Intel Threading Building Blocks

Parallel_reduce with partitioner

Parallel_reduce has an optional third argument to specify a partitioner. See the section “Automatic grain size” for more information.

Example 3-15 extends Example 3-10 and Example 3-11 by using an auto_partitioner.

Example 3-15. Parallel sum with partitioner

#include "tbb/parallel_reduce.h"
#include "tbb/blocked_range.h"

using namespace tbb;

struct Sum {
    float value;
    Sum() : value(0) {}
    Sum( Sum& s, split ) {value = 0;}
    void operator()( const blocked_range<float*>& range ) {
        float temp = value;
        for( float* a=range.begin(); a!=range.end(); ++a ) {
            temp += *a;
        }
        value = temp;
    }
    void join( Sum& rhs ) {value += rhs.value;}
};
float ParallelSum( float array[], size_t n ) {
    Sum total;
    parallel_reduce( blocked_range<float*>( array, array+n ),
                     total, auto_partitioner() );
    return total.value;
}

Two important changes from Example 3-11 should be noted:

• The call to parallel_reduce takes a third argument, an auto_partitioner object.

• The blocked_range constructor is not provided with a grainsize parameter.