MPLS Labels, Device Roles & LDP Protocol Explained Simply

Hello! This is Lazarus at telecom.Tech, where  telecom and networking technologies... are   simply explained! Today, we'll be simply  explaining MPLS labels, MPLS device roles,   and the Label Distribution Protocol. Now whether  you're new to MPLS or you just want to brush up on  

your understanding of the technology, this video  series is just for you! You can be notified of new   content within this MPLS video series by simply  subscribing and enabling notifications. If you do   so, you're all set, and if you find this video  useful, please do click that thumbs-up button.  

Now in the last video, we introduced MPLS as a  technology that is used to efficiently direct data   between nodes, and it's a technology that you'll  most often find within the core network of an ISP,   or other large geographically dispersed networks.  Now this here is an example of an MPLS core  

network connected to several customer sites. Now  the first thing we'll look at is what we call the   Customer Edge router or the CE router. These here  are all CE routers. These are the routers at the   edge of the customer network. They are equipment  provided typically by the customer at their own  

sites, and they connect to the MPLS network. Now  behind each CE router, we have the rest of the   customer network at that particular site. Now  these routers are completely oblivious to any   MPLS configuration. They don't know anything  about MPLS, but they, in turn, are connected  

to what we often call Provider Edge routers,  or PE routers. Now Provider Edge routers are   those routers that exist on the edge of the MPLS  infrastructure. These are the devices that connect   to the customer premises equipment which are the  CE routers. So by definition, PE routers connect  

to one or more CE routers. Now the term "Provider  Edge" or PE is somewhat of an unofficial name, but   it is used very often. The official name of these  routers within the framework of MPLS is Label Edge   Router or LER. Now these PEs or LERs have MPLS  configuration on them. Upon ingress of a packet  

into the MPLS infrastructure, they are the ones  that receive the packets from the CE and assign   a label to them based on the configuration that  has been applied for that particular customer.   Similarly upon egress of the packet from the MPLS  network, the PEs are the devices that forward  

a packet to the correct customer after having  removed the MPLS label. Now the PE and CE devices   will typically be configured as routing peers  using some dynamic routing protocol. The CE router   will inform the PE router of all of its own routes  within the customer network using either EIGRP or  

OSPF or BGP or even static routing, although  that's not typically recommended, but it can   be done. And finally we have what are often called  the P routers, or the Provider routers. These here   are the P routers within the MPLS topology. These  are essentially the devices that are not on the  

edge of the MPLS infrastructure, but they're part  of the core network interconnecting PE devices and   other P routers as well. Now like the PE device,  the P designation is also somewhat unofficial,   but you will see it written in this way in many  MPLS diagrams and documentation. The official  

name of this device in the MPLS framework  is Label Switch Router or LSR. So the P   routers or the LSRs are there to switch packets  based on their labels within the MPLS core. Now   here's an overall view with all of the routers  labeled in the role that they play within MPLS. 

Remember, CE routers are oblivious to MPLS  configurations and the CE and PE routers simply   have a regular routing protocol peering between  them, while PE and P routers are the only ones   that have any knowledge of MPLS infrastructure  and configurations. Now let's take a look at MPLS  

encapsulation in a little more detail. Here we see  the data going from the CE device to the PE device   where a label is added. This label is added in the  MPLS header which is inserted between the Layer 2   frame header and the Layer 3 IP header. Remember  MPLS is often called a Layer 2.5 protocol for  

this reason, because it fits in between these  two headers. Within the MPLS header, we see   various fields. For now we'll focus on the label  value which is a 20-bit numerical value that's   assigned to the particular packet. This label is  what is used throughout MPLS to direct the packet  

to its appropriate destination. Now if we look at  the MPLS forwarding table within the PE router,   we see something like this. This is the command we  would use in a Cisco IOS router to see this table.   Now don't worry about the details of this output  for now, just focus on this first entry here. In  

order to get to this destination, we are using a  label of 27 and we have our destination address   here. The exit interface and the next hop IP  that correspond with this particular label   are found here. So we know that this is the next  hop we need to send our packet to. Next when it  

arrives on the P router, we see this entry here.  You can see that the label is being switched to   label 28. The appropriate exit interface and  next hop IP are also indicated here, so that   packet can be forwarded appropriately on the next  P router. We see the process continues from node  

to node until it gets to the final PE router that  connects to the remote customer site. Now when   the P router receives something that is tagged  with label 28 with a destination of 4.4.4.4,   it's outgoing label says "Pop Label" which means  remove the label. The label is actually removed  

by the last P router in the path to our intended  destination. Now once this packet reaches the PE   router here, it no longer has an MPLS label,  and it's just normal routing that takes place.   Remember PE and CE routers have a regular IP  routing peering so it is only IP routing that's  

used to get that data to its intended destination  over that last link. So that's a brief, somewhat   simplistic overview of this whole process of MPLS  label switching, and the roles that the CE, PE   and P routers play in the process. Now these MPLS  forwarding tables are entities that are much more  

efficient to be searched in order to determine  in the next hop rather than searching through   hundreds or thousands of entries within a routing  table. Now one last question that may come to mind   is this how are these labels determined and how  is the MPLS forwarding table populated? That's  

done using the Label Distribution Protocol or LDP.  Now LDP is a protocol that automatically creates   LDP neighbor adjacencies between all MPLS-enabled  interfaces on PE and P routers. Adjacencies are   created and then neighbors can begin to exchange  information about labels. Each router locally  

generates labels for its prefixes and then it'll  advertises those labels to its LDP neighbors. Now   that's how each MPLS router populates its MPLS  forwarding table. Now the details of this actual   operation of LDP with the various control  plane and forwarding plane entities that are  

used in this process, we'll leave for another  video, but for now suffice it to say that LDP   is responsible for ensuring that those forwarding  tables are populated with the appropriate labels.   So that's the gist of MPLS labels, device roles,  and a brief introduction to the LDP protocol. So  

far so good I hope! For now, I hope you found this  video useful, and if you have, please click that   thumbs-up button! And please subscribe to get  updates to newly published videos I'm Lazarus   at telecom.Tech, I hope this has been helpful  for you, and I'd like to thank you for watching!