Networking becomes more relevant. I refer to this as my Theory of Network Relevancy.
While traditional data center networking may be slowing due to the adoption of cloud computing and edge computing, I would argue that networking solutions will become far more relevant in 2018 and beyond. So, while data center networking may be slowing, campus/access and WANs will continue to increase in relevance and importance to the business.
I refer to this as my Theory of Network Relevancy. Yes, this is a blatant rip off Einstein’s Theory of Relativity, just without any of the science or mathematical formulas to support it. I would think however, that additional complexity is not needed. There is enough of that already. Instead, this is based on common sense and looking at the drivers creating the need for better, faster, more secure, and more reliable networking. Some of these drivers include:
- The pendulum is swinging back. The cloud and IoT era are following a massive era of IT and data center consolidation, where organizations went from 80-100 data centers to six or seven. (HPE and IBM were very public consolidation efforts). So, everyone knew it would only be a matter of time until the pendulum swung back to distributed computing. The reasons are obvious, if you think about IoT and autonomous driving cars: Do you really want information sent all the way back to a centralized location, possibly located thousands of miles away (read latency), and back while your car tries to decide if it is going to stay on the road or plunge into the river? Just to make this somewhat scientific, the reason that would not be good is that speed of light is fixed, and going across the fiber (not in a vacuum), with the distance it can travel in a micro second being the speed of light (299.792m/us) multiplied by the index of refraction of the glass fiber (which can vary slightly, but let’s roughly call it 1.47), would result in unwanted latency (i.e., potentially you and your car in the river...not good). Okay, that is all the science for today. So, the local networks need to be robust as well the back haul to consolidate all the data for big analytics to take place.
- Cloud computing. ESG research points to the fact that more organizations are adopting cloud computing, and not just for CRM as used to be cited. Organizations are deploying production applications and many have cloud first policies. These organizations are also leveraging a multi-cloud policy for availability, and pricing competition. Therefore, secure and highly performant network connectivity will be essential to ensure a successful transition to the cloud. Keep in mind that this will not just be links from the data center, but potentially network connectivity from every remote and branch office as well. Global enterprises with hundreds of locations will find this a daunting task, but again will make the network much more relevant and important to get it right. If you are thinking this is a good fit for SD-WAN, you would be right. And because of the global logistics, expect to see global telcos and MSPs playing a more active role in this. Think of managing one or potentially two suppliers for all your cloud connectivity vs. tens or hundreds of suppliers.
- Campus/access networks. These networks are becoming increasingly dependent on wireless connectivity and the number of devices continues to increase, especially with IoT. As a result, it is imperative that employees and guests have a good experience on a WiFi connection, and it is just as important that it is a secure and segmented connection! While UC apps and voice in particular may require wired connections for now, as new technology emerges like 802/11.ax and 5G networks, users will have more options for more performant wireless technology.
- Intent-based networking. This is another area that will help make the network more relevant. Given the above points about cloud, IoT, and just the pace of business in general, the network is going to have to be an enabler, not an anchor. It will need to have more abstraction, awareness, automation, and intelligence. Those are all things that intent-based networking promises to deliver. I expect to see this deployed in stages, either in a specific domain, or perhaps getting started with visibility and verification to help change the culture. Don’t expect your network admins to have Alexa architecting and building out your network, but I also wouldn’t be surprised to see solutions like that become interfaces for network solutions.
No matter how you look at it, there needs to be a network to connect everything and everyone, such as applications to applications, applications to users, applications to machines, machines to machines, and so on. It only makes more sense that the network will be more relevant in 2018 and for years to come. Stay tuned for more on my Theory of Network Relevancy.