如何用 C++ 函数设计和实现可伸缩的并发系统？

#include <vector>
#include <thread>
#include <mutex>
#include <atomic>

std::mutex sum_mutex;
std::atomic_int total_sum;

void sum_partial(const std::vector<int>& numbers, size_t start, size_t end) {
  int partial_sum = 0;
  for (size_t i = start; i < end; ++i) {
    partial_sum += numbers[i];
  }

  // 使用锁保护共享变量
  std::lock_guard<std::mutex> lock(sum_mutex);
  total_sum += partial_sum;
}

int main() {
  std::vector<int> numbers = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
  const size_t num_threads = 4;
  std::vector<std::thread> threads;

  // 分割向量并创建线程执行并行求和
  const size_t chunk_size = numbers.size() / num_threads;
  for (size_t i = 0; i < num_threads; ++i) {
    size_t start = i * chunk_size;
    size_t end = (i + 1) * chunk_size;
    threads.emplace_back(sum_partial, std::ref(numbers), start, end);
  }

  // 等待所有线程完成
  for (auto& thread : threads) {
    thread.join();
  }

  // 打印总和
  std::cout << "Total sum: " << total_sum << std::endl;
  return 0;
}