A new race – making computation sustainable | Shivam Kajale | TEDxBoston

Computing infrastructure of the world is the fastest growing consumer of primary electricity, and on track to being a major contributor of greenhouse gases. Sustaining current growth rate of compute, fueled by artificial intelligence, blockchain and internet of things, might become practically impossible over the next two decades. Thus, we need to make quick advances towards making our computing hardware exceedingly energy-efficient, to make computation both environmentally and practically sustainable. Herein, I describe how we are developing “beyond CMOS” devices using two-dimensional magnetic materials to enable massive reductions in energy dissipation in computer hardware.

AI, Biomimicry, Brain, Climate Change, Computers, Electricity, Energy, Global issues, Machine Learning, Materials, Physics, Research, Sustainability, Technology Shivam is a PhD candidate and a research assistant at the MIT Media Labs. His research focuses on realizing the goal of highly energy-efficient “beyond CMOS” devices, based on magnetics, spintronics and magnetoelectrics.

Data centers across the world are consuming electricity at par with the electricity consumption of entire countries, like South Africa and the United Kingdom. The projected growth of energy consumption in computational hardware over the next few decades has serious global implications. Realizing this urgency, Shivam and his group of researchers are trying to build “beyond CMOS” devices, to replace the billions of silicon-based transistors in our computers with brain-inspired, atomically thin magnetic devices, which are projected to enable thousands of times more energy efficient computation for the sustainable growth of the IT infrastructure.

Shivam is an electrical engineer by training, with a B.Tech and M.Tech in Electrical Engineering from Indian Institute of Technology Bombay. He uses techniques from applied physics, material science and nanotechnology for building scalable and energy-efficient devices for an environmentally sustainable AI ecosystem growth. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at https://www.ted.com/tedx