Know about the openmp, recently wrote a summation of the program of array elements, but found that after running more slowly than serial sum, a lot of Internet, a variety of don't know the root cause, I'm quad-core CPU

 
Void parallel_sum () 
{
STD: : vector Nums (1000); 
For (int I=1; I & lt;=1000; + + I) 
Nums [I - 1]=I; 

Int sum=0; 

# pragma omp parallel for reduction (+ : sum) 
For (int I=0; I & lt; 1000; + + I) 
Sum=sum + nums [I]; 
} 

Consult!

CodePudding user response:

Don't Shared memory address,

CodePudding user response:

reference 1st floor GKatHere response:

don't Shared memory address,

mean sum, why