research ideas to further learn

Accelerated Computing CUDA CUDA Programming and Performance
1

I am unsure if I am posting this in the right forum. Over the summer, I taught myself CUDA C. I got into it because of the nVIDIA performance claims were simply unbelievable. I became I believable after running the demos from the SDK and from porting some of my old previous programs over.

Next week I will start another semester of my undergrad studies. My major is computer science. One of the classes I am taking is independent research where I will focus on CUDA C. Does anyone have any ideas on what I can concentrate on? Someone suggested see how many digits I can compute PI to because recently another undergrad student calculated PI to over 4 trillion digits [ [url=“y-cruncher - A Multi-Threaded Pi Program”]http://www.numberworld.org/y-cruncher/[/url] ].

I really don’t have any ideas so if you have any suggestions then I’m all ears. My goal in doing the research is to further understand the CUDA architecture and to be a better parallel programmer. One of my interests in computer science is security so I am currently looking online to see what will be a good topic.

Thanks for any input.

2

Take a look at several 3rd party libraries that are using CUDA to accelerate security applications. As for me, you can take a look at CUVI Lib which is a recently released GPU Accelerated Computer Vision and Imaging library. Let me know if this interests you and we can find a way in which you can contribute to it.

3

Take a look at several 3rd party libraries that are using CUDA to accelerate security applications. As for me, you can take a look at CUVI Lib which is a recently released GPU Accelerated Computer Vision and Imaging library. Let me know if this interests you and we can find a way in which you can contribute to it.