Need smarter insights in your inbox? Join our weekly newsletters to get solely what issues to enterprise AI, knowledge, and safety leaders. Subscribe Now
As AI functions more and more permeate enterprise operations, from enhancing affected person care by way of superior medical imaging to powering advanced fraud detection fashions and even aiding wildlife conservation, a crucial bottleneck usually emerges: knowledge storage.
Throughout VentureBeat’s Rework 2025, Greg Matson, head of merchandise and advertising and marketing, Solidigm and Roger Cummings, CEO of PEAK:AIO spoke with Michael Stewart, managing accomplice at M12 about how improvements in storage expertise allows enterprise AI use instances in healthcare.
The MONAI framework is a breakthrough in medical imaging, constructing it sooner, extra safely, and extra securely. Advances in storage expertise is what allows researchers to construct on prime of this framework, iterate and innovate shortly. PEAK:AIO partnered with Solidgm to combine power-efficient, performant, and high-capacity storage which enabled MONAI to retailer greater than two million full-body CT scans on a single node inside their IT atmosphere.
“As enterprise AI infrastructure evolves quickly, storage {hardware} more and more must be tailor-made to particular use instances, relying on the place they’re within the AI knowledge pipeline,” Matson stated. “The kind of use case we talked about with MONAI, an edge-use case, in addition to the feeding of a coaching cluster, are properly served by very high-capacity solid-state storage options, however the precise inference and mannequin coaching want one thing completely different. That’s a really high-performance, very excessive I/O-per-second requirement from the SSD. For us, RAG is bifurcating the kinds of merchandise that we make and the kinds of integrations we now have to make with the software program.”
Bettering AI inference on the edge
For peak efficiency on the edge, it’s crucial to scale storage right down to a single node, as a way to convey inference nearer to the information. And what’s key’s eradicating reminiscence bottlenecks. That may be performed by making reminiscence part of the AI infrastructure, as a way to scale it together with knowledge and metadata. The proximity of knowledge to compute dramatically will increase the time to perception.
“You see all the large deployments, the large inexperienced discipline knowledge facilities for AI, utilizing very particular {hardware} designs to have the ability to convey the information as shut as doable to the GPUs,” Matson stated. “They’ve been constructing out their knowledge facilities with very high-capacity solid-state storage, to convey petabyte-level storage, very accessible at very excessive speeds, to the GPUs. Now, that very same expertise is occurring in a microcosm on the edge and within the enterprise.”
It’s turning into crucial to purchasers of AI methods to make sure you’re getting essentially the most efficiency out of your system by operating it on all strong state. That lets you convey large quantities of knowledge, and allows unbelievable processing energy in a small system on the edge.
The way forward for AI {hardware}
“It’s crucial that we offer options which can be open, scalable, and at reminiscence pace, utilizing a number of the newest and biggest expertise on the market to do this,” Cummings stated. “That’s our objective as an organization, to offer that openness, that pace, and the size that organizations want. I feel you’re going to see the economies match that as properly.”
For the general coaching and inference knowledge pipeline, and inside inference itself, {hardware} wants will hold rising, whether or not it’s a really high-speed SSD or a really high-capacity answer that’s energy environment friendly.
“I’d say it’s going to maneuver even additional towards very high-capacity, whether or not it’s a one-petabyte SSD out a few years from now that runs at very low energy and that may mainly exchange 4 occasions as many exhausting drives, or a really high-performance product that’s virtually close to reminiscence speeds,” Matson stated. “You’ll see that the large GPU distributors are taking a look at find out how to outline the following storage structure, in order that it could possibly assist increase, very carefully, the HBM within the system. What was a general-purpose SSD in cloud computing is now bifurcating into capability and efficiency. We’ll hold doing that additional out in each instructions over the following 5 or 10 years.”
