Performance analysis and Tuning for GPGPUs

Held in conjunction with the 43rd Annual IEEE/ACM International Symposium on Microarchitecture

KACB 1447, Georgia Tech, Atlanta, Afternoon Sunday December 5, 2010


The goal of this half-day tutorial is to provide an introduction to the analysis, modeling, and tuning of performance for general-purpose graphics processing unit (GPGPU) platforms. Such systems feature prominently in current and future NSF TeraGrid acquisitions, including Lincoln at NCSA and the upcoming XD Keeneland system at ORNL, so that it is becoming increasingly important for developers to understand how best to use GPU resources. Fortunately, there is a wide variety of high-quality introductory material available on basic GPU programming, as well as numerous examples, making it easy to get started using GPUs. However, we believe there is far less such guidance available on the more advanced topics of performance modeling, analysis, and tuning, so that efficient use of GPUs remains a challenge for most developers. Our proposed tutorial aims to meet this need.

Tutorial Schedule

  • (1.5 hour) Part #1.0: Basic CUDA and GPU architecture, presenter Kim
  • (1 hour) Part #1.5: Performance Principles, Presenter Vuduc
  • (0.5 hour) Part #2.0: Analysis Tuning, Presenter Vuduc
  • (10 min) Part 2.5: Metrics and Analysis Tuning, presenter Kim
  • (10 min) Closing


    To register click here. For more details on MICRO-43, go here


    Hyesoon Kim (Georgia Tech)
    Richard Vuduc (Georgia Tech)