GPU Education Center at DA-IICT

Since April 2014, DA-IICT has been named a CUDA Teaching Center/ GPU Education Center by nVIDIA corporation, the world's leading developer of Graphics Processing Units (GPUs) and one of the world leader in visual and High Performance Computing (HPC).

CUDA is NVIDIA's parallel computing architecture that enables phenomenal increase in computing performance for processing-intensive applications by exploiting the power of the GPUs. The CUDA Teaching Center (CTC) program recognizes institutions that have integrated GPU computing techniques into their mainstream curriculum.

More than 500 universities and institutions around the world are now teaching parallel programming based on the nVIDIA CUDA parallel programming environment. As a CUDA Teaching Center, DA-IICT has the opportunity to keep pace with world renowned universities that have similar GPU computing educational resources. At DA-IICT, in combination with existing curriculum, it will cultivate student's theoretical and practical skills in HPC domain and also their competitiveness in the high-tech job market.

At DA-IICT, the CUDA Teaching Center program is tied with the "Introduction to GPU Programming" course (IT477) being offered during autumn semester. This is a technical elective course available to undergraduate as well as postgraduate students. It is being taught by Dr. Bhaskar Chaudhury, faculty of computational Science, and the principal investigator behind the GPU Education Center initiative at DA-IICT.

The above lab oriented course exposes students to design and write GPGPU based codes. A student who successfully completes this course will

  • Understand important concepts behind parallel computing.

  • Understand GPU architecture/hardware and GPGPU development frame-work (CUDA).

  • Develop the ability to solve computationally challenging and demanding problems using GPGPU programming.

  • Develop expertise in CUDA programming.

As part of the CTC award and as a support to the University's teaching efforts, DA-IICT received high-end graphics cards (Tesla K40 and GTX cards) from nVIDIA which were installed on four latest high-end workstations. This computing facility is utilized by students working on different areas of GPU and parallel programming in the broad domain of Computational Science and ICT.

From Autumn semester 2015, a new core course for Computational Science students has been introduced which focuses on latest High Performance Computing techniques including GPU programming. In this regard a new Computational Science and HPC laboratory has been set up at DA-IICT with 30 latest workstations with GPU cards in each of these machines which will be used by the students taking HPC courses. Several latest edition books on parallel programming and CUDA have been also included in the DA-IICT library. As part of the award nVIDIA also provided some books on GPU Programming.

Presently GPU based R&D activities at DA-IICT is primarily focused in the areas of Computational science, Computational Physics, Computational Electromagnetics, Computational Mathematics etc. Students are also exploring some new areas such as image processing for their projects.

For more information, contact the Principle Investigator (GPU Education center at DA-IICT):
Dr. Bhaskar Chaudhury
Faculty (Computational Science)
2205, Faculty Block - II, DA-IICT
Gandhinagar, Gujarat, INDIA 382 007
Phone: (+91) 79-30510-590
Email: bhaskar_chaudhury@daiict.ac.in