Cisco Tag Switching 18513 Essay Example

offers an effective solution to meet these requirements. Tag switching combines the functionality of a Network Layer router with the performance of an ATM switch. It utilizes the concept of "label swapping," where a label (or tag) is used for forwarding layer 3 packets.

The simplicity of the tag switching forwarding component allows for improved forwarding performance and a competitive price performance ratio. Cisco has chosen to support multiple granularities within the definition of a tag in order to enhance the flexibility of the system. This enables a tag switch to forward data based on various routing functions, including multicast, destination-based routing, hierarchy of routing knowledge, and flexible routing control. By combining all of these routing functions, tag switching provides a routing system that can meet the demand of emerging requirements in corporate and Internet backbones. This article will provide a description and definition of Cisco's Tag Switching, including its components, routing capabilities, ATM switching capabilities, and QoS implementation. Tag Switching is an innovative technique for high-performance packet forwarding that assigns "tags" to multi-protocol frames for transport across packet or cell-based networks.

Based on the concept of "label swapping," this technology uses packets or cells with short, fixed-length labels to guide data processing in switching nodes. There are several observations to consider from the given description of the forwarding component. Firstly, the forwarding decision utilizes an exact match algorithm that employs a fixed length tag as an index. As a result, the forwarding procedure is simplified compared to the longer match forwarding commonly used at the network layer. This simplicity allows for higher forwarding performance, resulting in faster throughput and a greater number of packets per second.

Furthermore, this straightforward forwarding mechanism enables easy hardware implementation.

1 Additionally, it is important to understand that the forwarding decisions made are not dependent on the level of detail for tagging. This means that the same method of forwarding applies to both unicast and multicast traffic. Specifically, a unicast entry would only have one subentry for [outgoing tag, outgoing interface, outgoing link level information]. Conversely, a multicast entry may have one or multiple subentries. This showcases how tag-forwarding criteria can be used for various routing functions in tag switching, such as unicast and multicast. Essentially, the simple forwarding procedure is separate from the control aspect of tag switching.

The addition of new routing functions does not require any changes to the forwarding criterion. This means that there is no need to modify the hardware or software in order to improve forwarding performance when new routing functionality is added.

Control Component

The control component ensures that there is a consistent set of tags across a group of TSR devices by establishing and maintaining the binding between a tag and network layer routing.

Generating a tag involves allocation and binding to a specific destination, such as a host address, address prefix, multicast group address, or other network layer information. Typically, the destination is a TSR. The control component consists of various modules that collectively facilitate different routing functions.

Adding new modules supports new routing functions. The following sections describe some of those modules.

Destination-Based Routing

Destination-based routing for unicast traffic is the most straightforward application of tag switching. In destination-based routing, an address prefix is associated with a forward equivalence class (FEC). Using information from unicast routing

protocols like OSPF, EIGRP, and BGP, a TSR router creates mappings between FECs (address prefixes) and their corresponding next hops.

The Tag Switching control component utilizes the mapping provided to form its Traffic Forwarding Information Base (TFIB). Unlike conventional routers, the TFIB is responsible for the direct forwarding of packets, rather than relying on the Forwarding Equivalence Class (FEC) to next-hop mapping. 6