2011-04-14 Keeneland Workshop

Workshop Tutorial 

The tutorial presentations can be found at:

 

On April 14-15, 2011, the Keeneland project and the Georgia Tech NVIDIA CUDA Center of Excellence presented a two day tutorial on GPU heterogeneous processing for computational science.

 

Tutorial topics included

  1. Architectural overview and motivation
  2. Introduction to CUDA
  3. Introduction to OpenCL
  4. Application case studies
  5. Overviews of GPU enabled libraries, correctness and performance tools, and compilers
  6. High level language interfaces, such as MATLAB, R, and Accelereyes, to GPUs
  7. Advanced optimization techniques for GPUs
  8. Hands-on time with the Keeneland Initial Delivery System on Friday (4/15) afternoon

Agenda

Thursday 4/14

Time/ Topic

08:00 Welcome and Introduction

08:15 Heterogeneous Architectures and KID Overview

08:45 Introduction to CUDA

10:15 Break

10:30 Introduction to OpenCL

11:30 Lunch

12:30 SDK Overview, incl debugging and performance

13:00 Libraries Overview: NVIDIA Libraries

13:40 Libraries Overview: MAGMA

14:00 Scalable Heterogeneous Computing

14:45 Break

15:00 Integrated Kernal Walk-throughs

15:45 Ocelot

16:15 Performance Analysis with Tau

17:00 Adjourn


 Friday 4/15

08:00 Announcements and Recap

08:05 Allinea DDT

08:35 Advanced Topics in CUDA, OpenCL

09:35 break

9:50  Additional Tools, Compilers, Libraries

10:30 Application Case Studies

11:30 Lunch

12:30 HANDS ON WITH KEENELAND - Bring your own code, use SHOC, or NVIDIA SDK

17:00 Adjourn
 

The Keeneland Workshop is sponsored by the Georgia Institute of Technology NVIDIA CUDA Center of Excellence and the Keeneland Project.  The Keeneland Project is funded by the National Science Foundation.