Here Are The Key Drivers That Are Pushing Cloud To The Edge
Today’s cloud computing architecture resembles 70s mainframes. The heavy lifting happens in dense data centers that act as the central point of gravity. In both the scenarios, applications share the underlying infrastructure. While this architecture works for the majority of the scenarios, the emerging use cases demand a different approach.
Edge computing fundamentally changes the cloud by making it distributed and decentralized. With edge, the core building blocks of cloud such as compute, storage, and the network will move closer to the applications. Cloud providers will move abstract layers such as machine learning models, serverless computing, and lightweight databases that run on the core infrastructure to the edge. Since the latency involved in making a round trip to the cloud is minimized, edge computing dramatically improves the user experience. The best thing about edge computing is that it is entirely transparent to the consumers. It is similar to what content delivery networks (CDN) do to static content and media streams.
Major public cloud providers including Amazon and Microsoft are heavily investing in edge computing. Industry consortiums such as EdgeX Foundry and OpenFog Consortium are trying to define the standards and reference architecture. Startups such as FogHorn Systems and Vapor IO are building new business models around edge computing.
Here are the key drivers that are pushing the cloud to edge:
Virtual, Augmented and Mixed Reality
The next generation user experience is going to be far richer than what it is today. Mobile devices pack more punch making it possible to double them up as virtual reality headsets. Oculus Rift, HTC Vive, Samsung Gear VR rely on mobile phones to deliver the experience. While these devices focus on virtual reality, Microsoft HoloLens, Magic Leap and Meta 2 are all set to unveil mixed reality headsets that take the user experience to the next level.
Though the initial versions of these experiences are focused on games like Pokémon Go, the use cases of VR and AR go far beyond that. Healthcare, automobile, construction, and education verticals will take advantage of these new technologies.
The new breed of VR and AR applications cannot afford to make a round trip to the public cloud. They require a single-digit millisecond response time to deliver a seamless experience. Cloud, in its current form, cannot be the backend for these applications. Edge computing becomes the essential backbone for the next generation VR and AR apps. Though rendering and processing will continue to happen on the public cloud platform, a large part of it will move to the edge.
It may be a while before we can find autonomous or self-driving cars picking us up from our doorstep. But smart cars and connected cars are a reality today. Major auto vendors such as Audi, BMW, Ford, Honda and Toyota are embedding sensors and self-diagnosing systems in the cars. According to Hitachi Data Systems (PDF), the amount of data generated by a connected vehicle may exceed 25GB/hour. Though the automobiles have inbuilt computers, they are not capable of dealing with extensive data.
Edge computing will make connected cars to become intelligent by providing diagnostics and predictive maintenance. The datasets streamed by smart cars will first hit the edge layer where it will get analyzed for anomalies. This layer can filter redundant data points and upload only the necessary data to the public cloud for further analysis. This architecture of connected cars talking to the edge layer will result in saving massive bandwidth costs.
Internet of Things
IoT is the primary driver of edge computing. Both consumer and industrial Internet of Things will benefit from the edge. Scenarios such as remote monitoring, predictive maintenance, smart factories will need local gateways. These appliances will perform protocol translation and data transformation before moving the data to the public cloud. The edge computing layer will run machine learning models that bring intelligence to connected devices.
Microsoft’s Azure IoT Edge and Amazon’s Greengrass are early examples of IoT-driven edge computing implementations.
Smart cities are a collection of smart buildings, smart homes, smart grids, smart hospitals and more. With everything from a street light to the water pump becoming connected, there is going to be a deluge of data. Streaming these massive datasets to the public cloud will prove to be very expensive.
Each smart city may have multiple edge locations to deal with the local data and intelligence. These micro data centers would be connected to the public cloud where the heavy lifting takes place. Similar to other use cases, the edge will become the backbone of smart cities.
The future of cloud lies at the edge. From IoT to smart cities, the next generation user experiences are heavily dependent on edge computing.
More on edge computing: More Organizations Will Look to Edge Computing for Data Center Insights
This blog is provided for informational purposes only and may require additional research and substantiation by the end user. In addition, the information is provided “as is” without any warranty or condition of any kind, either express or implied. Use of this information is at the end user’s own risk. CenturyLink does not warrant that the information will meet the end user’s requirements or that the implementation or usage of this information will result in the desired outcome of the end user.