通讯
SAN JOSE, Calif., April 9, 2019 – Dr. Zhimin Ding, Principal AI Processor Architect at Tachyum Inc. will be a featured presenter at the AI Accelerator Summit, April 11-12 at Silicon Valley’s San Jose Convention Center.
Ding is a pioneer in artificial intelligence, machine learning, and neural networks – and how these technologies intersect with semiconductors and system design. Since the early 1980s he has specialized in areas like pattern recognition, supervised and unsupervised learning, training, sensory fusion, and neuromorphic architectures. He holds several degrees in electrical engineering, computer science, and neuroscience, and holds or has developed 11 patents.
“In many ways AI has matured and evolved, yet in others we are still in the early phases of what can be accomplished with more efficient processors,” said Ding. “I will offer a different perspective on AI acceleration in favor of heterogeneous, universal, general-purpose processors, and how they will enable higher performance, higher densities, and higher intelligence.”
The AI Accelerator Summit attracts an elite group of leaders in AI hardware and architecture from the world’s largest organizations and most exciting AI chip startups in Silicon Valley. In 2019 it debuts a series of worldwide events, the AI Hardware World tour, in cities including London, Singapore, Shanghai, New York, Toronto, and Sydney.
The Tachyum ultra-low power Prodigy processor, will allow system integrators to build a 32 tensor Exaflops AI supercomputer This will enable users to simulate, in real-time, human brain- sized neural nets beginning in 2020, instead of 2028.
Prodigy is the smallest and fastest general purpose, 64-core processor developed to date, requiring 10x less processor power and reducing processor cost by 3x. Prodigy reduces data center TCO (annual total cost of ownership) by 4x, through a disruptive hardware architecture and a smart compiler that has made many parts of the hardware found in a typical processor redundant. Fewer wires and shorter wires, due to a smaller, simpler core, translates into much greater speed and power efficiency for the processor.
