Accelerated by 200 Tesla K80 GPUs, the Laconia Supercomputer was recently unveiled at the Institute for Cyber-Enabled Research at Michigan State University.
Named after a region in Greece that was home to the original Spartans, the mascot of Michigan State University, the supecomputer ranks in the TOP500 fastest computers in the world and is projected to be in the top 6% most energy efficient supercomputers.
The new supercomputer will be used by university researchers for a variety of projects, from running molecular simulations for new drug designs to developing neural networks for large-scale facial recognition.
University Distinguished Professor Anil Jain, along with his Ph.D. student Charles Otto, are able to train their convolutional neural network with 10 convolutional layers and millions of parameters in just two days on the new supercomputer, resulting in an effective solution for challenging face recognition applications.
Read more >>
Meet the New Tesla-Accelerated Supercomputer at Michigan State University
Jul 14, 2016
Discuss (0)
Related resources
- GTC session: AI Supercomputing: Pioneering the Future of Computational Science
- GTC session: Fueling the Future: How GM Motorsports Accelerates High Speed Racing with AI Physics
- GTC session: Empower Large-Scale AI Workloads With Google Cloud AI Hypercomputer Supercomputing Architecture (Presented by Google Cloud)
- NGC Containers: MATLAB
- SDK: AMGX
- Webinar: Want to drive innovation and speed up scientific workloads? It starts with the network.