Nervana NNPs: Intel unveils its high-performance AI chip for training and inference

Intel Nervana NNPs comprises of two different products: Intel Nervana NNP-T or Spring Crest for deep learning training and Intel Nervana NNP-I or Spring Hill for high-performing deep learning inference.

Chipmaker Intel unveiled on August 20 its upcoming high-performance artificial intelligence (AI) accelerators for deep learning training and industry-leading performance. The in-depth presentations about the capabilities and architecture of Intel Nervana Neural Network Processors (NNPs) were given at Hot Chips 2019, one of the semiconductor industry's leading conferences on high-performance microprocessors and related integrated circuits.

Intel Nervana NNPs comprises of two different products: Intel Nervana NNP-T or Spring Crest for deep learning training and Intel Nervana NNP-I or Spring Hill for high-performing deep learning inference. Both the products have been designed with a focus on AI and the future in mind. Dedicated accelerators like the Intel Nervana NNPs are designed with a focus on AI to provide customers the right intelligence at the right time, Intel said in a press release.

"To get to a future state of 'AI everywhere,' we'll need to address the crush of data being generated and ensure enterprises are empowered to make efficient use of their data, processing it where it's collected when it makes sense and making smarter use of their upstream resources," said Naveen Rao, Intel vice president and general manager, Artificial Intelligence Products Group.

Image Credit: Twitter / @intelnews

Data centers and the cloud need to have access to performant and scalable general-purpose computing and specialized acceleration for complex AI applications. In this future vision of AI everywhere, a holistic approach is needed—from hardware to software to applications," Rao further added.

Intel Nervana NNP-T is a new class of efficient deep learning system hardware specifically designed for training a network as fast as possible within a given power budget while maintaining a balance between computing, communication, and memory. NNP-T has been designed in a way such that it can be optimized for a wide variety of current and future workloads.

Intel Nervana NNP-I is purpose-built specifically to offer industry-leading performance and power efficiency for all major data center inference workloads and a high degree of programmability without compromising the performance or power efficiency. NNP-I solution will allow companies to harness the full potential of their data as actionable insights, Intel said.

We're sharing details on our Nervana NNP-I accelerator, designed to accelerate deep learning deployment at scale. https://t.co/tHEJsIXcDr #HotChips31 pic.twitter.com/CB2YPryK3n

— Intel (@intel) August 20, 2019