Monday October 27, 2014
Stamp Student Union, Atrium

8 – 8:30 a.m. Registration, Breakfast
8:30 – 10 a.m. Keynote Session

8:30 a.m.

Fran LoPresti
Welcome Address: GPUs at Maryland

8:40 a.m.

David Luebke
Computational Displays: How GPU Horsepower and Novel Optics to Enable Thin, Light, Wide-angle Virtual and Augmented Reality

9 a.m.

Amitabh Varshney
Augmented Reality Made More Real

9:20 a.m.

GPU Panel for Medicine
Chair: Terry Yoo
Panelists: Peter Bajcsy, Raj Shekhar and Oleg Kuybeda
10 – 10:30 a.m. Coffee Break
10:30 a.m. – 12 p.m. Research Overview Session I

10:30 a.m.

Alex Szalay
Data Intensive Science Using GPUs

10:50 a.m.

Michael Cummings
GPU Computing and the Tree of Life

11:10 a.m.

Jeff Hollingsworth
Automatically Tuning Performance and Power for GPUs

11:30 a.m.

Satyandra K. Gupta
GPU-Enabled Computing in Robotics and Advanced Manufacturing Applications

11:50 a.m.

Joseph JaJa
Mapping Biomedical Applications onto GPU Platforms
12:10 – 1:30 p.m. Lunch
1:30 – 3 p.m. Research Overview Session II

1:30 p.m.

Norman Wereley
Particle Simulations in Magnetorheological Flows

1:50 p.m.

Jeffery Klauda
Molecular Modeling of Biomolecules: How Can GPUs Advance Research?

2:10 p.m.

Shuvra Bhattacharyya
Vectorization and Mapping of Software Defined Radio Applications On GPU Platforms

2:30 p.m.

Lorena Barba
PyGBe for probing protein orientation near charged surfaces
3 – 3:30 p.m. Coffee Break
3:30 – 5 p.m. Research Overview Session III

3:30 p.m.

Ramani Duraiswami
Fast Heterogeneous Computing

3:50 p.m.

Laura Monroe
Probabilistic Computing on the GPU

4:10 p.m.

GPU Panel for High-Throughput Computing
Chair: Jimmy Lin
Panelists: Raju Namburu, George Stantchev, and R. Jacob Vogelstein

Tuesday October 28, 2014
Stamp Student Union, Prince George's Room

8 a.m. Intro to GPU Computing: High-level discussion of GPU computing.
Slides: CUDA Background, Driver/Toolkit/Samples
Code: Workshop Code
9:15 a.m. Programming with OpenACC: Using simple directives to accelerate code
11:30 a.m. Basics of CUDA Programming, Part 1: CUDA syntax, memory allocation, launching simple kernels
12:30 - 1:30 p.m. Lunch
1:30 p.m. Basics of CUDA Programming, Part 2: CUDA syntax, memory allocation, launching simple kernels
2:30 p.m. Fundamental GPU performance optimizations (3 hrs.): Using the performance profiler; global and shared memory optimizations

Wednesday October 29, 2014
Stamp Student Union, Prince George's Room

9:00 a.m. Intermediate Optimizations (3 hrs.): Overlapping communication with computation; streams and concurrency