Hi Mazlan,
Here's a quote from CCO:
"The differences between VPLS and H-VPLS are related to the scaling attributes of each solution. VPLS is a relatively simple architecture that is suitable for small-scale VPLS deployments. It requires that the edge device is MPLS-capable and it needs to participate in routing protocols and Label Distribution Protocol (LDP). This complicates the overall network operation and the edge device, which needs to hold VPLS forwarding tables, routing tables, and others. Additionally, all broadcast and multicast replication is performed at the edge device, which can decrease the efficiency of the network.
By contrast, H-VPLS partitions the network into several edge domains that are interconnected using an MPLS core. The considerations for the edge devices are now simplified, as they need only learn of their local n-PE devices and therefore do not need large routing table support. Alternatively, the edge domain can be built using Ethernet switches and techniques such as VLAN Tag Stacking, aka Q-in-Q, described within the IEEE 802.1ad Provider Bridges draft. Using Ethernet as the edge technology simplifies the operation of the edge domain and reduces the cost of the edge devices dramatically. Cisco was one of the first vendors to realize the scaling limitations imposed by having a nonhierarchical architecture, and developed the concept of a hierarchical VPLS architecture using standard Ethernet bridges at the edge and MPLS within the core."
I would also suggest that you peruse the following documents, which provide more info on H-VPLS:
http://www.cisco.com/en/US/tech/tk436/tk891/technologies_q_and_a_item09186a00801ed3bf.shtml
http://www.cisco.com/en/US/products/ps6603/products_white_paper09186a00801ed506.shtml
Hope that helps - pls rate the post if it does.
Paresh
