Compact Course: Introduction to GPU programming with CUDA
| When |
Dec 13, 2011 08:30 AM
to
Dec 14, 2011 05:00 PM |
|---|---|
| Where | PDC/KTH |
| Contact Name | Michael Schliephake |
| Contact Phone | +46 8 790 7822 |
| Add event to calendar |
|
Description
General-purpose graphics processing units (GPU) will provide a significant part of the computational power used in scientific computing applications. Mastering parallel computing with these devices is therefore very interesting and becoming increasingly important. This course will provide the basic knowledge and practical experience of parallel computing with GPU and how to use it for the development of programs in the field of scientific computing.
The lectures present an overview of hardware for parallel computing and the prevailing parallelization model using data parallelism. The CUDA programming model will be discussed in detail (CUDA threads, memories, floating-point computations ) as well as the advanced libraries helping in the application development (CUBLAS, CUFFT, CUSPARSE, CURAND).
Schedule
- Tuesday, 2011-12-13
- 08:30-09:00 Welcome, Lindstedtsvägen 5, room 4523
- 09:00-10:30 Lecture I, Lindstedtsvägen 5, room 4523
- 11:00-12:30 Lecture II, Lindstedtsvägen 5, room 4523
- 13:30-17:00 Computer Lab I, Lindstedtsvägen 3, room 4V2Röd
- Wednesday, 2011-12-14
- 09:00-10:30 Lecture III, Lindstedtsvägen 5, room 4523
- 11:00-12:30 Lecture IV, Lindstedtsvägen 5, room 4523
- 13:30-17:00 Computer Lab II, Lindstedtsvägen 3, room 4V2Röd
Prerequisites
It is necessary that you have a ready-prepared PDC account to attend the computer labs. Please apply for an accountin good time before the course if needed.The programming language used in the course will be C. However, FORTRAN developers are also encouraged to take part because there exist tools that allow to use CUDA in FORTRAN programs too.
Registration
Please send an e-mail to Michael Schliephake. The number of participants is limited to 20 persons.


