HOME
ABOUT
BLUE WATERS
CONTACT
Overview
Agenda
Travel Info
Contact Info
Discussion Board
Photos
Streaming Video Info

Agenda

Please note, all times are shown in Central Daylight Time (UTC-5).

Materials from the 2008 course on many-core processors are available for students to review.

Monday, August 10, 2009
8:00 a.m. - 11:00 a.m.
Registration / Check-in

7:30 a.m.
Continental Breakfast
9:00 a.m.
Beginner's Tutorial on Many-Core Processors, Multi-Core Processors, and C Programming (optional)
Wen-Mei Hwu and Nady Obeid, University of Illinois

Streaming video is available (Microsoft .asx)
11:00 a.m.
Lunch
11:50 a.m.
Welcome
Noon
Introduction to Many-Core Processors
David Kirk, NVIDIA

Streaming video is available (Microsoft .asx)
1:15 p.m.
Break
1:30 p.m.
Introduction to CUDA
Wen-Mei Hwu, University of Illinois, and David Kirk, NVIDIA

Streaming video is available (Microsoft .asx)
2:45 p.m.
Break
3:00 p.m.
Multidisciplinary Panel
Justin Haldar, Guochun Shi, Aaron Shinn, and John Stone, University of Illinois

Streaming video is available (Microsoft .asx)
4:30 p.m.
Adjourn
4:30 p.m.
EVL only — Rapid-fire session
The rapid-fire session is designed to help attendees connect with others with similar domain expertise. Each attendee will have up to 2 minutes to share their name, institutions, and a brief description of their research.
5:00 p.m.
University of Michigan only — Reception/Rapid-fire session
The rapid-fire session is designed to help attendees connect with others with similar domain expertise. Each attendee will have up to 2 minutes to share their name, institutions, and a brief description of their research.
Tuesday, August 11, 2009
7:30 a.m.
Continental Breakfast
8:00 a.m.
CUDA Threading Model
Wen-Mei Hwu, University of Illinois, and David Kirk, NVIDIA

Streaming video is available (Microsoft .asx)
9:15 a.m.
Break
9:30 a.m.
CUDA Memory Model
Wen-Mei Hwu, University of Illinois, and David Kirk, NVIDIA

Streaming video is available (Microsoft .asx)
11:00 a.m.
Lunch
Noon
Hands-on Lab
    -- Understanding the labs
    -- CUDA lab manual
    -- CUDA lab code
    -- CUDA lab matrix data
    -- CUDA lab solutions
Nady Obeid, Xiao-Long Wu and I-Jui "Ray" Sung, University of Illinois
1:30 p.m.
NCSA only — Tour (optional)
This optional one-hour tour will feature the GPU cluster students are using during the summer school and a trip through the NCSA's machine room. Space is limited, so please sign up for a tour at the check-in desk.

At all other sites, and for those at NCSA not taking the tour, hands-on lab will continue.
2:45 p.m.
Break
3:00 p.m.
Keynote: Algorithm Design for Manycore GPUs
Michael Garland, NVIDIA

Streaming video is available (Microsoft .asx)
4:30 p.m.
Adjourn
5:00 p.m.
NCSA only — Reception/Rapid-fire session
The rapid-fire session is designed to help attendees connect with others with similar domain expertise. Each attendee will have up to 2 minutes to share their name, institutions, and a brief description of their research.
Wednesday, August 12, 2009
7:30 a.m.
Continental Breakfast
8:00 a.m.
CUDA Performance Considerations
Wen-Mei Hwu, University of Illinois, and David Kirk, NVIDIA

Streaming video is available (Microsoft .asx)
9:15 a.m.
Break
9:30 a.m.
Floating-Point Precision and More Performance Considerations
Wen-Mei Hwu, University of Illinois, and David Kirk, NVIDIA

Streaming video is available (Microsoft .asx)
11:00 a.m.
Lunch
Noon
Hands-on Lab
    -- Understanding the labs
    -- CUDA lab manual
    -- CUDA lab code
    -- CUDA lab matrix data
    -- CUDA lab solutions
Nady Obeid, Xiao-Long Wu and I-Jui "Ray" Sung, University of Illinois

Streaming video is available (Microsoft .asx)
1:30 p.m.
NCSA only — Tour (optional)
This optional one-hour tour will feature the GPU cluster students are using during the summer school and a trip through the NCSA's machine room. Space is limited, so please sign up for a tour at the check-in desk.

At all other sites, and for those at NCSA not taking the tour, hands-on lab will continue.
2:45 p.m.
Break
3:00 p.m.
CUDA Tools, Libraries, and Software Resources
    -- CUDA Textual Profiler & Performance Tuning
    -- A Sampling ofCUDA Libraries
Teaching Staff

Streaming video is available (Microsoft .asx)
4:30 p.m.
Adjourn
Thursday, August 13, 2009
7:30 a.m.
Continental Breakfast
8:00 a.m.
Case Study - Accelerating Brain Functional Studies
Wen-Mei Hwu, University of Illinois

Streaming video is available (Microsoft .asx)
9:20 a.m.
Break
9:30 a.m.
Case Study - Accelerating Molecular Dynamics Experimentation
John Stone, University of Illinois

Streaming video is available (Microsoft .asx)
11:00 a.m.
Lunch
Noon
Hands-on Lab
Nady Obeid, Xiao-Long Wu and I-Jui "Ray" Sung, University of Illinois
1:30 p.m.
Case Study - Computational Fluid Dynamics (CFD) using Graphics Processing Units
Aaron Shinn, University of Illinois

Streaming video is available (Microsoft .asx)
2:45 p.m.
Break
3:00 p.m.
Keynote: Unlocking Biologically-Inspired Computer Vision: a High-Throughput Approach
David Cox, Harvard University

Streaming video is available (Microsoft .asx)
4:30 p.m.
Adjourn
5:00 p.m.
NCSA only — Pizza Reception
301 Coordinated Science Lab
Friday, August 14, 2009
7:30 a.m.
Continental Breakfast
8:00 a.m.
Building Many-Core Communities
    -- Future Directions and Conclusions
    -- Using the Internet to Build Community
Wen-Mei Hwu and Andy Schuh, University of Illinois

Streaming video is available (Microsoft .asx)
9:00 a.m.
Sharing of Plans for Using GPUs in Research
Attendees

Streaming video is available (Microsoft .asx)
10:30 a.m.
Break
10:45 a.m.
Participant Feedback - online survey
11:15 a.m.
Boxed lunches to stay or go
Noon
Adjourn

Video presentations, lessons, and other resources from the first summer school offered by the Virtual School of Computational Science and Engineering, which focused on GPU and multicore processors, are available online.

The Great Lakes Consortium for Petascale Computation (GLCPC) facilitates the widespread and effective use of petascale computing to address frontier research questions in science, technology, engineering, and mathematics at research, educational, and industrial organizations across the United States. The National Center for Supercomputing Applications, on behalf of the University of Illinois at Urbana-Champaign, is the lead institution for organizing and managing the GLCPC. ©2022 Board of Trustees University of Illinois. All rights reserved. Web privacy notice.