ViBram Silicon Hub is a research wing of National Institute of Technology, Raipur (NITRR), Government of India and VividSparks. We are the world's first POSIT™ based General Purpose (GP) Graphics Processing Unit (GPU) R&D group. The team at ViBram Silicon Hub is a mix of diverse and experienced talent. This diversity is what makes ViBram Silicon Hub a unique resource. The engineers and scientists bring to the research efforts the experience of handling complex and large projects and programs for the government and industry. ViBram Silicon Hub is excited by the journey from creating an idea, forming a working concept, building a prototype, and solving the problem. Partnering in this journey are industries and research organizations.
Dr. Sonal Yadav is the Director of ViBram Silicon Hub. She has almost ten years of teaching and research experience. She has been Assistant Professor in the Department of Computer Science and Engineering at the National Institute of Technology, Raipur, since 2021. Her research interests include networks-on-chip, computer architecture, computer vision, and AI/ML.
She completed her two-year postdoctoral from the Department of Computer Science and Engineering at the Indian Institute of Technology (IIT) Guwahati in March 2020. She obtained her PhD and MTech degrees from the Malaviya National Institute of Technology (MNIT), Jaipur, India. She received her BE (honours) in Computer Science and Engineering from the Government Engineering College Bikaner. She investigated power-gating methods for improving static power efficiency during her postdoc experience. Prior this, she explored static power and performance efficiency for manycore network-on-chip architectures during her PhD. Her thesis work was awarded travel grant in DATE 2019 and VLSI-SoC 2020 conferences. In 2018, she got the Best Paper Award by TCVLSI at IEEE-iSES. She has published more than 20 articles in prestigious international conferences and journals. She wrote a book on Computer System Organization published by AICTE, New Delhi, for Indian and international students.
Dr. Vijay Holimath, Deputy Director of ViBram Silicon Hub, a prominent figure in the world of computing. He is also CEO of VivdSparks, Dr. Holimath's expertise and contributions to the industry are truly exceptional. VividSparks operating at the forefront of innovation in AI/ML, Automotive, Graphics Computing, HPC, Edge Computing, Cloud Computing, and IoT. Throughout his career, Dr. Holimath has been involved in key projects with companies like Intel AMD, NXP, C-DAC Government of India, etc. His initiatives aimed at solving critical computational and memory bottleneck challenges faced by his clients. His ability to provide innovative solutions has earned him a stellar reputation in the field.
One of his most remarkable accomplishments has been raising funds, publishing papers, establishing offices across multiple countries, and cultivating a substantial base of satisfied customers. His visionary leadership has steered VividSparks IT Solutions toward continuous growth and excellence.
Dr. Holimath's journey into this dynamic industry began with a solid educational foundation, including a Bachelor of Engineering, Master of Philosophy, and a PhD in Electronics and Computer Science. His academic prowess, coupled with hands-on skills in chip design and application development, has been instrumental in his career's success.
Looking ahead, Dr. Holimath envisions the industry heading toward cloud computing and edge computing in the next 5-10 years. Hardware for computing tasks may increasingly reside in the cloud, transforming the way we use mobile devices and computers for everyday activities.
We are working on VividSparks RacEr GPGPU project. The RacEr GPGPU has 512 CPU cores in 2D mesh network, it has 64GB RAM expandable to 512GB RAM. It supports most programming stacks such as CUDA, Python, PyTorch, Tensor Flow, etc. The RacEr can be programmable to any applications, using VividProfiler one can pinpoint hotspots in applications as well as in hardware fabric and meet application power performance requirements. VividKompler supports native POSIT computation which supports C, C++ and Fortran. Visit VividSparks