necessary to set the number of threads equal to the number of cores:
Parallel.For(0,n, new ParallelOptions { MaxDegreeOfParallelism = Environment.ProcessorCount }, i =>
{
result[i]=func(data[i]);
});
According to Microsoft’s documentation, the workload is distributed dynamically, so there’s no need to work with different scheduling schemes.
