The processing demands of artificial-intelligence and deep-learning applications are sparking a renaissance in hardware development. Old commodity stuff isn’t cutting it, even with spanking-new software on top. Chip makers are now called upon to delver rip-sawing, purpose-made graphics processing units, and memory chips, etc., to provide the extra oomph.

Applications like image recognition for medical research are among those that require an architecture that fires on all cylinders, according to Tom Eby (pictured, right), senior vice president and general manager of the Compute & Networking business unit at Micron Technology Inc.

“To support those in an optimized way really does require the mating of the optimal processing architecture — things like GPUs with the right high-bandwidth, low-latency memory and storage solutions,” Eby said.

Micron is among the lucky group of companies to score a dance with chip-maker Nvidia Corp. They’ve partnered on end-to-end architectures that cover the whole life of AI apps from model-training to inferencing at the edge.

Eby and Premal Savla (pictured, left), director of product management, deep learning systems, at Nvidia, spoke with Dave Vellante (@dvellante) and David Floyer (@dfloyer), co-hosts of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, during the Micron Insight event in San Francisco. They discussed the two companies’ deep engineering partnership, as well as how AI is breathing new life into hardware. (* Disclosure below.)

Chipping away at AI apps

GPUs are still the foundation of Nvidia’s business, but it’s now building out whole end-to-end architectures, software and AI training libraries. “We have something called TensorRT, which is basically Tensor real-time that gives the capability to get these answers very quickly,” Savla said.

The Nvidia Tesla T4 GPU is allegedly the world’s most advanced inference accelerator and is ideal for aforementioned image recognition, according to Savla.

Nvidia’s recently-released GeForce RTX graphics cards leverage Micron’s GDDR5X  and GDDR6 high-speed memory chips.

These chips are showing up in a range of use cases inside and outside of data centers, Eby pointed out. “We’re seeing technologies like GDDR6 being applied to a much broader range of applications — like automotive, like networking, like edge AI — to provide the performance to get that real-time response, but in a form factor and at a cost point that’s affordable to the application.”

Micron has announced a $100 million investment fund targeting AI startups.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of Micron Insight 2018. (* Disclosure: TheCUBE is a paid media partner for the Micron Insight event. Neither Micron Technology Inc., the event sponsor, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE