为什么这个OpenMP程序比单线程慢？ [英] Why is this OpenMP program slower than single-thread?

问题描述

请看这段代码。

单线程程序： http://pastebin.com/KAx4RmSJ 。编译：

g ++ -lrt -O2 main.cpp -o nnlv2

具有openMP的多线程： http://pastebin.com/fbe4gZSn
编译为：

g ++ -lrt -fopenmp -O2 main_openmp.cpp -o nnlv2_openmp

我在双核系统上测试了它（所以我们有两个并行运行的线程）。但是多线程版本比单线程版本慢（并且显示不稳定的时间，尝试运行它几次）。怎么了？

一些测试：

单线程：

 层神经元输入---时间（ns）
 
 10 200 200 --- 1898983 
 
 10 500 500 --- 11009094 
 
 10 1000 1000 --- 48116913 
  

多线程：

 层神经元输入---时间（ns）
 
 10 200 200 --- 2518262 
 
 10 500 500 --- 13861504 
 
 10 1000 1000 --- 53446849 
  
解决方案
是你的目标这里研究OpenMP，还是让你的程序更快？如果是后者，那么写入乘法代码，减少遍历次数，并结合SIMD将是更值得的。
 
 
 
步骤1：组合循环并使用乘法 - add：
  //完全删除变量'temp'
 for（int i = 0; i< LAYERS; i ++）
 {
 for（int j = 0; j  {
 outputs [j] = 0; 
 
 for（int k = 0，l = 0; l  {
 outputs [j] + = inputs [ ] [k]; 
} 
 
 outputs [j] = sigmoid（outputs [j]）; 
} 
 
 std :: swap（inputs，outputs）; 
} 
  
 
Please look at this code. 

Single-threaded program: http://pastebin.com/KAx4RmSJ. Compiled with: 

  
g++ -lrt -O2 main.cpp -o nnlv2
Multithread with openMP: http://pastebin.com/fbe4gZSn
Compiled with: 

  
g++ -lrt -fopenmp -O2 main_openmp.cpp -o nnlv2_openmp
I tested it on a dual core system (so we have two threads running in parallel). But multi-threaded version is slower than the single-threaded one (and shows unstable time, try to run it few times). What's wrong? Where did I make mistake?

Some tests:

Single-thread:
Layers Neurons Inputs --- Time (ns)

10 200 200 --- 1898983

10 500 500 --- 11009094

10 1000 1000 --- 48116913
Multi-thread:
Layers Neurons Inputs --- Time (ns)

10 200 200 --- 2518262

10 500 500 --- 13861504

10 1000 1000 --- 53446849
I don't understand what is wrong.
解决方案
Is your goal here to study OpenMP, or to make your program faster?  If the latter, it would be more worthwhile to write multiply-add code, reduce the number of passes, and incorporate SIMD.

Step 1: Combine loops and use multiply-add:
// remove the variable 'temp' completely
for(int i=0;i<LAYERS;i++)
{
  for(int j=0;j<NEURONS;j++)
  {
    outputs[j] = 0;

    for(int k=0,l=0;l<INPUTS;l++,k++)
    {
      outputs[j] += inputs[l] * weights[i][k];
    }

    outputs[j] = sigmoid(outputs[j]);
  }

  std::swap(inputs, outputs);
}


                        
这篇关于为什么这个OpenMP程序比单线程慢？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！