VxLAN | Part 1 – How VxLAN Works

now the story of on what network with distributed hosts and the one technology has no choice but to keep them all together is VX Lane I think we've all been there sitting at your desk frustrated working on a network design trying to figure out how to get some virtual machines into the same land segment even when there's a whole network between them what you really wish you could do is run a layer tool and segment over the top of your existing network I have good news my friend this is one of the things that VX Lane does it helps simplify the underlying network and reduces the need for spanning tree trunk and stretching VLANs virtual machines can move between hosts on different VLANs all without the need to fiddle with different IP addresses and on top of it all the X LAN is defined in RFC 73 for 8 so it's standards-based there is no need to lock into a particular vendor let me take a moment to briefly show you what we're going to cover in the series in this video we're going is get an overview of the ex LAN and have a look at the basics of how it works in part two we'll dig into the VX line headers and what they do part three introduces the spine leaf topology which often goes hand-in-hand with VX Lane things will get really interesting in part 4 which shows us how addresses are learned if you've heard about evpn this is the video for you and if you want to see how it all fits together then you'll definitely want to see parts five and six where we will configure some nexus 9000 switches now let's get under the hood and see how VX lane works every villain has a VLAN ID this ID is added to a frame to keep traffic unique the VLAN ID is 12 bits long allowing around 4,000 unique VLANs the X LAN is quite similar HVX LAN segment also has an identifier called a VN I this is 24 bits long which gives us around 16 million segments it is enough to give VLAN some sort of inferiority complex just as with VLANs traffic in one vni is kept separate from every other VN I if you want to mix traffic you will need to use a router you may think though 4000 VLANs should be enough for anybody well imagine that you're providing services to customers if you give each customer 8 VLANs you can have only 500 customers before the VLAN pool is exhausted after that you have to get creative and reuse VLAN IDs across different hardware or use something like qinq this is where VX Lane helps with 16 million addresses to choose from you can have unique IDs for your customers across all your data centers or sites vx land creates virtual networks on top of the existing infrastructure this makes it an overlay technology the infrastructure it runs on is called the underlay the underlay network is all layer three all ports in the underlay are routed so there is no need for trucking or spanning tree dynamic routing of some sort is used OSPF and EIGRP are good choices is itis would be fine too this makes good use of ecmp for load sharing and fast recovery you could use BGP in the underlay but this can get quite complicated so I won't get into it here the X Lane itself is the overlay Network each vni is a separate virtual network that runs over the underlay each of these VN i's are called a bridge domain to create this virtual network traffic is encapsulated with UDP and IP before it is sent out when it reaches the destination switch it is d capsulated a hidden advantage to separating the overlay and underlay is that the underlay can be changed without needing to redesign the overlay as long as there is IP connectivity from one end of the underlay to the other the overlay will be fine this leaves us with a question how does encapsulation work switches and routers that participate in VX land have a special interface called a V tip the Vita provides the connection between the overlay and the underlay each V tip has an IP address in the underly network it also has one or more VN eyes to deliver traffic from one host to another a source and destination V tip will create a stateless tunnel these tunnels exist only long enough to deliver the VX Lam frame when a frame for a remote host reaches a switch the frame is encapsulated in IP and UDP headers the switch then forwards a traffic over the underlay but how does the switch know which destination V tap to send traffic to how does even find the other V tips we'll investigate these questions and more in part four of this series the excellen can be supported on hardware or software an advantage of this is that it can run on a hypervisor like ESX or hyper-v this is the host based method the V switch on the host has a V tip which encapsulates traffic from VMs before it touches any physical switches the physical switches just see IP traffic and are unaware of VX lam the advantage of this is a simplified physical network which can simply focus on transport the V tips could also be on physical switches or routers an example of this is configuring VLAN on a nexus switch this is called VX LAN gateway the VM send traffic and the V switches passed the traffic as normal when the traffic arrives on the physical switch the V tap encapsulate s– the frames the advantage is that VX LAN may be implemented in Hardware improving performance not all platforms support VX Lane Hardware so watch out when you're choosing your switches of course a hybrid approach is possible this is where some devices are behind VX Lane gateways and some hosts are VX Lane enabled natively in the second part of this series we're going to dig deeper into the excellent headers and encapsulation before moving on please hit that like button and take a moment to subscribe I'll see you in part 2 you