Cloud computing is a bigger thing than most people realize. Its influence is not only felt in the business world, where earlier local servers have been replaced by more nimble external alternatives. Thanks to services like Google Photos and Netflix, even the layman mixes parts in the ethereal data center in the sky. But is another revolution afoot?
We are talking about edge computing. This new paradigm in IT aims to bring remote data centers closer to the people who actually use them. It is particularly ideal for time-sensitive applications where low latency is a must. Here̵
In the beginning there was the server
For edge computing to make sense, it’s helpful to put it in a historical context. So we start at the very beginning.
Corporate IT used to be a static matter. People worked in huge cabin farms and worked under the harsh halogen light. It made sense to have their data and business-critical applications nearby. Companies moved servers to well-ventilated premises on the premises or rented space in a local data center.
Then things changed. People started working more from home. Companies grew and opened offices in other cities and countries. The local server quickly made no sense – especially in view of the enormous increase in Internet usage by consumers. Technology companies find it difficult to scale when they have to buy, provision, and provision new servers every few days.
Cloud computing services like Microsoft Azure and Amazon Web Services (AWS) have solved these problems. Companies could rent storage space on a server and expand as it grows.
The problem with the cloud in its current incarnation is that it is centralized. Vendors like Amazon, Microsoft, and Google have data centers in most locations, but these are often hundreds – if not thousands – of miles away from their customers.
For example, if you are in Edinburgh, Scotland, the closest AWS data center is in London, which is approximately 100 km away. If you are in Lagos, Nigeria, your closest AWS continental location is in Cape Town, South Africa, which is nearly 3,000 miles away.
The further the distance, the higher the latency. Remember that data is just light traveling through a fiber optic cable and is therefore limited by the laws of physics.
So what’s the solution? The answer arguably lies in history repeating itself and bringing the servers closer to the people who use them.
A life on the edge
In summary, edge computing means that applications and data storage are brought closer to the location of the user. For large companies, this could include a dedicated server facility in close proximity to their main offices. On the consumer side, it can be helpful to envision IoT devices that perform certain tasks, such as facial recognition, using their own local computing resources rather than handing them off to a cloud service.
This has several advantages. First, it helps reduce the network traffic to be sent. Given that many large companies often pay high fees for mixing bits between data centers, it makes sense to move them closer to their home.
Second, it reduces latency. Often, a large portion of the time required to complete a task is spent moving traffic across the network. Bringing the computing power closer to your home can reduce that latency and increase the speed.
This could potentially open the door to new forms of computing for which immediacy is key. An often touted example is a “smart city” where the local government can gather real-time information about things like utility usage and road traffic patterns and act quickly.
There are also potential uses for edge computing in the industrial sector. This includes allowing manufacturers to collect data about devices and make quick adjustments to reduce energy consumption and device degradation.
On the consumer side, edge computing has the potential to make things like cloud gaming a more satisfying experience. If the graphical calculation is closer to the players, there is less chance of an uncomfortable delay. This can be the deciding factor in who wins an online game.
The 5G factor
The introduction of 5G connectivity goes hand in hand with the steady rise in edge computing. Although it is still in its infancy, 5G promises significantly lower latencies than previous cellular standards. As a result, you can expect it to play a huge role in the evolution of edge computing as a paradigm.
What does that mean? In the logistics industry, the focus is on analysis and data, as trucks and vans transmit information that needs to be analyzed and processed in real time. There is also the prospect of “smart agriculture” that will automate large parts of agricultural production. This not only improves crop yields, but also prevents waste.
Then there is the consumer side. By bringing computational “heavy lifting” closer to people’s phones, you can unlock newer, richer entertainment experiences for things like virtual reality (VR), augmented reality (AR), and games.
That is of course still a long way off. Carriers and developers have to build it first. If it does, however, you can expect the same seismic change that occurred when cloud computing first hit the scene.
CONNECTED: What is 5G and how fast will it be?