GPU-accelerator technology to help in engineering, scientific research, and data analytics
BANGALORE, December 18: NVIDIA Corporation (NASDAQ:NVDA), the visual computing company, announced in a select media briefing today that it aims to drive the adoption of graphics processing unit (GPU)-accelerated computing technology in partnership with International Business Machines Corporation (NYSE:IBM). As part of its plans, NVIDIA will also launch products in the areas of engineering, scientific research, and enterprise-scale data analytics in India.
IBM and NVIDIA plan to collaborate on GPU-accelerated versions of IBM’s wide portfolio of enterprise software applications, enabling the adoption of GPU-accelerator technology for the first time in enterprise-scale data centers.
The collaboration aims to enable IBM customers to more rapidly process, secure, and analyze massive volumes of streaming data. The partnership also plans to integrate the processing capabilities of NVIDIA’s Tesla GPUs with IBM POWER8 central processing units (CPUs) in IBM Power supercomputers to deliver a new class of technology that maximizes performance and efficiency for scientific, engineering, big data analytics, and other high-performance computing (HPC) workloads.
This week, NVIDIA named the Indian Institute of Technology, Bombay (IIT-B) as India’s first CUDA Center of Excellence (CCoE), the highest honor that have earlier been awarded to the University of Harvard and the University of Stanford for their ground-breaking educational programs and research using NVIDIA GPU accelerators and the CUDA parallel programming environment.
“Compared with other major regions around the world, India is experiencing the fastest growth in new CUDA developers,” said Vishal Dhupar, Managing Director, South Asia, NVIDIA. “We’re boosting our efforts to support this growth, providing the most advance technologies and resources to bring the game-changing benefits of accelerated computing to academics, researchers, engineers, and enterprise IT managers throughout the country.”
According to the company officials, NVIDIA Tesla K40 GPU accelerator is the world’s “highest performance accelerator ever built”, delivering extreme performance to a widening range of scientific, engineering, HPC, and enterprise applications. It provides double the memory and up to 40% better performance than its predecessor, the Tesla K20X GPU accelerator, and 10 times better performance than today’s fastest CPU. The Tesla K40 GPU is also the world’s first and highest-performance accelerator optimized for big data analytics and large-scale scientific workloads.
Explaining how GPU accelerators will help reduce power consumption drastically and shorten the simulation time, Shankar Trivedi, Vice President Industry Business Development, NVIDIA, said, “Utilizing GPUs as companion processors has accelerated its application performance by 40 times compared to running them on CPUs alone and delivered over 80% savings, enabling greener data centers that deliver better performance for the same power input. Take the case of JPMorgan. Earlier, the bank used to take at least eight hours to do its intra-day risk calculations across a range of products, but now they can carry out the same task in just an hour.”
Asked whether GPU accelerators would help companies like Google Inc. (NASDAQ:GOOG) and Facebook Inc. (NASDAQ:FB) that run 180,000 and 217,000 servers, respectively, Trivedi said, “Yes, it will certainly help them reduce power consumption drastically.”
NVIDIA will also target software companies such as Infosys Ltd. (NSE:INFY) and Tata Consultancy Services Limited (NSE:TCS) for adopting GPU accelerators into their systems.