The word “Edge” when used in the world of data centers can be difficult to define. It seems that if you ask 10 people in IT or Facilities to define edge, you will likely get 10 very different answers. Is it because it is misunderstood and mystified? Or is it because its potential use is so diverse? Yes, perhaps… Let us first look at the word edge. Edge is defined as “the outside limit of an object, area, or surface; a place or part farthest away from the center of something.” So, in data center terms, this means we will move something (like an application, IT hardware, or a process) further away from the center (like an enterprise data center) and bring it closer to where it will be consumed. Its principle is quite simple; bring the application closest to where it is intended to be used.
In today’s day and time I am sure you have heard someone say something like “we are getting out of the data center, as it’s not core to our business”, or “we are getting rid of our data center because we’re going to the cloud”. While many can agree that the operation of actually running a data center may not be a core competency for all organizations, the question isn’t whether you should be in the data center business; the question is how do you want to be in the data center business. With this question in mind, most medium to large enterprises are leveraging a hybrid strategy that includes cloud, co-location and on-premise; with the on-premise option looking more like an edge deployment. With that mentality; the practical application of edge can be almost limitless. Let’s look at a few and why they work well.
An office setting requires the need to house various types of technology infrastructure. These technologies can range from local and wide area networks that connect users and systems to data and to the outside world, as well as basic office systems like file, print, and storage. Edge solutions in these use cases will range from MPOP, MDF, IDF, network closets, and often small server rooms. These types of environments will likely only include a few racks of equipment (or less…), but they are critical to keeping a business operational. For any company that has workers onsite at some type of facility, it is highly likely that you need to connect that user and their device(s) to a network. Let’s face it, the network will never go away. If and when these classifications of edge take an unplanned outage, users and systems connection to the network is lost without access critical to applications. The office example is the most basic; and likely the least complicated example of edge, but non the less this is edge in its very definition.
Healthcare providers are about one thing. Patient care. On the surface, this sounds basic and straight forward, but believe me, it is anything but… In healthcare today, providers need to bring the best intelligence to the edge of their network, closest to the patient and healthcare delivery team. They need to deliver critical analysis in real-time. Just like all other industries, automation and connected systems are paramount in modern healthcare, hence the birth of “smart hospitals and clinics”. These “smart” facilities include a series of sophisticated integrated systems that have almost no tolerance for network latency. Healthcare professionals need to make the best decisions about a patient’s healthcare plan, sometimes in the matter of minutes or even seconds. This becomes even more complicated and critical when you combine this with the need for artificial intelligence and the ability to rationalize that data in a moment’s notice. No doubt that our healthcare industry has outpaced most other industries in technology adoption. Their need for stable and reliable real time solutions has increased the healthcare providers reliance on edge solutions.
Industrial and Manufacturing
For companies in the manufacturing business, edge is already (or should be) in your vocabulary. The delivery of information to and from an application to a device or process on a plant floor is typically time sensitive and has a specific tolerance for latency. In today’s cloud enabled everything world we live in, many manufacturing businesses continue to invest, or are in the planning stages to invest in edge solutions. The reasoning here is straight forward; applications that support a manufacturing process tend to break when taken off-premise from the plant regardless of the amount of network bandwidth you throw at it. Now add in IoT (Internet of Things) and Industry 4.0, many manufactures are moving to smart machines that automate the manufacturing process and produce mass amounts of data. These types of smart machines require precise sets of instructions, and the manufacture will want to utilize the data coming from that machine to learn, adjust and streamline operations. Again, the importance of a well-rounded hybrid strategy comes in to play, even for companies with a cloud only, or cloud first mentality.
Retailers today, are highly focused on strategies to increase margins, decrease costs and attract and retain potential customers. In addition to store operations, retailers are tirelessly looking for ways to enhance the guest experience, further develop loyalty programs, how to best time special offers and match or beat their competitors pricing. So, in addition to just keeping cash registers and credit cards readers operational, retailers now have a host of other technologies that are relevant to their success and these technologies require some level of care and feeding from an onsite perspective. This includes dedicated systems that connect in store technologies to a local network in addition to connections to corporate operations, warehousing, and suppliers. Retailers have a keen desire to learn as much as possible about their customers buying habits, how they consume marketing and instore messaging and how that impacts traffic patterns in a store. These types of technology all require highly integrated systems that rely on an edge infrastructure somewhere within that facility.
Telecomm, ISP and content delivery
Telco’s and Internet Service Providers are no strangers to edge, or even cloud for that matter. Content streaming services, bandwidth intensive mobile applications and on-demand everything has service providers rethinking their strategy on how they deliver their network to end users. As our reliance on information and our desire for limitless amounts of bandwidth increase, these service providers must bring their core processing as close to the end user as possible. This means deploying edge solutions in several different ways that will include multiple smaller edge sites in specific areas that are dense and have lots of users or high bandwidth requirements and larger edge solutions that might support multiple less populated areas. As the era of 5G comes upon us, edge deployments from telco’s, ISP’s and content providers will grow at an unprecedented pace.
Other applications for Edge
As mentioned, the opportunity for edge is almost limitless. There will be very few industries that do not adopt an edge solution of some sort to accomplish their goals, solve a business challenge, streamline a process, save money and/or a myriad of other drivers. Other practical uses for edge that we have not mentioned yet include things like; housing legacy applications that are not cloud ready or cloud enabled, intellectual property or applications that give a unique competitive advantage that businesses require to be kept on-premise. Things like 5G, autonomous vehicles and smart city will likely be the fastest growing segment in the world of edge, and an entire article could be written on this topic (and probably will…).
Written by: Jason Tannahill – Sales Executive, 2NSystems