Hey friends,
As you know, we had our first session as per the schedule on 22-09-11.
It was a very interactive session and everyone showed interest in learning it further.
We started with definition of Parallel Processing and its levels followed by difference in various type parallel processing architectures.
Then we discussed GPU computing, its history, present scenario and the expected future.
We started CUDA by first understanding its programming model and then developed a O (1) code for "Squaring a numeric array using CUDA".
We tried to relate things with our previous knowledge of serial processing like as that through C. (PPT attached)
Those who attended the session and those who couldn't, I recommend them to go through this article
I am also attaching a PDF file containing description of some small CUDA projects done by people around the world. You can get idea for your project from them.
Our next session will be on 26-09-2011 (Monday) from 8:00 to 10:00 pm in the same room (3250). We will be discussing CUDA Threading Hierarchy and Execution Configuration. (No prior reading required)
Hope to see you there. Feel free to contact anytime for any query/doubt.
-Saiyedul Islam
ME CS 2011
PS - If anybody is thinking to join us then please be present in next session because after this session it would be too difficult to start over from beginning for any new entry.