Its that time of the year again when we try to predict the future based on everything we have seen and heard over the course of the prior year (or years). This is a fun exercise that makes us put on our thinking caps and try to determine how new technologies or initiatives will help to shape the upcoming year.
So, with that preamble, how do I see 2019 shaping up for the networking space? Well, let’s start with a quick recap of 2018. It was a good year for networking, with several higher-level trends influencing the direction and importance of networking—things like digital and IT transformation, IoT, distributed or edge computing, and cloud computing. These initiatives have already started to have an impact on the network space as they all make the network more relevant to enterprises and force organizations to focus on connectivity and positive experiences. Other network-focused technologies emerged as well, including 5G, 802.11ax or WiFi6, CBRS, intent-based networking, and disaggregated network OSes, and we witnessed the SD-WAN proliferate.
So, what does the next year hold for networking? Here is what I am expecting for 2019:
- The 5G rubber will meet the proverbial road. With a number of standards-based (hopefully) 5G trials going live in late 2018 across a number of US cities, 2019 will be critical in determining the effectiveness of 5G networks and how enterprises can best leverage them. While early programs will tend to be deployed in parts of dense metro areas where small cells will be deployed in droves, don’t expect any 5G deployments in rural areas where distance and obstacles, like trees, walls, tinted windows, etc., can attenuate the signal. As providers look to monetize 5G with enterprises, more common applications could include the rapid delivery of a 10G connection for an enterprise. Especially when the alternative connectivity method includes digging up the street, which could cause significant delays (think northern cities in the winter). Other areas of focus for 5G will include large event centers like stadiums and potentially in robotic manufacturing plants for real-time control. Of course, to complete the testing we need 5G chips in the phones, tablets, computers, etc., which will be available in 2019.
- WiFi6 gains traction. Okay, no points for this, as it is more of an inevitable conclusion with the vendors releasing enterprise-grade technology this fall and consumer devices (phones, pcs, tablets) coming in the first half of next year. What I do find interesting and believe will really take hold is the name change because, for one, 802.11 ax doesn’t quite roll off the tongue and two, if competition with 5G is expected, why wouldn’t you want 6-? Can you picture the scene from This Is Spinal Tap? “Our amps go to 11.” So ultimately, my prediction is that the WiFi6 nomenclature will gain rapid adoption and the technology, which will drive far greater efficiencies, especially for high density environments, will be adopted over the course of regular refresh cycles. Organizations should ensure the supporting PoE(+ or ++) equipment is capable of supporting the WiFi6 APs’ appetite for power. Quick side note on the name change: Fibre Channel navigated this same transition years ago from speeds and feeds (2, 4, 8, 16G FC) to Gen5, Gen6, etc., and it caught on nicely.
As long as we are talking about wireless technologies, for US only, it is worth mentioning that CBRS looks like it may become a commercial reality in 2019 as well. It also has potential to provide value in high density environments or for in building LTE networks. And best of all, the spectrum will be free—it will still be licensed, but free. Expect to hear more about initial deployments in 2019.
- Automation gains importance. As networks scale to meet the needs of the business and accommodate the aforementioned transformations, the DC, campus, branch, and wide area networks will need to be more flexible and agile, and it will be increasingly difficult to operate these networks manually via CLI. Even though it is possible for IT staff to maintain complex environments through individual acts of heroism, history has shown us that it is not a long-term solution and automation will win out. This is especially true for cloud-native environments leveraging containers and microservices. The sheer volume of containers and rate of change requires automated solutions. As a result, organizations need to investigate solutions and learn new skills. Which brings up the fact that this transition is about both technology and cultural issues. I would strongly recommend embracing the new technology and helping to lead the charge toward increased levels of automation. How can you do this? First, get the right mindset that change is good and automation will help your business (and you personally) and second, seek out vendor training and education programs that will accelerate your knowledge. Regardless of whether your approach to network automation is intent-based or disaggregated, understanding closed loop automation leveraging AI/ML and enhancing your programming skills (instead of memorizing CLI commands) will be critical for your future growth and will help enable your business.
- The connectivity and experience mantra will continue to grow in importance. Organizations need to understand the value of being able to connect from anywhere and to anything, safely and securely. It includes connecting applications to data sources, IoT devices, clouds, users, and customers or other applications regardless of location. Instead of talking about quality of service (QoS) as the key performance indicator (KPI), organizations need to think in terms of quality of experience (QoE) as a new metric for success. As the pendulum continues to swing toward distributed compute (i.e., the edge), the network will be critical in enabling those connections and ensuring positive quality of experience. Organizations need to have high performance, end-to-end connectivity between all locations and public cloud services to remain competitive.