CUDA Programming – C/C++ Basics


Speaker: Mr. Oren Tropp (Sagivtech)
“Prace Conference 2014”,
Partnership for Advanced Computing in Europe,
Tel Aviv University,



  1. thank you for your video. it is very helpful/

    here is the man who teach CUDA in korea.

    he is AI proffessional lecturer teaching Machine learning, CUDA, Aibril, etx

    website ''

  2. Hello, when i want to add thread into global function, will i add third value on fucntion call like -> <<<N,1,threadNumber>>>

  3. "Compute Unified Device Architecture" ?? Don't just repeat any old BS you learn. CUDA is not an acronym! It's the name that was chosen by the original author(s).

  4. Seems like a good lecture but wow these students are really stupid. It says in the first few slides C/C++ experience is expected. Half the class shouldn't even be in that room, they are just ruining the talk with their stupidity.

  5. Seems like a lot of time was lost because of Nvidia's examples – I noticed that in the documentation for the hello world program that you compile with nvcc, that the necessary include statement "#include <stdio.h>" at the beginning of the program is completely missing. I think this instructor lost a lot of time to that.

  6. 24:40 _global_ attribute indicates that the function can be called either from host or from device, not only from host. You could call _global_ function from _device_ or _host_ function. _host_ keyword means that function can be invoked only from host.


Please enter your comment!
Please enter your name here