Researchers from the University of Illinois Win Second Annual Award from NVIDIA’s CUDA Centers of Excellence

by Chandra Cheij

Researchers from the University of Illinois at Urbana-Champaign snagged the Second Annual Achievement Award for CUDA Centers of Excellence (CCOE), for their research with Blue Waters.

The team was among three other groups of researchers from CCOE institutions, which include some of the world’s top universities, engaged in cutting-edge work with CUDA and GPU computing.

Each of the world’s 21 CCOEs was asked to submit an abstract describing their top achievement in GPU computing over the past year.  A panel of experts, led by NVIDIA Chief Scientist Bill Dally, selected four CCOEs to present their achievements at a special event during our annual GPU Technology Conference  this week in San Jose. CCOE peers voted for their favorite, who won bragging rights recipient of the second CUDA Achievement Award 2013.

The four finalists — each of whom received a GeForce GTX Titan – built around the same Kepler chip that powers the world’s fastest supercomputer, the Titan system at the Oak Ridge National Laboratory.  The University of Illinois not only won bragging rights and a Titan but also got to take home a  Microsoft Surface tablet and special NVIDIA green keyboard.

The winning team
Researchers from the University of Illinois at Urbana-Champaign snagged the Second Annual Achievement Awards for CUDA Centers of Excellence (CCOE).

The four finalists who presented were:

Barcelona Supercomputing Center/Universitat Politecnica Catalunya – Supercomputing for the “Small” Masses

The Barcelona Supercomputing Center/Universitat Politecnica Catalunya CCOE defined and coordinated the design of the first hybrid ARM + CUDA GPU platform, where SECO provided the integrated board design and product manufacturing and validation, and NVIDIA contributed the silicon technology and software stack support for the ARM platform. The research project has also been part of the PRACE initiative to explore a set of prototypes to test and evaluate promising new technologies. The result of this project was a commercial product called the CARMA Development Kit, released in summer 2012.

Lomonosov Moscow State University – Developing GPU HPC Infrastructure at MSU and Beyond

CUDA Center of Excellence of Moscow State University presented a project “Developing GPU HPC Infrastructure at Moscow State University and Beyond.” Hardware and software, supercomputing applications and super-scalable applications, education, popularization and many other issues are necessary components of this work. As a result of the project, there are dozens of efficient supercomputing applications in the MSU HPC center, 35 Russian universities are involved in GPU educations, and hundreds research groups use GPU technology in their practice.

Technische Universität Dresden, Revolutionizing Microscopy – One Image at a Time

Automated computer-aided workflows have changed how biologists can understand the inner workings of a cell. Experiments with thousands of parallel samples, all just slightly different, can now be carried out, analyzed and influenced in almost real time. The compute power of GPUs has further enhanced this method to open up new dimensions and even increase image resolution beyond the limits of the microscope’s optics. The presentation is an example how the ground-breaking methods, developed by the team of Yannis Kalaidzidis can be used to understand the inner workings of liver cells – and better understand the causes of cancer, and how it can prevented. The project used special GPU-enabled motion tracking software to make the transition from 2D to 3D microscopy.

University of Illinois at Urbana-Champaign – GPU Computing in Blue Waters

In 2011, when the University of Illinois at Urbana-Champaign (UIUC) CCOE proposed the inclusion of Kepler GPUs in Blue Waters skeptics expressed concern about a lack of applications for GPU accelerated supercomputers and perceptions that GPUs are hard to program. To address these concerns, a joint teamof UIUC CCOE researchers, NVIDIA engineers, Cray engineers, staff from the National Center for Supercomputing Applications and others documented huge gains in whole-application performance when using GPU acceleration. The UIUC CCOE also established a training program that resulted in more than 2,600 students getting Certificate of Achievement (with a grade of 70% or higher); and 2,281 of them received Certificate of Distinction (with a grade of 85% or higher).