Questions and Answers :
GPU applications :
Cuda and Astropulse re-process behaviour
Message board moderation
Author | Message |
---|---|
![]() Send message Joined: 22 Nov 05 Posts: 8 Credit: 340,251 RAC: 0 ![]() |
I am running SETI on a dual core system. I have a new GPU that runs CUDA. The speed is phenominal. (but never fast enough:) - I have a 1400+ hour unit running in astropulse. Yesterday it reached the end of processing and then BACKUP UP about 166 points and began reprocessing that last bit of data again. This morning it reached the end and backed up again. Is this normal for astropulse or should I assume there is a data glitch and abort the processing? |
![]() Send message Joined: 22 Nov 05 Posts: 8 Credit: 340,251 RAC: 0 ![]() |
Okay - The data finished processing and has been uploaded. So, we can assume that the reprocessing of the last 200 or so records can be normal using CUDA. Carry on having fun. |
![]() Send message Joined: 22 Nov 05 Posts: 8 Credit: 340,251 RAC: 0 ![]() |
After that, only one SETI unit is being processed. The other is idle (2 CPUs) |
![]() Send message Joined: 22 Nov 05 Posts: 8 Credit: 340,251 RAC: 0 ![]() |
Another autopulse unit has downloaded. Still only one work unit is being processed although I have two processors. |
OzzFan ![]() ![]() ![]() ![]() Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 ![]() ![]() |
There are several threads here in the CUDA forum that explain why one core remains idle when using CUDA. |
©2023 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.