NVIDIA & Virginia Tech GPU Computing Workshop
Overview
Virginia Tech and NVIDIA are collaborating to host a two-day, hands-on workshop to survey the broad range of GPU accelerated applications across all domains of scientific research and engineering. The workshop will take place on Aug 14 and 15, and the location will be 2150 Torgersen Hall. This workshop will address topics such as why heterogeneous computing with GPUs is important to sustaining and advancing the state of the art in scientific and research computing. 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.
Participants will learn how to program GPUs via the use of libraries, OpenACC compiler directives, and CUDA programming. The workshop will incorporate hands-on exercises on Virginia Tech's HokieSpeed to ensure that participants will have ample opportunities to develop the skills required to use and develop GPU aware applications. Please bring your laptop to participate in hands-on exercises.
This workshop is recommended for graduate students, postdocs, researchers, and professors. Space is limited, reservations will be granted on a first come, first serve basis. Lunch will be provided. Interested parties can register here.
Agenda
Wednesday, August 14
- 09:00-09:45 Overview & Introduction to GPU Computing (Feng, VT)
- 09:45-11:00 Accessing HokieSpeed and Set-Up (McClure, VT)
- 11:00-12:00 Intro to GPU Computing: Apps, Libraries, and OpenACC (Crovella, NVIDIA)
- 12:00-01:00 Lunch
- 01:00-01:30 Library Porting. Example: Using CUBLAS and CUSPARSE in a CG Solver [videos]
- 01:30-02:00 Hands-On Session 1: CG Solver [videos]
- 02:00-03:30 Intro to OpenACC [videos]
- 03:30-05:00 Hands-On Session 2: OpenACC Grid Example [videos]
Thursday, August 15
- 09:00-10:00 Hands-On Session 3: OpenACC Jacobian Relaxation Example [videos]
- 10:00-11:00 Intro to CUDA [videos]
- 11:00-12:00 CUDA Basic Optimizations [videos]
- 12:00-01:00 Lunch
- 01:00-02:00 Hands-On Session 4: CUDA hello world, grid example in CUDA C [videos]
- 02:00-03:30 Visual Profiler and Debugger Intro, Optimizing Grid Example [videos]
- 03:30-05:00 Q&A
Miscellaneous Documents
Setup Instructions
- Mac OS X users will be able to log into HokieSpeed using the 'Terminal' application, which is located under Applications > Utilities
- Windows users should do the following in advance of the workshop:
- Go to the following link: http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html
- Download 'putty.exe'
- Double click the executable that is downloaded
- Click 'Connection' on the left-hand side
- Set 'seconds between keepalives' to '30'
- Your username and password will be provided to you at the workshop
Many of the Folk Assisting with Instructions
Pictures