Via Network World : Predicting the end of cloud computing
This Andreesen Horowitz VC believes the cloud will be replaced by edge computing
Here’s a prediction you don’t hear very often: The cloud computing market, as we know it, will be obsolete in a matter of years.
The provocateur is Peter Levine, a general partner at venture capital firm Andreessen Horowitz. He believes that the increased computing power of intelligent Internet of Things devices, combined with ever-increasingly accurate machine learning technologies, will largely replace the infrastructure as a service public cloud market.
The cloud as its know today is a “very centralized model” of computing, Levine says. Information is sent to the cloud where it is processed and stored. Many applications live in the cloud and whole data centers are being migrated to it.
Levine says we’re already starting to see signs of Internet of Things devices replacing some of the computing power of the cloud. There are smart cars, drones, robots, appliances and machines. Each of these devices collects data in real time. “(Given) the latency of the network and the amount of information, in many of these systems there isn’t time for that information to go back to the central cloud to get processed,” he says. So, he argues, the edge of the network will be forced to become more sophisticated. He adds: “This shift is going to obviate cloud computing as we know it.”
Take a self-driving car as an example. It needs to be able to identify a stop sign or a pedestrian and act on that information instantaneously. It can’t wait for a network connection to the cloud to tell it what to do.
This new world will not eliminate the need for a centralized cloud, Levine says. The cloud will still be where information is offloaded to, where it is stored for long periods of time and where machine-learning algorithms get access to the vast troves of data they need to become ever smarter.
The idea of edge computing becoming more powerful is not original. Cisco is credited with coining the term Fog Computing, which is the idea of analyzing and acting on time-sensitive data at the network edge.
All this is a ‘back-to-the-future’ moment, Levine notes. Computing began with a centralized model focused on the mainframe. Then came the distributed client-server world. The cloud has swung computing back to a centralized platform. This dawn of edge intelligence will once again swing the world back to distributed system.