Databases
StackOverflow2013
Postsdbo
POFill histograms (array reduction) in parallel with OpenMP without using a critical section
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POFill histograms (array reduction) in parallel with OpenMP without using a critical section
    primarykey
    data
    text
    <p>I would like to fill histograms in parallel using OpenMP. I have come up with two different methods of doing this with OpenMP in C/C++. </p> <p>The first method <code>proccess_data_v1</code> makes a private histogram variable <code>hist_private</code> for each thread, fills them in prallel, and then sums the private histograms into the shared histogram <code>hist</code> in a <code>critical</code> section. </p> <p>The second method <code>proccess_data_v2</code> makes a shared array of histograms with array size equal to the number of threads, fills this array in parallel, and then sums the shared histogram <code>hist</code> in parallel. </p> <p>The second method seems superior to me since it avoids a critical section and sums the histograms in parallel. However, it requires knowing the number of threads and calling <code>omp_get_thread_num()</code>. I generally try to avoid this. Is there better way to do the second method without referencing the thread numbers and using a shared array with size equal to the number of threads?</p> <pre><code>void proccess_data_v1(float *data, int *hist, const int n, const int nbins, float max) { #pragma omp parallel { int *hist_private = new int[nbins]; for(int i=0; i&lt;nbins; i++) hist_private[i] = 0; #pragma omp for nowait for(int i=0; i&lt;n; i++) { float x = reconstruct_data(data[i]); fill_hist(hist_private, nbins, max, x); } #pragma omp critical { for(int i=0; i&lt;nbins; i++) { hist[i] += hist_private[i]; } } delete[] hist_private; } } void proccess_data_v2(float *data, int *hist, const int n, const int nbins, float max) { const int nthreads = 8; omp_set_num_threads(nthreads); int *hista = new int[nbins*nthreads]; #pragma omp parallel { const int ithread = omp_get_thread_num(); for(int i=0; i&lt;nbins; i++) hista[nbins*ithread+i] = 0; #pragma omp for for(int i=0; i&lt;n; i++) { float x = reconstruct_data(data[i]); fill_hist(&amp;hista[nbins*ithread], nbins, max, x); } #pragma omp for for(int i=0; i&lt;nbins; i++) { for(int t=0; t&lt;nthreads; t++) { hist[i] += hista[nbins*t + i]; } } } delete[] hista; } </code></pre> <p><strong>Edit:</strong> Based on a suggestion by @HristoIliev I have created an improved method called <code>process_data_v3</code></p> <pre><code>#define ROUND_DOWN(x, s) ((x) &amp; ~((s)-1)) void proccess_data_v2(float *data, int *hist, const int n, const int nbins, float max) { int* hista; #pragma omp parallel { const int nthreads = omp_get_num_threads(); const int ithread = omp_get_thread_num(); int lda = ROUND_DOWN(nbins+1023, 1024); //1024 ints = 4096 bytes -&gt; round to a multiple of page size #pragma omp single hista = (int*)_mm_malloc(lda*sizeof(int)*nthreads, 4096); //align memory to page size for(int i=0; i&lt;nbins; i++) hista[lda*ithread+i] = 0; #pragma omp for for(int i=0; i&lt;n; i++) { float x = reconstruct_data(data[i]); fill_hist(&amp;hista[lda*ithread], nbins, max, x); } #pragma omp for for(int i=0; i&lt;nbins; i++) { for(int t=0; t&lt;nthreads; t++) { hist[i] += hista[lda*t + i]; } } } _mm_free(hista); } </code></pre>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    plurals
    1. PL
      singulars
      1. LTLinked
    1. PL
      singulars
      1. LTLinked
    2. PL
      singulars
      1. LTLinked
    3. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POFill histograms (array reduction) in parallel with OpenMP without using a critical section
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POFill histograms (array reduction) in parallel with OpenMP without using a critical section
      1. USZ boson
      1. VTFavorite
    3. VO
      singulars
      1. POFill histograms (array reduction) in parallel with OpenMP without using a critical section
      1. This table or related slice is empty.
      1. VTUpMod
    1. COCould you please explain why you are using nested parallel regions? (I am referring to your process_data_v1 approach). Maybe I am not understanding something, but according to your code, it seems to me that you are asking for Nthreads**2. It is to say, you are asking for more resources than the available ones. Is that correct? In other words, could you explain the behaviour of parallel regions inside the outer one? Thanks...
      singulars
      1. POFill histograms (array reduction) in parallel with OpenMP without using a critical section
      1. USAlejandro
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload