Applications are the ravenous beast of today’s dynamic hybrid-cloud environment, causing network demand to spike dramatically in milliseconds. Fiber-optic plans offer high speeds, but most businesses still subscribe to fixed network packages and have not yet recognized the interdependence of storage, compute and network in cloud computing, according to Paul Savill (pictured), senior vice president of core network and technology solutions at CenturyLink Inc.

“When you deploy an application in the cloud, you can’t just think about the application,” Savill said. “You have got to think about the network that ties it all together. It is a sea change …  we’re starting to see the network really start to play a major role in how cloud services and how performance of cloud-based applications are delivered.”

Savill spoke with John Furrier (@furrier), host of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, during AWS re:Invent in Las Vegas. They discussed how cloud adoption and intelligent analytics are making connectivity a hot topic once again.  (* Disclosure below.)

Microservices for cutting-edge retail technology

The synergy from its merger with Level 3 Communications Inc. in November 2017 has encouraged CenturyLink to push the boundaries of technology, expanding its microservices for artificial intelligence and the internet of things and providing dynamic content delivery with over-the-top video solutions and Vyvx live event broadcasting.

“Level 3 was thinking about how to make the network consumable on a dynamic basis and on-demand basis the same way cloud is … [while] CenturyLink was really focused around building out cloud services,” Savill said. Combining the two is “opening up so many new things for us to do, so many new ways that we can deliver value to our enterprise,” he added.

Retail robotics and video retail analytics are two cutting-edge use cases where CenturyLink is putting its new network technology to work. Hybrid is essential for these use cases, according to Savill, as on-premises deployment (putting a hardware stack in each retail store) is prohibitively costly, while latency issues make performance a problem for cloud.

“We believe that what we’re starting to see is this transformation where applications are going to be broken up into these microservices where parts of it’s going to run in the cloud core, part of it’s going to run in the prem, and part of it’s going to run on the near-edge,” he said.

Using video analytics to monitor customer’s facial expressions and identify where in the shopping process they are most happy, or if there is a part of the store where shoppers often become frustrated, is driven by advances in artificial intelligence. But simultaneously feeding from multiple video sources to a distant AI in the cloud “is just too much of a lift in terms of bandwidth and too much of a cost,” Saville said.

The answer is a distributed model where “portions of the application in the AI act at different locations in the network and the network is tying it all together,” Saville concluded.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of AWS re:Invent. (* Disclosure: CenturyLink Inc. sponsored this segment of theCUBE. Neither CenturyLink nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE