Category: Design

Two weeks ago we discussed whether it makes sense to use BGP as the routing protocol in a data center fabric. Today we’ll tackle three additional design challenges:

• Should you use IBGP or EBGP?
• When should you run BGP on the spine switches?
• Should every leaf switch have a different AS number or should they share the same AS number?

A while ago I helped a large enterprise redesign their data center fabric. They did a wonderful job optimizing their infrastructure, so all they really needed were two switches in each location.

Some vendors couldn’t fathom that. One of them proposed to build a “future-proof” (and twice as expensive) leaf-and-spine fabric with two leaves and two spines. On top of that they proposed to use EBGP as the only routing protocol because draft-lapukhov-bgp-routing-large-dc – a clear case of missing the customer needs.

One of my friends wanted to design a nice-and-easy layer-3 leaf-and-spine fabric for a new data center, and got blindsided by a hyperconverged vendor. Here’s what he wrote:

We wanted to have a spine/leaf L3 topology for an NSX deployment but can’t do that because the Nutanix servers require L2 between their nodes so they can be in the same cluster.

I wanted to check his claims, but Nutanix doesn’t publish their documentation (I would consider that a red flag), so I’m assuming he’s right until someone proves otherwise (note: whitepaper is not a proof of anything ;).

I published the last part of my Optimize Data Center Infrastructure series: build an optimized data center fabric.

To learn more about data center fabric designs, check the new online course or enroll into the Spring 2018 session of Building Next-Generation Data Center course.

One of my readers sent me this question:

I'm having an internal debate whether to use firewall-based VPNs or DMVPN to connect several sites if our MPLS connection goes down. How would you handle it? Do you have specific courses answering this question?

As always, the correct answer is it depends, in this case on:

Everyone knows that Service Providers and Enterprise networks diverged decades ago. More precisely, organizations that offer network connectivity as their core business usually (but not always) behave differently from organizations that use networking to support their core business.

Obviously, there are grey areas: from people claiming to be service providers who can’t get their act together, to departments (or whole organizations) who run enterprise networks that look a lot like traditional service provider networks because they’re effectively an internal service provider.

One of the design scenarios we covered in Leaf-and-Spine Fabric Architectures webinar is a pure layer-3 data center, and in the “how do I do this” part of that section Dinesh Dutt talked about the details you need to know to get this idea implemented on Cumulus Linux.

We covered a half-dozen design scenarios in that webinar; for an even wider picture check out the new Designing and Building Data Center Fabrics online course.

Sometimes I have this weird feeling that I’m the only loony in town desperately preaching against the stupidities heaped upon infrastructure, so it’s really nice when I find a fellow lost soul. This is what another senior networking engineer sent me:

I'm belonging to a small group of people who are thinking that the source of the problem are the apps and the associated business/security rules: their nature, their complexity, their lifecycle...

Sounds familiar (I probably wrote a few blog posts on this topic in the past), and it only got better.

In June 2017, we concluded the Building Next Generation Data Center online course with a roundtable discussion with Andrew Lerner, Research Vice President, Networking, and Simon Richard, Research Director, Data Center Networking @ Gartner.

In the second half of our discussion (first half is here) we focused on these topics:

One of my friends sent me this design challenge:

Assume you’re migrating from another WAN transport technology to MPLS. The existing network has 3000 routes but the MPLS carrier is limiting you to 1000 routes. How could you solve this with MPLS?

Personally, I think MPLS is a red herring.