Code Bytes #1 – PLINQ Basics

PLINQ /Parallel LINQ is part of the TPL (Task Parallel Library) and it makes your life easier when it comes to multi-core processor programming which is totally different from multithreading which allows more than one thread per process and you have no idea if they will be equally distributed across CPU cores. To use PLINQ your objects have to be in memory. This means you can’t use AsParallel on LINQ to SQL until you bring all your query results over to the local machine. When it comes to running your code in parallel the key to remember is that the AsParallel method is you friend. Every result that gets returned after your first call to AsParallel is always a ParallelQuery object. You can get more theory here. Now go code!

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Linq.Expressions;

namespace OliverCode.CodeBytes
{
    class ParallelLinqCB
    {
        static void Main(string[] args)
        {
            Action action = (int itemFromList) => Console.Write(itemFromList + ",");
            var lst = Enumerable.Range(1, 10);

             //PLINQ will decide on the number of processors to use in this run up to 64 threads (if you have that much)
             lst.AsParallel().Select(i => i * i).ForAll(action);
            Console.WriteLine();
            //My machine has 4 cores but i only need it to use up to 2 cores so I use the WithDegreeOfParallelism to restrict it
            //Or I could have use all 4 with the statement WithDegreeOfParallelism(4) hereby not letting PLINQ choose for me
            lst.AsParallel().WithDegreeOfParallelism(2).ForAll(action);

            Console.Read();
        }
    }
}

Cropped diagram courtesy of MSDN PFX (Parallel Programming Framework)

CodeProject