The cloud is coming to the edge because latency matters

The past two weeks have marked a milestone for edge computing. In that time period, both Amazon and Microsoft have signed deals with large telco carriers to bring their cloud services to the edge. Both announcements focused on 5G and edge computing, but the word you should focus on in both these announcements is latency, because both deals are designed to bring computing closer to people and machines, with an eye toward reducing response time.

This week, Amazon announced partnerships with Verizon, KDDI, Vodafone and SK Telecom to offer its cloud services inside carriers’ networks to help ensure super-fast response times for applications on 5G networks. This deal is similar to a deal announced back in November between AT&T and Microsoft, which brings Azure’s services to AT&T’s cell towers.

With the Verizon-Amazon deal, Amazon is putting its hardware inside metro data centers operated by carriers. The goal is to allow developers building latency-sensitive use cases access to networks closer to their end users. It’s part of a new service AWS calls Wavelength. These data centers will act like an Amazon availability zone, although Amazon’s Raj Pai, vice president of EC2 for AWS, explained that Amazon is calling these “Wavelength Zones.” Developers will simply choose Wavelength Zones in cities where they want to deploy their latency-sensitive services, and Amazon will ensure the Wavelength Zones closest to the user field the traffic.

Initial customers include the NFL, which wants to build new services for fans in stadiums (think of a connected jersey that lights up when the home team scores a touchdown), and Mapbox, a mapping service that pulls in real-time traffic and street data to serve up to developers. Lower latency helps Mapbox adjust its maps immediately, which could help deliver real-time parking information, real-time traffic data and even influence wayfinding by autonomous vehicles.

Pai says that Amazon will continue signing partnerships with other carriers as it seeks to bring the Wavelength service to more places. AT&T’s deal with Microsoft Azure is similar, but AT&T is letting Microsoft put gear inside its radio access network. So far, the Microsoft Azure services will only be available in Dallas. Next year, AT&T plans to add Los Angeles and Atlanta. The idea behind both networks is the same, and I expect we’ll see AT&T sign deals with Amazon, and Verizon sign deals with Microsoft Azure.

The rationale for taking the cloud out of the data center and embedding it in the cellular network is to reduce latency, the time it takes data traffic to hop from server to server over the internet. While gaming and voice calls have always been latency-sensitive, basic web search and buffered video streaming have had more room for forgiveness. But with more devices connected, and those devices needing split-second reaction times, latency is becoming the buzzword that matters.

Consumer publications have been focused on 5G because it can offer a lot of capacity (those gigabit speeds are enticing), but the real game changer is latency that can be measured a few milliseconds. For example, when a sensor detects a problem in an industrial setting, it may need to transmit a command to stop an action before a machine is destroyed or a person is injured. Those commands need sub 200-millisecond latency. In virtual reality gaming, developers are trying to deliver sub 50-millisecond latency to avoid gamers becoming disoriented.

This reduction in latency is where 5G shines and where the upgrade to the 5G network will matter for many IoT applications. But those millisecond latencies don’t happen if the data is traveling over the 5G network, then hopping to the internet, then over the public network to the cloud and then back again. That’s the problem these partnerships are aiming to solve. And while it may not make sense for many applications today, it’s going to matter tremendously in the near future.