Over the last several years, the edge has been one of the most over-hyped and misunderstood areas in the telco space. Although the debate goes on about exactly what and where it is, one thing that analysts and futurists have agreed on is that the edge will be an integral part of IT and network strategy going forward, discusses Ben Baker, senior director of cloud & DC marketing and business analysis at Juniper Networks.
The number of active sensors in households, factories, smart cities and even people’s wrists is growing at a tremendous rate. It’s estimated that the amount of data generated in 2025 will be well over 175 zettabytesOpens a new window .Â
The capacity to carry all this data from its sources to conventional, core data centers doesn’t exist, and the chances are that it never will. Currently, roughly 10% of enterprise data is already processed outside a traditional data center or cloud. According to Gartner, that figure could reach 75%Opens a new window by 2025.
Given these numbers, the status of edge computing and the edge cloud technology that’s an integral part of it has evolved from an innovative idea to an urgent necessity. It’s time for companies to turn their attention away from futuristic use cases hyped in the media and focus on the practicalities of deploying and operating the infrastructure required to deliver new services that are in increasing demand right now — services that can provide ROI in an acceptable time frame. This means focusing on the edge.Â
See More: The Unexpected Role of 5G and Telcos in Helping Industries Reduce Carbon Emissions
Business and Technology Drivers
There are several use cases where the physical proximity of users, data and applications can result in a user experience that’s a competitive differentiator. Two prominent examples are:
-
- Content delivery networks (CDNs): These were the early adopters of edge computing, and they continue to grow in many dimensions – quantity, coverage and breadth of services, to name a few.
- Communication service providers (CSPs): These companies are making sizable edge deployments to serve their users with low-latency compute and storage for activities ranging from gaming to augmented reality to real-time machine learning to foundational radio access network RAN infrastructure.Â
Many large enterprises in other sectors are becoming interested in edge computing, notably retail (for inventory tracking) and manufacturing (for factory floor automation). When the IoT data created by these activities is substantial, for example, in applications such as video surveillance, that data needs to be processed locally. Any other approach is simply not practical.
Beyond operational improvements, edge computing provides another important benefit: the significant savings companies can achieve in network traffic costs when data is processed locally rather than transmitted to a distant, centralized cloud.Â
In addition to the business drivers that make edge computing attractive, there is a hugely important technology driver, 5G. It’s linked with the edge in two ways. First, edge clouds will be a part of most 5G deployments because the radio access network (RAN) and other components are virtualizedOpens a new window . Second, the new capabilities of 5G will spawn a host of multi-access edge computing (MEC) applications, which will require more edge data centers.Â
Practical Paths to Profit
Whether driven by operational considerations, cost savings or the rise of 5G technology, the benefits of edge computing — at least for now — derive from very practical reasons that don’t make for catchy headlines. Distributed data processing can improve reliability as well as the user experience. Data sovereignty requirements and privacy and security regulations may dictate local storage and processing. Long-distance xHaul transmissions may result in unacceptable latency.Â
In contrast to exotic applications, addressing these practical issues through a successful transition to the edge will deliver immediate business benefits.
Challenges and Solutions
The need to deal with almost unimaginable quantities of data will force many companies to transform their operational paradigm to maintain profitability. CSPs are one of the best examples of this new reality. Edge computing in this vertical will require large numbers of small, geographically dispersed data centers, which may consist of as few as two hosts. Managing hundreds and sometimes thousands of these data centers is a new and daunting challenge.Â
From a human capital perspective alone, it’s unlikely that each of these data centers will have dedicated staff. In fact, it’ll be challenging to staff network engineers on-site during the test and turn-up processes, not to mention the inevitable issues that come up after the initial deployment.Â
It’s simply impossible to economically manage a multitude of remote, dispersed data centers without a heavy dose of automation. This begins with an operational template that can easily be replicated wherever a new point of presence is required. The commonality and consistency of such a deployment strategy are the basis for reliability and, ultimately, speed. Common data center fabric management and high levels of automation are imperative. This is the general approach the cloud hyperscalers have taken, and it has been proven to work.Â
As with any new operational paradigm, shifting to the edge will present challenges. Some companies will face skill set issues. There will be organizational issues as well. The most significant may be cultural. Telcos have been focused on transmission for more than a century, but cloud infrastructure will be as crucial as transmission in the new services (and sources of profit) they are beginning to offer. Fortunately, the technology to support these services with reliable, field-proven automation is available now.Â
How are you dealing with the operational challenges of moving to the edge? Tell us on FacebookOpens a new window , TwitterOpens a new window , and LinkedInOpens a new window . We’d love to know!