AT&T says it is embracing edge computing to boost the potential of self-driving cars, augmented and virtual reality, robotic manufacturing, and other 5G apps by moving the data crunching from the device to the cloud.
The challenge with next-gen applications like autonomous cars and AR/VR, AT&T said, is that they demand massive amounts of near-real time computation.
For example, according to some third-party estimates, self-driving cars will generate as much as 3.6 terabytes of data per hour, due to the clusters of cameras and other sensors required to enable their digital vision. While some functions, such as braking, turning and acceleration, will likely always be managed by the computer systems in the cars themselves, secondary systems like detailed navigation maps could be offloaded to the cloud.
In the case of AR/VR, the industry is moving to a model where those apps will be delivered through smartphones, but most devices don’t have the processing power to deliver a rich experience – and if they did, the tradeoff would be extremely short battery life.
Edge computing addresses those obstacles by moving the computation into the cloud in a way that feels seamless, said Andre Fuetsch, president of AT&T Labs and chief technology officer
“Edge computing fulfills the promise of the cloud to transcend the physical constraints of our mobile devices,” said Fuetsch. “The capabilities of tomorrow’s 5G are the missing link that will make edge computing possible.”
The faster speeds and particularly the lower latency expected with 5G will be critical elements to enabling edge computing. But latency is also determined by the physical distance between a mobile device and the network resources to which it’s connected.
For example, VR users are likely to experience a noticeable delay between the moment they turn their head and the moment the image moves to follow their gaze if the data center powering the app is hundreds of miles away.
The solution, AT&T says, is to leverage nearby central offices, macro towers, and small cells to do the edge computing necessary to reduce latency. AT&T says it will outfit such facilities with high-end graphics processing chips and other general purpose computers, and coordinate and manage them via its software-defined network.
Eventually, AT&T adds, it could embed such systems in traffic lights and other infrastructure.