USC Information Technology Services and NVIDIA are organizing a two day High Performance Computing event. NVIDIA GPUs are the world’s fastest and most efficient accelerators delivering world record scientific application performance. NVIDIA’s CUDA Technology is the most pervasive parallel computing model, used by over 250 scientific applications and over 150,000 developers worldwide.
This Programming Workshop will focus on introducing scientific computing programming utilizing NVIDIA GPUs to accelerate applications. The workshop will introduce programming techniques using CUDA and OpenACC paradigms as well as optimization, profiling, and debugging methods for GPU programming. Following topics will be covered: Introduction to OpenACC, Introduction to CUDA, CUDA Libraries, and CUDA tools such as NVIDIA Visual Profiler.
Who is it for: Graduate Students, Postdocs, Researchers, and Professors.
Why you should attend: NVIDIA GPUs are the world’s fastest and most efficient accelerators delivering world record scientific application performance. NVIDIA’s CUDA Technology is the most pervasive parallel computing model, used by over 250 scientific applications and over 150,000 developers worldwide.
Day 1, June 12th - 9AM to 4PM
1. CUDA 101 - A basic introduction to the C/C++ CUDA language
2. CUDA Basic optimizations - Covering GPU architecture and key optimizations (threads, memory) for GPU performance.
3. GPU Computing with OpenACC - Introduction
4. GPU Computing with OpenACC - Advanced - Development, analysis, and optimization of jacobian relaxation application.
No previous GPU programming experience is required. However, beginner-level C and Linux experience will be expected.
In preparation, you may watch a few short (approx. 5 min) YouTube videos on introductory GPU programming topics. These CUDACasts can be found here: http://www.youtube.com/user/NVIDIADeveloper
Day 2, June 13th - 9AM to 12PM
1. CUDA Libraries - Introduction to CUBLAS and CUSPARSE, with a walkthrough sample application porting of a Conjugate Gradient Solver.
2. CUDA Tools - CUDA-GDB, Visual Profiler, Insight Eclipse Edition (IDE)
IMPORTANT: Please bring your laptop to participate in hands-on exercises. No GPU in your laptop is required.
This event is only for USC faculty, staff scientists and graduate students.
Space is limited, reservations will be granted on a first come, first serve basis.
Lunch will be provided.