how can I do parallel reduction approach to combine the partial sums in c
I have to do partial sums using parallel reduction approach in C. but I doesn't have any idea about it. So, I need guidance of Community to achieve this.
What I need to achieve: for example, computational thread, then in first reduction step add two element in array, thread 4 should wait for thread 0 done, same as thread 5 has to wait for thread 1 done, thread 6 waits for thread 2 & thread 7 should waits for thread 3 done .
now in second step, , thread 6 waits for thread 4 finished, and thread 7 waits for thread 5 finished. , thread 6 waits for thread 4 finished, and thread 7 waits for thread 5 finished.
In the third step, thread 7 waits for thread 6 done. then need to print whole array
Please help me, give me a guidance to achieve this one.
Comments
Post a Comment