如何:使用 combinable 提高性能

本示例演示如何使用 Concurrency::combinable 类计算 std::array 对象中的质数和。 combinable 类通过消除共享状态来提高性能。

示例

下面的示例使用 std::accumulate 函数计算数组中质数元素的和。 在此示例中,aarray 对象,is_prime 函数将确定其输入值是否为质数。

prime_sum = accumulate(a.begin(), a.end(), 0, [&](int acc, int i) {
   return acc + (is_prime(i) ? i : 0);
});

下面的示例演示一种将上一示例并行化的低级方式。 此示例使用 Concurrency::parallel_for_each 算法以并行方式处理数组和 Concurrency::critical_section 对象,以同步对 prime_sum 变量的访问。 此示例不会缩放,因为每个线程都必须等待共享资源变为可用。

critical_section cs;
prime_sum = 0;
parallel_for_each(a.begin(), a.end(), [&](int i) {
   cs.lock();
   prime_sum += (is_prime(i) ? i : 0);
   cs.unlock();
});

下面的示例使用 combinable 对象提高上一示例的性能。 该示例不再需要同步对象;由于 combinable 对象使每个线程能够单独执行其任务,因此该示例将可以缩放。

通常分两个步骤使用 combinable 对象。 第一步,通过并行执行工作产生一系列细化的计算。 第二步,将这些计算合并(或减少)为一个最终结果。 此示例使用 Concurrency::combinable::local 方法获取对局部和的引用。 然后,此示例使用 Concurrency::combinable::combine 方法和 std::plus 对象将本地计算合并为最终结果。

combinable<int> sum;
parallel_for_each(a.begin(), a.end(), [&](int i) {
   sum.local() += (is_prime(i) ? i : 0);
});
prime_sum = sum.combine(plus<int>());

下面的完整示例按串行方式和并行方式分别计算质数的和。 该示例会将执行两个计算所需的时间输出到控制台。

// parallel-sum-of-primes.cpp
// compile with: /EHsc
#include <windows.h>
#include <ppl.h>
#include <array>
#include <numeric>
#include <iostream>

using namespace Concurrency;
using namespace std;

// Calls the provided work function and returns the number of milliseconds 
// that it takes to call that function.
template <class Function>
__int64 time_call(Function&& f)
{
   __int64 begin = GetTickCount();
   f();
   return GetTickCount() - begin;
}

// Determines whether the input value is prime.
bool is_prime(int n)
{
   if (n < 2)
      return false;
   for (int i = 2; i < n; ++i)
   {
      if ((n % i) == 0)
         return false;
   }
   return true;
}

int wmain()
{   
   // Create an array object that contains 200000 integers.
   array<int, 200000> a;

   // Initialize the array such that a[i] == i.
   int n = 0;
   generate(a.begin(), a.end(), [&] {
      return n++;
   });

   int prime_sum;
   __int64 elapsed;

   // Compute the sum of the numbers in the array that are prime.
   elapsed = time_call([&] {
      prime_sum = accumulate(a.begin(), a.end(), 0, [&](int acc, int i) {
         return acc + (is_prime(i) ? i : 0);
      });
   });   
   wcout << prime_sum << endl;   
   wcout << L"serial time: " << elapsed << L" ms" << endl << endl;

   // Now perform the same task in parallel.
   elapsed = time_call([&] {
      combinable<int> sum;
      parallel_for_each(a.begin(), a.end(), [&](int i) {
         sum.local() += (is_prime(i) ? i : 0);
      });
      prime_sum = sum.combine(plus<int>());
   });
   wcout << prime_sum << endl;
   wcout << L"parallel time: " << elapsed << L" ms" << endl << endl;
}

下例是四处理器计算机的输出结果。

1709600813
serial time: 6178 ms

1709600813
parallel time: 1638 ms

编译代码

若要编译代码,请复制代码并将其粘贴到 Visual Studio 项目中或一个名为 parallel-sum-of-primes.cpp 的文件中,然后在 Visual Studio 命令提示符窗口中运行以下命令。

cl.exe /EHsc parallel-sum-of-primes.cpp

请参见

参考

combinable 类

critical_section 类

概念

并行容器和对象