5G, the next generation of wireless network technology, is causing a radical shift in the telecom space and soon in enterprise networks. While 2G, 3G and 4G were faster versions of the original mobile network, 5G didn’t derive from antiquated standards. It is the first technology standard in networking that was born in the cloud era and utilizes software-defined principles, strict policy enforcement and cloud-native microservices.
The transformative principles of 5G are paving the way for a new “edgeless” enterprise architecture, which would allow organizations to quickly deploy key network and policy services wherever their computing resources may be. No wonder edgeless enterprise is one of the hottest terms in networking right now. But it’s also one of the least understood.
So, what does the term really mean, and how will 5G help organizations achieve an edgeless enterprise? To help me answer those questions, I recently interviewed Rajeev Shah, co-founder and CEO of Celona, a startup that developed a platform for enterprises to deploy private 5G wireless networks. Highlights of my ZKast video with Shah, done in conjunction with eWEEK eSPEAKS, are below:
What is an edgeless enterprise?
- The edgeless enterprise describes the current state of the network where the traditional hard “edge” has been diffused into the cloud, into a worker’s home and other locations.
- By having an artificial definition of “edge,” enterprise networks are siloed into narrow architectures that depend on where the users and applications are located.
- In reality, the edge is wherever the apps need to be, which changes all the time.
- The edgeless enterprise calls for the convergence of network services, apps, cybersecurity tools and wide area network (WAN) optimization on cloud-native, software-defined edge computing platforms deployable anywhere.
- As a result, enterprise apps can be implemented across various on-premises, private and public clouds, depending on computing requirements, eliminating the need for a network edge.
How does 5G fit into the edgeless enterprise?
- First, 5G revolutionized the telecom sector. Now the innovation can be applied in the enterprise to build highly agile networks.
- 5G is cloud-native and highly adaptable to changes in the network. In fact, 5G is the first network architecture that mandated the use of cloud principles.
- While many vendors use the term “cloud networking,” they are referring to cloud management and not cloud-native microservices delivering network capabilities.
- Rather than simply managing networking from the cloud, organizations can tap into 5G’s cloud-native software principles to create a new service layer that automates policies across enterprise networks.
- This service layer enables any network and some security services to be quickly deployed across the enterprise. This is critical in the edgeless enterprise because connections are highly ephemeral today.
- Celona uses the Citizens Broadband Radio Service (CBRS) spectrum in the U.S. to deploy 5G in the enterprise, which can run alongside existing WiFi networks. 5G is ideally suited when maintaining connectivity is critical where WiFi provides a ubiquitous, best-effort network.
What is micro-slicing?
- One cloud-native principle of 5G is multi-tenancy—a single physical infrastructure that can have multiple tenants or slices.
- Celona is bringing slicing into the enterprise with a technology called micro-slicing, which automatically enforces and tracks key service levels—such as latency, jitter and packet error rates—without the IT department’s intervention.
- Micro-slicing provides the assurance that an app is performing as it should over the wireless network, much like it would over a wired network.
Will the edgeless enterprise remain at the edge?
- It’s inevitable for the “brains of the network” to be wherever it needs to be. With apps becoming mobile, the network needs to be wherever the apps are.
- Whether it’s an individual site, such as a store, multi-tenant edge cloud offerings such as AWS or Azure, private data centers or the public cloud, apps could be in any of those places.
- The network needs to be instantly available for apps regardless of their location.
Is cloud-native networking complementary or conflicting to software-defined networking (SDN)?
- Taking control plane and user plane functions and splitting them apart, so there is more flexibility, is the foundation of the 5G cloud-native architecture. This term known as Control-/User Plane Separation (CUPS), is essentially SDN in another form.
- It’s unnecessary to have separate networking hardware, separate security hardware and separate app hardware if single hardware can run all three.
- With the data center revolution, we saw the convergence of computing and storage controlled by software. The next step is the convergence of computing, storage and networking controlled by software in the edgeless enterprise.
Is artificial intelligence for IT operations (AIOps) part of the journey?
- AIOps, which refers to technology that automates IT operations through analytics and machine learning (ML), can react to things that happen, but it cannot proactively influence the network to do things.
- With cloud-native networks, organizations can proactively meet the needs of their apps, as opposed to only reacting to problems.
- AIOps is going through a natural evolution for networking, moving closer toward self-driving networks rather than self-operating networks.
- Today, all the hype is around automation and digitization, although the focus should be on predictable connectivity. People, apps and networks need to be connected everywhere. That’s the ultimate benefit of the edgeless enterprise.