 Now imagine multiple customer networks are connected to some provider networks. Now these provider networks may also be connected to other provider networks. You can take an analogy from the customer premises equipment that is connected to the access network and that access network is connected to the core network. Now since the core network is basically a very large network so the traffic within the core is carried on a backbone and from that backbone the traffic is relayed from one access network to another access network. So here also we are going to extend and build on that concept through the concept of provider backbone bridges. These provider backbone bridges essentially are used to provide connectivity between provider networks. Since we are talking about provider networks that is source provider network and destination provider network and since we have already talked about utilizing s tag in addition to the vlan id or the or the c tag here we are somehow forced to think about some kind of addressing of every network. For that the source and destination MAC addresses are now used which identify the provider networks. So it means that in order to enter the ethernet based traffic from one provider backbone bridge to another provider backbone bridge we actually are interested to use MAC addressing to transfer from one provider's network to another provider network. It means now the backbone is going to be the connectivity between two provider networks each identified with their own MAC address. Now the operation within the provider backbone network is that the switches which belong to the provider backbone network look at the MAC addresses and these MAC addresses are then referred to the forwarding table. The forwarding table how it is developed in a soft way at a particular switch is something that we are already aware of. This provider backbone network uses an additional tag in addition to the c tag and the s tag that we call the i tag. The i tag has as such a length of 24 bits. Now you may wonder why it is given a length of 24 bits. There's a very interesting reason to it when we talk about intra provider network traffic or intra customer network traffic we are talking about a limited number of nodes but when we talk about the provider networks there could be numerous or uncountable provider networks that the backbone is supposed to provide connectivity to. So that is why in order to keep it scalable in order to provide backbone services to multiple provider networks the length has been kept double to what 12 bits we have seen before. The provider backbone bridges use this i tag which separates the traffic as it enters from one provider network to another and the only thing that modifies is the source and destination MAC addresses. So we have the i tag we have the source and destination MAC addresses. The only thing that is of interest is and it's very very relevant that is when we talk about the provider backbone network there is an entry and exit point to the to the backbone. So it means that the intermediate switches within that backbone are not necessarily required to be provider backbone enabled. They just have to behave like provider bridges because if we are talking about a backbone connecting two endpoints then we are only interested in the ingress and egress behavior of the provider backbone bridge. Let's look at this figure once again. Here we see we have multiple customer networks. These customer networks are connected to each other through a provider network. Now the provider networks are connected to each other via provider backbone bridge network which uses additional information like the i tag and it also has a b tag that is the backbone tag. Now this backbone tag identifies the backbone as such and we see that now it is using the source and destination MAC addresses. The overall header now includes two MAC addressings that is source and destination MAC addressing both for the customer as well as for identifying the backbone endpoints. Then we have the tags the c tag s tag and the i and b tag.