Edge computing sounds exactly how a futuristic networking architecture should. Not only is this sharp, crackling phrase free from all the fluffy trappings of the cloud, it’s reminiscent of technology that would be featured in “Blade Runner” —perhaps something you’d use to power your flying car between neon skyscrapers.
But behind the cool name lies a pretty practical concept: Computing is swiftly moving away from mere cloud storage and sprawling data centers to include mesh networks of microdata centers. When developers talk about “the edge,” they’re referring to a network where the majority of processing occurs near devices rather than traveling to a data center and back. And that difference could transform how networks are structured.
How the internet of things is driving developers to the edge
The biggest factor propelling networks toward an edge model is speed. In today’s cloud storage architectures, the majority of processing happens in faraway data centers, which means computing isn’t as efficient as it could be. Those delays have been acceptable so far because they’re so small—we’re talking milliseconds. This is fine for conventional processing needs, as you’re not going to notice the lag when you’re downloading a file or streaming a video, but when you start adding IoT devices to the equation, those delays suddenly become untenable. For automated industrial applications like an agricultural drone or a smart traffic light, a split-second delay could make a huge difference.
Meanwhile, the sheer volume of IoT data presents a problem, too. Take health care monitoring, for example. Soon, sensors and wearable devices will allow health care professionals to harvest huge quantities of patient data. It’s easy to see how continuously tracking vital signs like glucose levels and heart rate could improve the efficiency and accuracy of health treatments, but the amount of data generated makes conventional cloud processing cumbersome—particularly when a real-time response is called for. Edge computing moves data processing away from data centers and closer to actual devices and applications, offering the ability to handle more data quickly.
Security predictions in the age of edge computing
As with any new technology, there are tradeoffs. For instance, the transition to the edge could potentially open up new security issues, since IoT devices don’t have the best reputation for user security and privacy. Edge networks, therefore, will have to be incredibly well designed in order to keep data safe. Moreover, developers may have to increase their reliance on VPN and encryption, especially since sensor networks could be processing sensitive data (as in the health care use case).
Adding to that is the vast range of technologies connecting to the network, each with their own security protocols and access levels. In the future, the technologies using edge networks may include wireless sensor networks, peer-to-peer ad hoc networking, local clouds, distributed data storage and retrieval, augmented reality and the like. Securing all those different systems will take plenty of work and coordination.
Furthermore, IoT devices open up new threats beyond leaks and data breaches. Once users start handing over control to smart devices, even their physical safety could be compromised. The FBI has already expressed concerns about vehicle hacking, in which cybercriminals remotely hijack automotive systems online. While both are very serious, the idea of a data breach seems like nothing compared to your transmission shutting off when you’re going 70 mph down the highway!
Bridging the gap between today and tomorrow
These issues present real challenges for developers hoping to take advantage of edge computing in the near future. That said, the transition won’t happen all at once, which gives developers time to address security issues. Similarly, it’s unlikely that edge computing will ever totally replace conventional cloud storage and processing; rather, it represents an extension of the existing cloud applied to a specific subset of use cases. As computing applications gain complexity, we’re more likely to see a mixture of different network architectures working together than one technology replacing another.
Of course, knowing about the outlook for the future will only take you so far in terms of addressing your present needs. For more insight into today’s cloud storage options, we recommend downloading our “Does Your Business Belong in the Cloud?” e-book to help you navigate different commercial applications as move your business from the center of the cloud to the edge.