Decentralized Cloud: The Next Big Step

By Michal Katz
Published March 1, 2019 | 5 min read

RBC’s Michal Katz looks at a paradigm shift set to unleash the power of computing devices at the edge of today’s hyper-connected world.

A paradigm shift is set to unleash the power of computing devices at the edge of today’s hyper-connected world. By 2020, over 50 billion devices will be connected to the Internet – all generating data that cannot be effectively and efficiently processed and managed in real time by a central cloud. Future demand will be immense - bandwidth, latency, security and cost will be big challenges. As such, the edge must become more sophisticated.

Instead of deploying new infrastructure, the industry is looking at how we take advantage of the ever-increasing computing capabilities of edge devices (mobile phones, cars, drones, robots and IoT objects).  For example, today’s car has approximately [100] CPUs, iPhone 7 has 3.3 billion transistors, and the collective computing power of 50 million Sony PlayStation 4s is roughly four times that of Amazon Web Services 27 data centers around the world.

Put simply, there are billions of connected devices which collectively have millions of times the computing resources of central cloud.  Those become a massive distributed computing system at the edge of the network. Most of these devices are under-utilized, except for perhaps a few hours in the day, meaning vast unused computing resources at the edge sit idle most of the time.

The Impact of a Decentralized Cloud

So what impact will this decentralized model have, not just on the technology sector, but on investors and the wider economy?

The move to the edge is reminiscent of the evolution from mainframe to distributed computing of the 1980s, with the big difference being that today’s transformation presents an exponentially greater opportunity. Whereas prior generations of computing objects were tethered to humans (users), the Internet of Things opens the market to billions, perhaps, trillions of devices, which will need to be managed and secured.

With the massive amounts of data collected at the edge, there will also be a need for machine learning technologies to process and analyze information at the end point.  This creates an opportunity for data scientists and new solutions for deciphering the unstructured data collected and supported by the edge’s backbone.

Edge computing solutions can also take many forms: self-driving vehicles, smart phones or sensors in hospitals, manufacturing plants or offshore oil rigs. These will open up opportunities across various traditional industries such as insurance, maintenance and repair and logistics.

It will have social and economic implications. Decentralized cloud is more private in theory since it minimizes central trust entities. And it should be more cost-efficient since it leverages unused computing resources at the edge.

But make no mistake. The edge cloud is not a replacement for central cloud. Some applications will remain better suited to computing resources in data centers. However, central cloud (servers in data centers) should not be the bottleneck, but rather work along with edge devices to build distributed edge cloud architecture.

As with all rapidly evolving technologies, evaluating, deploying and operating edge computing solutions will create friction and disruption. Business models will need to change in order to adapt to this high tech, decentralized marketplace.

The transition to the edge is already underway. Edge computing’s primary gift is that of power: it re-distributes it, which will undoubtedly cause a significant disruption in industries and business models.

Here we go again . . .


Michal Katz
Global Co-Head, Technology Investment Banking


CloudConvergenceDisruptionEdge ComputingInnovationMichal KatzTechnology