How Edge Computing Will Change Data Centers
Over the past decade or so, the trend in computing has been away from having the computing power primarily in individual devices, and toward data centers. This is primarily seen with cloud computing where businesses would have much of their CPU processing, data storage, and even memory located in large data centers, which were often built and managed by a third party.
This trend helped to reduce the overall cost of computing, and also increased reliability since most things were spread out over multiple physical servers. Thanks to the fact that the main Internet backbones run on fiber cables, and a growing number of secondary circuits are also on these types of circuits, the time it took for data to get to and from the data centers was insignificant.
Today, however, there are many applications being developed that rely on near instant communications, and as Cole Crawford, the founder and CEO of Vapor IO says, the “Speed of light sucks.” To put it simply, this means that even fiber connections, which transmit data using light impulses, isn’t fast enough in many cases. To address this issue, the industry is beginning to shift away from cloud computing, and toward edge computing.
What is Edge Computing?
Due to the fact that this is still a very new concept, many people don’t have a great understanding of what it is. Without getting into technical specifics, suffice it to say that the concept of edge computing is the idea of putting powerful processing and data storage as close to the end user as possible with the intent of minimizing the distance the data has to travel. Since the laws of nature prevent any possibility of transmitting data faster than the speed of light, and nobody has come up with a way around that, the only option is to reduce the distance.
Why is Edge Computing Necessary?
Looking at what is driving the need for edge computing can also help to understand what it will be used for. The following are different types of technologies that are either pushing the need for this type of computing, or will be taking direct advantage of it as it becomes more widely available.
- Internet of Things – The IoT is growing at an incredible pace, with millions of new devices being made and purchased each year. These devices often have minimal on-board computing power, but need to transmit and receive data immediately in order to function properly.
- Augmented Reality – Augmented reality is growing rapidly for entertainment, medical, industrial, and many other industries. In order for AR to function in a way that users can benefit from, it is necessary to transmit and receive data at near-instant speeds.
- Autonomous Vehicles – Self-driving vehicles need to be able to gather and process huge amounts of information extremely fast. Some of this is done locally, but much of it will need to be sent to edge computing data centers to be handled by faster and more powerful equipment.
There are many other things that will take advantage of this type of computing, and while it won’t entirely eliminate the cloud computing industry, it does represent a shift in what types of services data centers will need to provide.
How Will Edge Computing Impact Data Centers
Over the past several years, the industry has been focused primarily on building extremely large data centers, often with 1 million+ square feet of floor space. These data centers could hold thousands of servers, and meet the huge needs that came with cloud computing. These data centers were built in specific geographic locations that were chosen based on things like tax benefits, weather patterns, workforce availability, and more.
While the demand for these mega-data centers will continue for some time, it will also be necessary to have a larger number of smaller data centers, which are placed near population centers to meet the demand of edge computing. This demand may be met by existing data center companies that are well-established in the industry, or it could drive innovation from new or existing companies that focus on this emerging market.
Whether this means constructing all new facilities, retrofitting existing facilities, or creating an entirely new solution has yet to be seen. What is almost certain, however, is that edge computing will need to be available not only to major cities like New York, Chicago, and Houston, but also ‘second-tier’ cities, and possibly even mid-sized communities.
How this will be accomplished is not yet entirely known, and will likely take multiple different approaches. We can be confident that the data center industry will innovate and adapt to meet this growing demand.