You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Parallel and Distributed Systems at SFU (431)
How do we know how many CUDA threads to launch? wouldn't we need to know how many cuda cores are available for the csil computer?
Thank you
Arrvindh Shriraman
unread,
Mar 28, 2020, 4:33:45 PM3/28/20
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Parallel and Distributed Systems at SFU (431)
You distribute based on problem size (not CUDA cores).
Thread blocks is 1D or 2D (limited by the amount of shared memory you want to cumulatively use across all threads in block). Typically thread blocks would be 128 or 256 threads.