VPLS (Virtual Private LAN Service) and h-VPLS (Hierarhical VPLS) has some common and different properties. So they have different aadvantages and disadvantages. In this article, we will focus on the comparison of these two concepts and we will see Hierarhical VPLS versus VPLS. The most important different between VPLS and h-VPLS is in the interface between the CE (Customer Edge) and the PE (Provider Edge). Until know, when we talk about Provider Edge devices of an MPLS network, we always use PE abbstraction. With the concept of h-VPLS, instead of one PE device, there are two PE device. u-PE (User facing Provider Edge) and n-PE (Network facing Provider Edge).
In VPLS (Virtual Private LAN Service), the CE devices are directly connected to the MPLS Core n-PE (Network facing Provider Edge) devices. In other words, CE device connects directly to an MPLS-enabled device. This device will be participate in routing (routing protocols) and label distribution (LDP).
In this type of VPLS, the Edge Device will perform many heavy facilities. For example it needs to hold routing tables, forwarding tables etc. It does many jobs as an only PE device. This is not an efficient solution for large scale networks.
In h-VPLS, the disadvantage of normal VPLS is bypassed with the usage of additional edge points . This point is called u-PE (User face of Provider Edge). The customer traffic is meet on the u-PE device. And this first point do not need to hold whole routing tables. It needs only the address of its peer n-PE device. This u-PE device can be a ethernet switch. And it transfers the different trunk traffic to the n-PE device. u-PE device uses Q-in-Q tagging for this.
For large scale networks, h-VPLS is a good and scalable choice.
h-VPLS (Hierarchical VPLS) has many advantages if you compare with flat VPLS. h-VPLS has also some disadvantages. Lets check the advantages of h-VPLS.