Parallel Accumulate using tbb::parallel_reduce

readme.md

Call this function the same way you would with std::accumulate:

std::vector<int> v{1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
int sum = parallelAccumulate(v.begin(), v.end(), 0);

Using Intel TBB (Threading Building Blocks).

parallelAccumulate.cpp

#include <numeric>                // std::accumulate
#include "tbb/parallel_reduce.h"
#include "tbb/blocked_range.h"

template <class Summable, class SummableIt>
Summable parallelAccumulate(
    SummableIt beginIt, 
    SummableIt endIt, 
    const Summable initial
  ) const {
  Summable* start = &(*beginIt);
  Summable* end = &(*endIt);
  return tbb::parallel_reduce(
    tbb::blocked_range<Summable*>(start, end), // range
    initial, // identity; initial value
    [](const tbb::blocked_range<Summable*>& chunk, Summable initial)->Summable {
      return std::accumulate(chunk.begin(), chunk.end(), initial);
    }, // function to accumulate subrange
    std::plus<Summable>() // reduction operator
  );
}