Ok, let’s start with what the Autonomous Enterprise is – it is really a continuation of the digital transformation process we have all been experiencing over the last several years. At its foundation, the autonomous enterprise is really an architecture that leverages machine learning, artificial intelligence, policy-based decision making, and automation to drive workflows that enable self-configuration, self-repair, and automatic mitigation of network issues, whether performance or security related.
To be honest, all major and some not-so-major, enterprise networking vendors either have a solution, components of a solution, or at least some messaging around the autonomous enterprise as it does seem to be the topic of the day.
I don’t mean to sound flip as the autonomous enterprise is an important step in the evolution of enterprise networking, however when companies post information to their web pages stating essentially, ”We will tell you more about our solution when we actually figure it out all its components,” it is a clear indication that they not only don’t have a solution but are not even really sure what the architecture will look like (name withheld for my own safety).
The concept of the autonomous enterprise comes out of something called Intent Based Networking (IBN), it has actually been around for several years and there are plenty of research and informational materials on this topic, although not everyone agrees on its scope, Tech Target has a really good paper to get you started.
Additionally, vendors use different terms to describe their solution, you will see Autonomous Enterprise, Autonomous Networking, Intent-Based Networking, Experience-Driven Networking, Self-Driving Networking and all will define it using terms like Machine Learning, Artificial Intelligence, Cognitive Thinking, Proactive, Automatic, ok you get the idea.
One of the biggest challenges for many networking vendors is to provide a solution that is not siloed by solution sets, like Data Center, Campus Fabric, Edge Switches, WLAN, Cloud, SD-WAN, etc. All major enterprise networking vendors have good-to-great hardware solutions and management solutions and even a good autonomous enterprise message in one or more areas of the network, but for most cases, there is not a single framework that ties them together. In this case, you may see them pitching their message in only one area.
Next area to watch out for is vendors who try and lock you into their proprietary solution, the message focuses on the automation without mentioning all hardware has to be theirs, all software has to be theirs, and there are a few other optional/mandatory software packages that are required to make it all ‘seamless.’ Well, most networks aren’t built that way so make sure to look for terms like multi-vendor, open-ecosystem, standards-based, and open standards. Then make them demonstrate it.
There is also the question of scalability, and not just how many devices, users, and traffic loads, but also ask about horizontal scalability, across campus, and out to branches. My final point here is to ask about how difficult it is to scale? If I need to add another platform, is it plug & play, do I need to bring the SE team in for a week or is it somewhere in between?
Don’t forget we are typically focused on the enterprise, some vendors may attempt to demonstrate how successful they were deploying in a service provider network or a carrier network, both operate very differently than your typical enterprise. Ask for examples in your industry.
An industry trend that networking vendors speak to in different ways. All versions are based on or identified to leverage cognitive thinking, machine learning, and artificial intelligence capabilities to proactively detect and remediate network & security events.
Solution Name Intent-Based Networking Experience-Driven Networking Self-Driving Network
The Messages Intent-Based Networking captures business intent and uses analytics, machine learning, and automation to align the network continuously to changing business needs. Experience-Driven service management platform, 5 core feature sets; planning & deployment, analytics, performance, security & extensibility, to automate and optimize the delivery of network services. Self-Driving Network, similar to a self-driving car, is an autonomous network that is predictive and adaptive to its environment. Supports multi-vendor environment.
The Limitations Multiple proprietary automation & orchestration solutions (data center, campus, SD-WAN, Cloud, etc.) that do not interwork as a single system. System-wide modifications still require multiple touches of components. Yet to clearly define their architecture or solution. Current web message - developing features and collateral to tell story for our management and operations solutions... As the content is created we'll expand on what falls into each category to solve use cases in a holistic manner. Service provider focused, trying to leverage architecture to enter enterprise market. No integrated edge to data center solution, another solution comprised of multiple silos with no single orchestration/
Extreme’s software-driven network infrastructure is a foundational layer for enabling a secure, scalable, and efficient use of cloud, edge, and IoT applications. We offer pervasive real-time visibility from our cloud to your to deliver insights and security across a broad range of infrastructure and cloud environments. With our policy-driven closed-loop automation, the network can detect anomalous behavior from IoT devices, users, or applications, and automatically resolve issues without intervention. Unlike other vendors with proprietary solutions, Extreme offers multi-vendor support in an open and standards-based ecosystem.
Now as I said, everyone has a message, we do too. Extreme has over 30 years building enterprise networks, starting in the early thicknet days of Cabletron. Yes, I was one of those guys. Through the switching revolution, to wireless, and then all areas of the network. Over the years it was always about connecting people to resources and applications via a combination of hardware, software, services, and applications. Providing a network able to automatically detect errors and correct them, baseline client behavior, and implement policy control process if an anomaly is detected. Essentially a network that understands what the business-critical applications and experiences are, and then automatically optimizes the system to support this level of service.
This blog was originally authored by Perry Correll, Director, Product Marketing.