- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have a problem when i try parallelization with OpenMP on a Intel Core 2 Quad(processor X5355)
when i'm using a single proccessor the running time is : 48 minutes, but if i use four proccessors the running time is 23 minutes, but if use 5, 6,7,8 the running time is the same 21 MIN !!!... when i'm using 5 to 8 processors the paralelizacion time does not decreases ...
what is the problem..??
Link Copied
2 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
For the multi-processor system, did you use the OMP_NUM_THREADS setting before you ran your executable, as in this example?
> set OMP_NUM_THREADS=4
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Setting that environment variable is usually not necessary, as it defaults to the number of execution units (cores). My guess is that the program and its data simply don't scale past four threads, based on size of data, work done in each thread, etc.
You can get more information by running it under Intel Thread Profiler and see what it's doing.
You can get more information by running it under Intel Thread Profiler and see what it's doing.
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page