Pure Storage has unveiled a new data-centric architecture in a bid to enable businesses to put machine learning at the heart of their operations.
The US firm made a suite of announcements at its annual Pure/Accelerate summit in California this week, as part of plans to help customers’ build storage solutions around their data and applications.
“Our customers aggressively seek to use data to improve their customer experience and outdistance their competition,” said Pure’s CEO Charles Giancarlo. “[Our] data-centric architecture enhances and simplifies their ability to use data for intelligence and advantage.”
The company unveiled a new FlashArray//X product line equipped with NVMe to make databases, virtualised environments and test/dev initiatives faster. The lineup features five configurations and is priced at the same level as the lineup it replaces. A new “Evergreen Storage Service” allows business to use on-premises storage on a “cloud-like consumption model”.
Pure also unveiled AIRI Mini, a new AI-Ready Infrastructure solution for businesses that want to get machine learning projects up and running quickly. Like the original AIRI storage solution announced in March, the AIRI Mini features Nvidia servers and software.
“While AIRI simplifies AI infrastructure for initiatives at any scale, AIRI Mini offers a powerful and affordable entry point for organizations to explore and scale as they grow into AI,” said Pure’s GM of FlashBlade Matt Burr. “AIRI Mini removes the final barriers for any organization to leverage AI, and scales as AI workloads grow.”
A survey of 2,300 business leaders conducted by MIT on behalf of Pure revealed that 82 per cent think AI will have a positive impact, 83 per cent think AI is important for analytics and reducing human error and 79 per cent think the legal and ethical implications of the technology still need to be clarified.
Speaking to NS Tech, Pure’s EMEA CTO Alex McMullan said that while AI is still a long way off from mimicking human intelligence, the technology is more developed than many might assume.
“I think the biggest gap we still have in this space is how we teach the machines. Our biggest problem is that we don’t quite understand how we work yet. So translating that into machine-based teaching is our biggest gap,” he said.
“Once we’re able to teach the machines what we really want to teach them, that curve will pick up hugely. We’re not even in the early days, we’re in year two of five in terms of getting to that huge tipping point. We’ve gone beyond the hype now and I think we’re genuinely at that point of real inception.”