Category: data center

I’ve already written about the stupidities of risking the stability of two data centers to enable live migration of “mission critical” VMs between them. Now let’s take the discussion a step further – after hearing how critical the VM the server or application team wants to migrate is, you might be tempted to ask “and how do you ensure its high availability the rest of the time?” The response will likely be along the lines of “We’re using VMware High Availability” or even prouder “We’re using VMware Fault Tolerance to ensure even a hardware failure can’t bring it down.”

It seems that most networking vendors consider the Flat Earth architectures the new bonanza. Everyone is running to join the gold rush, from Cisco’s FabricPath and Brocade’s VCS to HP’s IRF and Juniper’s upcoming QFabric. As always, the standardization bodies are following the industry with a large buffet of standards to choose from: TRILL, 802.1ag (SPB), 802.1Qbg (EVB) and 802.1bh (Port extenders).

The following design challenge landed in my Inbox not too long ago:

My organization is the in the process of building a completely new data center from the ground up (new hardware, software, protocols ...). We will currently start with one site but may move to two for DR purposes. What DC technologies should we be looking at implementing to build a stable infrastructure that will scale and support technologies you feel will play a big role in the future?

In an ideal world, my answer would begin with “Start with the applications.”

I was sort of upset that my vacations were making me miss the VMware vSphere 5.0 launch event (on the other hand, being limited to half hour Internet access served with early morning cappuccino is not necessarily a bad thing), but after I managed to get home, I realized I hadn’t really missed much. Let me rephrase that – VMware launched a major release of vSphere and the networking features are barely worth mentioning (or maybe they’ll launch them when the vTax brouhaha subsides).

A comment left on my dense-mode FCoE post is a perfect example of the dangers of using vague, marketing-driven and ill-defined word like “switching”. The author wrote: “FC-SW is by no means routing… Fibre Channel is switching.” As I explained in one of my previous posts, switching can mean anything, from circuit-based activities to bridging, routing and even load balancing (I am positive some vendors claim their load balancers application delivery controllers are L4-L7 switches), so let’s see whether Fibre Channel “switching” is closer to bridging or routing.

Yandy sent me an interesting question:

Is it just me or do you also see the Nexus 2000 series not having any type of distributed forwarding as a major design flaw? Cisco keeps throwing in the “it's a line-card” line, but any dumb modular switch nowadays has distributed forwarding in all its line cards.

I’m at least as annoyed as Yandy is by the lack of distributed switching in the Nexus port (oops, fabric) extender product range, but let’s focus on a different question: does it matter?

I had to check the Microsoft clustering terminology a few days ago, so I used Google to find the most relevant pages for “Windows cluster” and landed on the Failover clustering home page where the Multisite Clustering link immediately caught my attention. Dreading the humongous amount of layer-2 DCI stupidities that could lurk hidden behind such a concept, I barely dared to click on the link… which unveiled one of the most pleasant surprises I’ve got from an IT vendor in a very long time.

J Michel Metz brought out an interesting aspect of the dense/sparse mode FCoE design dilemma in a comment to my FCoE over Trill ... this time from Juniper post: FC-focused troubleshooting. I have to mention that he happens to be working for a company that has the only dense-mode FCoE solution, but the comment does stand on its own.

Before reading this post you might want to read the definition of dense- and sparse-mode FCoE and a few more technical details.

You probably know the old saying – if the mountain doesn’t want to come to you, you have to go out there and climb it. vCider, a brand-new startup launching their product at Gigaom Structure Launchpad, decided to do something similar in the server virtualization (Infrastructure-as-a-Service; IaaS) space – its software allows IaaS customers to build their own virtual layer-2 networks (let’s call then vSubnets) on top of IaaS provider’s IP infrastructure; you can even build a vSubnets between VMs running within your enterprise network (private cloud in the cloudy lingo) and those running within Amazon EC2 or Rackspace.

Full disclosure: Chris Marino from vCider got in touch with me in early June. I found the idea interesting, he helped me understand their product (even offered a test run, but I chose to trust the technical information available on their web site and passed to me in e-mails and phone calls), and I decided to write about it. That’s it.

One of the implications of Virtual Machine (VM) mobility (as implemented by VMware’s vMotion or Microsoft’s Live Migration) is the need to have the same VLAN configured on the access ports connected to the source and the target hypervisor hosts. EVB (802.1Qbg) provides a perfect solution, but it’s questionable when it will leave the dreamland domain. In the meantime, most environments have to deploy stretched VLANs ... or you might be able to use hypervisor-aware features of your edge switches, for example VM Tracer implemented in Arista EOS.

A tweet from J Michel Metz has alerted me to a “Why TRILL won't work for data center network architecture” article by Anjan Venkatramani, Juniper’s VP of Product Management. Most of the long article could be condensed in two short sentences my readers are very familiar about: Bridging does not scale and TRILL does not solve the traffic trombone issues (hidden implication: QFabric will solve all your problems)... but the author couldn’t resist throwing “FCoE over TRILL” bone into the mix.

Some people are changing round wheels to heptagonal format because they will roll better. Some other people are building stretched high-availability clusters – clusters of servers stretched over multiple data centers. Unfortunately only one of these claims is false.

Similar to the stretched firewalls design, stretched tightly coupled HA clusters are vulnerable – you lose the inter-DC link for long enough time (depending on how the cluster heartbeat is configured, a few seconds could be enough) and you have a total disaster on your hands.

A few years ago Cisco introduced an interesting concept to the data center networking: fabric extenders, devices acting like remote linecards of a central switch (Juniper’s “revolutionary” QFabric looks very similar from a distance; the only major difference seems to be local switching in the QF/Nodes). Cisco’s proprietary technology used in its FEX products became the basis for 802.1Qbh, an IEEE draft that is supposed to standardize the port extender architecture.

If you’re not familiar with the FEX products, read my “Port or Fabric Extenders?” article before continuing ... and disregard most of what it says about 802.1Qbh.

Three months after the QFabric launch, the details remain shrouded in mystical clouds, so let’s try to speculate what they could be hiding. We have two well-known facts:

• QFabric has three components: QF/Node (edge device), QF/Interconnect (high-speed core device) and QF/Director (the brains).
• Juniper is strong in the Service Provider technologies, including MPLS, MPLS/VPN, VPLS and BGP. It’s also touting its BGP MPLS-based MAC VPN technology (too long to write more than once, let’s call it BMMV).

I am positive Juniper would never try to build a monster single-brain fabric with Borg or Big Brother architecture as they simply don’t scale (as the OpenFlow crowd will learn in a few years).

The Edge Virtual Bridging (EVB; 802.1Qbg) standard solves two important layer-2-based virtualization issues:

• Automatic provisioning of access switches based on hypervisor-signaled information (discussed in the EVB eases VLAN configuration pains article)
• Multiplexing of multiple logical 802.1Q links over a single physical link.

Logical link multiplexing might seem a solution in search of a problem until you discover that VMware-related design documents usually recommend using 6 to 10 NICs per server – an approach that either wastes switch ports or is hard to implement with blade servers’ mezzanine cards (due to limited number of backplane connections).