Como: Acelerar corpos de loop pequenos

Quando um loop Parallel.For tem um corpo pequeno, ele pode executar mais lentamente do que o loop sequencial equivalente, como o loop para em c# e o loop Para no Visual Basic. O desempenho mais lento é causado pela sobrecarga envolvida no particionamento dos dados e do custo de invocação de um representante em cada iteração do loop. Para resolver esses cenários, a classe Partitioner fornece o método Partitioner.Create, que permite que você forneça um loop sequencial do corpo do representante, de modo que o representante seja chamado apenas uma vez por partição, em vez de uma vez por iteração. Para saber mais, veja Particionadores personalizados para PLINQ e TPL.

Exemplo

using System;
using System.Collections.Concurrent;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;

class Program
{
    static void Main()
    {

        // Source must be array or IList.
        var source = Enumerable.Range(0, 100000).ToArray();

        // Partition the entire source array.
        var rangePartitioner = Partitioner.Create(0, source.Length);

        double[] results = new double[source.Length];

        // Loop over the partitions in parallel.
        Parallel.ForEach(rangePartitioner, (range, loopState) =>
        {
            // Loop over each range element without a delegate invocation.
            for (int i = range.Item1; i < range.Item2; i++)
            {
                results[i] = source[i] * Math.PI;
            }
        });

        Console.WriteLine("Operation complete. Print results? y/n");
        char input = Console.ReadKey().KeyChar;
        if (input == 'y' || input == 'Y')
        {
            foreach(double d in results)
            {
                Console.Write("{0} ", d);
            }
        }
    }
}
Imports System.Threading.Tasks
Imports System.Collections.Concurrent

Module PartitionDemo

    Sub Main()
        ' Source must be array or IList.
        Dim source = Enumerable.Range(0, 100000).ToArray()

        ' Partition the entire source array. 
        ' Let the partitioner size the ranges.
        Dim rangePartitioner = Partitioner.Create(0, source.Length)

        Dim results(source.Length - 1) As Double

        ' Loop over the partitions in parallel. The Sub is invoked
        ' once per partition.
        Parallel.ForEach(rangePartitioner, Sub(range, loopState)

                                               ' Loop over each range element without a delegate invocation.
                                               For i As Integer = range.Item1 To range.Item2 - 1
                                                   results(i) = source(i) * Math.PI
                                               Next
                                           End Sub)
        Console.WriteLine("Operation complete. Print results? y/n")
        Dim input As Char = Console.ReadKey().KeyChar
        If input = "y"c Or input = "Y"c Then
            For Each d As Double In results
                Console.Write("{0} ", d)
            Next
        End If

    End Sub
End Module

A abordagem demonstrada neste exemplo é útil quando o loop executa uma quantidade mínima de trabalho. Como o trabalho se torna mais dispendioso computacionalmente, você provavelmente obterá um desempenho igual ou melhor usando um loop For ou ForEach com o particionador padrão.

Confira também