SAN JOSE, Calif.—Nvidia refreshed its lineup of GPUs for deep studying and synthetic intelligence purposes on Wednesday with the brand new 5,120-core, 7.5 teraflop Tesla V100 Volta.
The brand new processor is a part of Nvidia’s quest to give you a brand new technique to constantly enhance computing efficiency within the aftermath of Moore’s Legislation, which many trade leaders agree is just about lifeless. As a substitute of boosting processor speeds or cramming extra transistors onto already-crowded silicon, Nvidia is championing GPU-accelerated computing, which the corporate’s CEO Jensen Huang (pictured above) mentioned can provide a 150 p.c efficiency enhance yearly.
The Tesla V100, with a brand-new structure referred to as “Volta,” represents that newest enhance. It is the “subsequent large leap into the brand new world” of AI and high-performance computing, Huang mentioned at Nvidia’s builders convention right here. The V100 will begin delivery by the tip of the 12 months to knowledge facilities owned by Amazon, Microsoft, and different cloud computing providers in a number of completely different configurations.
The most typical off-the-shelf model of the V100 is a $149,000 supercomputer referred to as the DGX-1, which accommodates eight V100 processors. There’s additionally a $69,000, liquid-cooled mini supercomputer referred to as the DGX aimed toward researchers who aren’t utilizing the cloud, which is powered by 4 V100s.
The DGX-1 made its debut final 12 months with the V100’s predecessor, the Tesla P100. Because of the Volta structure, the V100-powered supercomputers provide 5 occasions extra peak teraflops enchancment over Pascal chips, the current-generation Nvidia GPU structure, and 15 occasions greater than the Maxwell structure affords.
The V100 will energy servers within the Microsoft Azure cloud and Amazon Web Services, that are the 2 titans of cloud choices for deep studying. As soon as put in, they will help software program like Microsoft’s Cognitive Toolkit and Mxnet, an open-source deep studying framework that’s Amazon’s most well-liked AI framework. Nvidia’s V100 can even possible energy Fb’s Caffe2 framework.
“Our technique is to create the best platform for deep studying,” Jensen mentioned. He provided a number of demonstrations of the P100’s potential purposes, from an art work algorithm that may merge the colour palette of 1 photograph (an orange sundown, for example) with the topic of one other (a seashore and clouds, for example) into a wholly new picture.
After all, the chief good thing about internet hosting AI processing energy within the cloud is that engineers can do just about no matter they need with it. To that finish, Nvidia additionally plans to launch its personal GPU Cloud service, presently in an invite-only beta, which permits anybody to add their machine algorithms to Volta servers.
“Top-of-the-line methods to take pleasure in deep studying is for another person to construct this extremely sophisticated supercomputer in your behalf,” Jensen mentioned.