The GPU Center of Excellence program recognizes, rewards, and fosters collaboration with institutions at the forefront of massively parallel computing research. GPU COEs have demonstrated a unique vision for improving the technology and application of parallel computing. Through the GPU COEs NVIDIA intends to:

  • Train the next generation of computer scientists and computational researchers to leverage the power of parallel computing to accelerate research and discoveries.
  • Enable and empower academic researchers to do world-changing research and discovery by dramatically increasing access to NVIDIA parallel computing platfrom and resident parallel computing expertise.
  • Establish research, educational, and recruiting relationships with the foremost academic institutions in the world.

The GPU COE program is an invitation-only program. Institutions whose primary goals include research, education and outreach as well as those having a current collaboration with NVIDIA may be invited to submit a proposal for consideration.

Invited institutions must satisfy three minimal requirements across multiple disciplines:

  • Teach: Offer multiple courses on a recurring basis that introduce students to GPU Computing and CUDA C/C++.
  • Publish: Demonstrate promising research results and generate academic papers and publication involving GPU Computing in at least 3 separate research laboratories or projects.
  • Outreach: Proactively outreach support  and influence students, researchers, thought leaders in academia, government and industry to learn and adopt GPU computing, the introduction of CUDA within their region and influence.

Each GPU COE award is tailored specifically for the institution and based on a variety of factors.  The duration is intended for up to three years, subject to approval of annual reports. Renewal each year is at the sole discretion of NVIDIA. GPU COEs are renewable beyond the three-year duration upon invitation.

The GPU COE program is an exclusive and limited program. Most research institutions should consider applying to the GPU Research Center program. University faculty who are not yet teaching or plan to teach CUDA C/C++ should consider applying to the GPU Education Center program.

See our current list of  WorldWide GPU Centers of Excellence.

For questions about the GCOE Program contact us at GCOEproposal@nvidia.com.

 

GPU COE Achievement Awards

GPU Centers of Excellence include some of the world's top universities engaged in cutting-edge work with CUDA and GPU computing. To highlight and reward the excellent research taking place, each of the world's COEs are asked to submit an abstract describing their top achievement in GPU computing over the past year. A panel of NVIDIA experts select four COEs to present their achievements at a special event during GTC. COE peers at the event then vote for their favorite presentation and research. The results of the past GPU COE Achievement Awards are below: