Unveiled Monday at HPE Discover EU in London, The Machine is a reborn computer that flips conventional IT architecture on its head. To discuss how storage is impacted by Hewlett Packard Enterprise Co.’s groundbreaking efforts to significantly speed up data retrieval, Milan Shetti, CTO of the Data Center Infrastructure Group at HPE, joined Dave Vellante (@dvellante) and Paul Gillin (@pgillin), co-hosts of theCUBE*, from the SiliconANGLE Media team.

Shetti’s expectation is that in the near future, the interactions between storage and computing would be focused in three “topologies.” The first two of these cover fairly traditional forms, under the groupings of rack and stack and direct external attachment to computers.

The third would be memory-driven computing, with the idea that “the working set [of data] … is really small,” so by putting it into a centralized place and allowing simultaneous access by multiple computing devices, each application would theoretically need less computing power.

“Storage will actually exist in all three places,” Shetti stated, but from the perspective of his division at HPE, that third place will be the one driving innovation.

Enter HPE’s The Machine, a computing architecture intended to drive computer functionality with data and memory rather than processors.

“What we wanted to show with The Machine initially was that when persistent memory comes in … what the world would look like if you never have to be afraid of reboots,” Shetti explained. “The way you would handle things would be very different.”

Faster links

Part of this change was motivated by the rising utility of containers in services and applications, and the delays that were being discovered in larger environments relying on that arrangement. As Shetti noted, data proximity to containers is a factor to be considered in the timing of reboots, accessibility and other qualities of operation.

“When you’re scaling services, and when you’re thousands and thousands of machines … there are large outages … because the limits of the cloud infrastructure erupts very fast,” he said.

One benefit mentioned in making this new shift would be that less infrastructure would be required to accomplish the same regular tasks, decreasing resource consumption, while decoupling the servicing of hardware from servicing of the applications would bring its own host of improvements.

Looking at the rise of Internet of Things dependencies, Shetti acknowledged that there simply was not enough bandwidth availability to shuttle their data back and forth from central data locations, necessitating improvement of edge computing. Speaking for HPE, he shared his company’s belief “that the center of the universe is going to be not compute … [but] data and memory,” with expectations that the new architecture would do for persistent memory what smartphones did for flash storage.

Watch the complete video interview, and be sure to check out more of SiliconANGLE and theCUBE’s coverage of HPE Discover EU. (*Disclosure: HPE and other companies sponsor some HPE Discover EU segments on SiliconANGLE Media’s theCUBE. Neither HPE nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo by SiliconANGLE