This workshop is now over

Slides presented by Mark Harris at the workshop

• Tesla CUDA Introduction
• CUDA Parallel
• Programming Model
• CUDA Toolkit Libraries
• Optimizing CUDA (full)
• Optimizing CUDA (short)
• CUDA Directions

Video of Workshop

We did video the workshop. Unfortunately the first hour or so is missing. The total size of the video is currently around 800MB. Subject to space we will make this available on-line. In the meantime if you desperately need to see the video you can email me (Alistair Rendell).

---

The workshop is free, but please email Alistair.Rendell@anu.edu.au for catering purposes. (We have capacity for at most 50 attendees)

Abstract:

This workshop and discussion forum aims to provide a detailed introduction to GPU computing with CUDA and NVIDIA Tesla computing solutions.

CUDA is NVIDIA's revolutionary parallel computing architecture that is built up on a unified Computing architecture and multiple software components. The architecture consists of an ISA and hardware compute engine. The available software tools include a C compiler and NVIDIA drivers for developers to build applications using C for CUDA, as well as useful libraries for high-performance computing (BLAS, FFT, etc.).

This half-day workshop will provide an introduction to the CUDA architecture, programming model, and the programming environment of C for CUDA, as well as an overview of the Tesla GPU architecture, a live programming demo, and strategies for optimizing CUDA applications for the GPU. The workshop will also include a brief presentation of some of the projects using CUDA within the ANU College of Engineering and Computer Science, and of the hardware requirements for getting started with CUDA.

Agenda

8:45	Arrival and registration
9:00	Introduction to NVIDIA CUDA and Tesla Sample Applications from ANU College of Engineering and Computer Science (Alistair Rendell) CUDA Parallel Programming Model and C for CUDA Live Programming Demo (part 1)
10:50	Coffee Break
11:10	Live Programming Demo(part 2) CUDA Toolkit and Libraries Optimizing Performance CUDA Directions Getting Started - CUDA enabled Hardware Options (Dragan Dimitrovici, Xenon Computer) General Q&A
13:00	Light lunch with discussion

Background

Mark Harris is in Canberra as part of a CSIRO organised workshop on "GPU Computing Workshop at Computational and Simulation Sciences Conference" that will be held on March 19th. As there is limited space for external participants at the CSIRO event, but also to engage with ANU students and discuss other projects that use CUDA, Mark has kindly agreed to repeat his presentation at the ANU.

More generally, within the ANU College of Engineering and Computer Science we have been working with CUDA on the GPU (and other special purpose processors, such at CellBE, or FPGAs) for a couple of years now and are interested in building up a local community with similar interests, hosting subsequent events, having ANU students work on interesting projects in this space etc. We will be discussing this with our colleagues in CSIRO, but in the meantime please feel free to contact me Alistair.Rendell@anu.edu.au (phone 02 6125 4386) if you have an interest in such things and are not attending on March 20th.

Bio: Mark Harris

Mark Harris is a Senior Developer Technology Engineer at NVIDIA, where he works with developers around the world on software for computer graphics and high-performance computing. His research interests include parallel algorithms, general-purpose computation on GPUs, physically based simulation, real-time rendering, and gastronomy. Mark earned his Ph.D. in computer science from the University of North Carolina at Chapel Hill in 2003 and his B.S. from the University of Notre Dame in 1998. Mark founded and maintains GPGPU.org , a web site dedicated to general-purpose computation on GPUs. Mark has recently moved to Brisbane after living in the United Kingdom for five years.