With the ever-growing computing capabilities edge computing, there’s a new stage of computing coming with the Internet of Things (IoT). As we have explained in previous articles, edge computing isn’t going to push cloud computing into being an outdated technology. Instead, they’re going to work in conjunction with each other to bring faster speeds and more possibilities. Call it edge cloud computing.
When is edge cloud computing useful?
By 2020, it’s estimated that “as many as 50 billion devices will be connected” to the cloud, including our cars, communication devices, and even clothing. This means that the traffic generated from these devices will skyrocket even more.
With this excess of data, we can use IoT devices to “turn them into cloud servers and extend central-cloud capabilities to the edge.” Soon, edge devices are expected to handle the majority of the cloud computing burden.
In the opinion of Gartner’s 2017 Hype Cycle for Emerging Technologies, edge computing has the opportunity to become an “innovation trigger” based on the fact that microchips and sensors are becoming prevalent in numerous objects that we use daily.
While the number of potential edge servers is already huge, it’s expected to continue growing. The reasons these IoT devices will become a part of an edge computing network are easily seen if you look at some case examples.
For instance, there’s no question that this speed is necessary when self-driving cars need to make split-second decisions on the road, but it’s also useful when companies want to stream augmented reality experiences to their users. Augmented reality wouldn’t feel much like reality if everything that was happening on your device lagged, after all.
If a self-driving car needs to know which road it should take or what the weather conditions are, the faster it can get these responses, the better. The cars need a lot of computing power to do things like avoiding accidents, optimizing traffic flow, and finding the best routes for you to get to your destination quickly and safely.
While one might immediately think of those giant server farms to get enough power for this, it’s simply not fast enough to be a really effective solution.
The amount of data they’d send and receive is massive, expected by Toyota to reach 10 exabytes per month by 2025. Unfortunately, sending and receiving all of this data is expensive. In fact, “there are no networks in the foreseeable future that can be used to send all this data back to the central cloud for processing.” Most, instead, should be locally processed.
Not to mention, these cars need the utmost speed, and there’s no question that we can’t put these giant datacenters that can take up as many as 19 city blocks everywhere. According to Zachary Smith, CEO and co-founder of New York City startup Packet, this is exactly why we should distribute these edge networks. This would make sure client devices can get processing power more quickly, making things like autonomous cars much safer.
Using distributed edge networks rather than large, centralized datacenters would make sure client devices can get processing power more quickly, making things like autonomous cars much safer and more efficient.
And here is where we see exactly why edge cloud computing will become so important in the future. Considering all of these potential downfalls, the cars can become “‘data centers on wheels’ where most of the communication and processing is performed as close as possible to the edge.”
If we switch to utilizing edge computing, cars will be able to process this large amount of data quickly and efficiently by “leveraging not only the processing power at the ‘central cloud’ but also their own collective computing, storage and memory resources in a collaborative fashion with other cars on the road.”
What’s already changing?
Microsoft has already announced it is testing Azure IoT Edge service, which is made to help transfer some of the cloud computing functions to the developers’ devices. You can now preview this via their website. Almost immediately after this announcement, Amazon Web Services also allowed access to their AWS Greengrass software that performs a similar goal.
AT&T also recently announced its own plans to “build a mobile edge computing network based on 5G, with the goal of reaching ‘single-digit millisecond latency.’” They would accomplish this by having the data travel the short distance between the device and closest cell tower or central office rather than the much larger distance to a cloud datacenter.
Alongside Packet, many other startups are moving toward edge computing. Vapor IO, for example, has also begun building micro-datacenters next to cell towers that are already there. In fact, they recently announced a partnership with Crown Castle, “the largest US provider of shared wireless infrastructure.” This, called Project Voluntus, gives Vapor IO access to 40,000 existing cell towers and 60,000 miles of fiber optic lines in metropolitan zones.
One of the main goals of this startup is to remotely operate and monitor these datacenters so customers won’t ever have to experience interruptions in service, even if some computer servers do go down, according to Vapor IO’s founder and CEO, Cole Crawford.
As we’ve explained before, though, don’t expect edge cloud computing to push typical cloud computing and massive datacenters out of existence. Instead, think of the edge more of an extension of typical cloud computing.
A large push toward edge cloud computing is expected in the future, with services such as Amazon CloudFront already using edge locations to speed up the delivery of their data. Even the game-streaming company Hatch teamed up with Packet to deploy their service in a way that lets customers instantly play games on their smartphones without waiting for a download.
This growth is expected to continue, particularly considering the amount of data people typically generate today using things like mobile Internet and social media along with the profound growth of IoT devices.
With edge cloud computing, we can transform this massive amount of data from a challenge to an opportunity, making the large number of devices and data a part of the solution.
Photo credit: Pixabay