Network Fabrics: 5 Common Misconceptions Dispelled

A clarifying review of fabric technology and its use cases

istock 544672688

What’s old is new again. This statement rings true as the industry rallies behind fabric networking again. Fabrics are not a new technology; they’ve been around since 2011/2012. However, we’ve seen increased uptake for fabrics outside of the traditional data center and into the enterprise campus. As the use cases and the technology continue to evolve, this blog clarifies some common misconceptions you may have about fabric technology.

1. They are only for the data center

Not anymore. Going back 5-6 years, network fabrics were originally designed to solve how to stretch L2 VLANs across subnets for VM migrations and mobility. We now see far more use cases for network fabrics that extend to the enterprise campus portion of the network. Examples include network automation, zero-touch provisioning, simplified network segmentation, and even high-performance multicast without the use of any complex PIM protocols.

2. You need to hire a network programmer  

Not necessarily. What we are seeing in the industry is that there are different types of network fabrics that appeal to different types of users. Some users value plug-and-play simplicity because they have a lean IT staff and are looking for a solution that is easy to deploy and manage. Both Extreme’s VCS Fabric for the data center and Fabric Connect for the campus are well positioned for these types of deployments.

In another case, some users value openness and programmability and are looking for a data center fabric that they can customize and integrate easily with their ecosystem tools and cloud offerings. Extreme’s IP Fabric is well suited for this type of user. 

3. It’s based solely on network overlays

Not always! Again, it really depends on the fabric technology. It might be fair to say that the vast majority of campus fabric solutions are overlay, but there is at least one fabric in the industry - Extreme Fabric Connect – that actually streamlines the network and reduces the number of network protocols down to one. This contributes to the plug-and-play simplicity that is becoming increasingly important as complexity in the campus increases due to IoT, mobility, real-time demands, and the ever-growing threat of a cyberattack.

4. They are only for wired devices

Not with Extreme! Extreme enables true unification of wired and wireless access through our unique fabric-based, policy-enabled architecture that extends from core to wired/wireless access. This provides a uniquely simple, easy-to-manage consistent network that not only enables plug-and-play deployment of the APs, but also dynamic auto-attach of users and IoT devices into the appropriate fabric-enabled service—regardless of their connection point.

5. They are either L2 or L3 centric

Not necessarily! Extreme is delivering fabric technologies for both the enterprise campus as well as the data center that support full L2/3 functionality. In fact, Fabric Connect, which is based on an enhanced form of Shortest Path Bridging, is one of the few protocols to be jointly standardized by both the IEEE and IETF. In addition to providing scalable L2 services, it also provides integrated VRF technology and integrated, scalable, high-performance IP multicast.

I’ve only briefly touched on the some of the misconceptions about network fabrics here, but if you'd like to learn more, watch the on-demand webinar, Demystifying Fabric Networking. You’ll hear from Paul Unbehagen, Extreme’s Senior Director of Product and Strategy, as he discusses fabrics in-depth and shares more details about Fabric Connect specifically. As an active member of the IEEE and IETF and co-author of the IEEE 802.1aq standard, Paul will be able to clarify any misconception about fabric networking.

Camille Campbell is a Product Marketing Manager at Extreme Networks.