Microsoft and OpenAI collaborate on a monster of a supercomputer
In quick: Microsoft constructed a gargantuan supercomputer for artificial intelligence, which is the newest expansion of its Azure cloud computing platform. And by helping OpenAI bring brand-new refinements to AI models, it got a brand-new vote of confidence for Azure as one of the leading platforms for future development.
At its virtual Build 2020 designer conference, Microsoft exposed numerous intriguing preview into its latest efforts to unify Windows 10 app advancement, use better tools for power users, and integrate more Linux goodies by means of the Windows Subsystem for Linux.
A less talked about but still intriguing statement at the occasion is that Microsoft has actually partnered with OpenAI to make one of the leading 5 AI supercomputers out there. The monster is basically an Azure-linked supercomputer that’s developed particularly to evaluate OpenAI’s synthetic intelligence models, from the ones that compose believable phony news and beat you at Dota 2, to the more practical ones that can enhance robotics and decision-making.
Microsoft didn’t go unfathomable into the technical information of the brand-new supercomputer, but it did state that it consists of more than 285,000 CPU cores combined with 10,000 GPUs grouped into servers equipped with 400 gigabits per second of network bandwidth.
The primary reason the business is tossing so much calculate power into AI applications is that models are practically doubling in size every few months. For circumstances, Microsoft’s Turing design for natural language processing has swollen to 17 billion parameters from simply one billion it had one year back.
Last year, Microsoft entered a multi-year collaboration with OpenAI, putting $1 billion into the AI business began by Elon Musk to have it utilize Azure in training AIs. But more notably, the truth that OpenAI picked Azure as its cloud platform of option is yet another win for Microsoft on that front.