Amrita University named NVIDIA CUDA Research Center
Amrita University will now be able to access new technology after being named a CUDA Research Center by NVIDIA. With this Amrita will join Imperial College London, Purdue University, IIT Rourkee and others as a NVIDIA designated research center.
With several studies on GPGPU computing, a form of high performance computing, Amrita School of Biotechnology is now called aCUDA Research Center as declared by NVIDIA, a world leader in high performance computing and the makers of CUDA. CUDA, which stands for Compute Unified Device Architecture, is a parallel computing platform and programming model created by NVIDIA and implemented using graphics processing units (GPUs).
Amrita School of Biotechnology has also been nominated as CUDA Teaching Center for a period of 1 year starting March 13, 2015. As part of the grants, the School will receive CUDA computing devices as equipment and a TA support grant of US $2500. Students of various programmes, specifically M.Sc. Bioinformatics have been using CUDA as a technology to improve computing in order to model and solve problems in Neuroscience, Computer-aided Drug Design and bioinformatics.
Dr. Shyam Diwakar is the principal investigator of this research project and has been working on modelling brain tissue function on these high-performance computing devices. For Amrita’s computational neuroscience group, this recognition will help design prediction models for a serious cerebellar disease, called ataxia. The open challenge is to understand the complex computation that takes place in our brains in order to predict behaviour, understand function and dysfunction and to look at abstracting the understanding to solve problems in engineering and robotics.
The focus here is a part of the brain, the cerebellum, which is of critical importance for sensory-motor control and learning and its disruption causes a dramatic neurological syndrome called ataxia.The group at Amrita School of Biotechnology modelled a sample of firing neurons in spatial structure, as shown in the animated figure, as a realistic simulation of how rat cerebellum granule neurons respond to certain inputs, the red color indicating that the neurons (coloured dots or small circles) are firing or producing electrical signals called action potentials. Such studies provide valuable information to understand disease conditions.
The GPGPU study is also implemented by Mrs. Manjusha Nair, a faculty member of Computer Science and Engineering at Amrita School of Engineering, Amritapuri who is also a part-time Ph.D. student at the Computational neuroscience lab. The School of Biotechnology had already setup a 7-teraflop GPGPU supercomputer “MANAS” and had several publications from simulations run on the machine. The future role of these awards will surely help accelerate life science–related computational modelling research. The recognition will also be reported soon via NVIDIA academic page.