General Manager, Tesla GPU Accelerated Computing, NVIDIA
Sumit Gupta is responsible for the overall development and marketing of the NVIDIA Tesla GPU accelerator products for the high-performance computing (HPC) and technical computing industries. Following his stellar career path at NVIDIA, Sumit’s keen intellect, drive, and energy has taken him from what was an interim position to the full-time GM of the Tesla business unit. How long can he keep Tesla up on top?
Gupta has been at the eye of storm of GPU computing from the beginning, when NVIDIA erupted onto the high performance computing scene in 2007. Since then, he’s had the opportunity to work with scientific and computing visionaries in universities and throughout the technology industry to developing and applying advanced GPU accelerator solutions to solve some of the world’s most complex computational problems. He is intensely focused on advancing the global awareness and use of GPU accelerated computing, which offers the potential to fundamentally accelerate the pace of innovation and discovery in science and industry.
One of his key strengths he possesses is his strategic perspective from his past experience living in four countries – Kuwait, India, United States, and Belgium – giving him a keen global perspective on the intersection of technology, industry and society. His professional background includes experience in range of industries, including roles as a system software engineer at IBM, a hardware chip design engineer at S3 Graphics, and a researcher at University of California at San Diego and Irvine, before making a dramatic change to serve an entrepreneur-in-residence at Tallwood Venture Capital. This broad and culturally diverse background served him well as he prepared for his career in marketing and business management at Tensilica, and ultimately led him to join NVIDIA in 2007.
Gupta has a Ph.D. in computer science and engineering from the University of California, Irvine, and a bachelor of technology in electrical engineering from the Indian Institute of Technology, Delhi. He has authored a book on parallelizing compilers for hardware design, holds a patent related to the same, and has contributed to more than 20 books and technical publications on embedded system design, FPGAs and reconfigurable computing.
When he can find the time, he enjoys multiple outdoor sports including rock climbing, tennis and hiking with his family. He also enjoys yoga, blogging, and reading business books to keep his life/balance order in place.
Sumit’s Top 4 HPC initiatives or technologies to watch in 2013:
- Accelerated computing: 2013 is the year of accelerated computing. Every HPC processor vendor (Intel, NVIDIA, AMD) has new accelerators for the market and every OEM has differentiated accelerated systems to offer.
- User-friendly software tools: Tools like OpenACC that move the burden of parallelizing from the computational scientists to the compiler/software tools.
- Affordable HPC: Wider access to inexpensive HPC systems, whether it is PCs that are supercharged with accelerators or small accelerated clusters that provide 100 teraflops for every researcher in the world.
- ARM: ARM-based servers will make their debut in the enterprise and HPC market.