**1. Introduction**

*Multi Protocol Label Switching (MPLS) is today mostly used for traffic engineering therefore we start by describing what traffic engineering is and why traffic engineering is needed.* 

Traffic engineering and fast reroute are the two major applications of constraint based routing Traffic engineering is the process of controlling how traffic flows through a service provider's network so as to optimize resource utilization and network performance[1]. Traffic engineering is needed in the Internet mainly because the shortest path is used in current intra- domain routing protocols (e.g., OSPF, IS-IS) to forward traffic. The shortest path routing may give rise to two problems.

**First**, the shortest paths from different sources overlap at some links, resulting in congestion at those links.

**Second**, at some time, the traffic volume from a source to a destination could exceed the capacity of the shortest path, while a longer path between these two nodes remains underutilized. The reason why conventional IP routing cannot provide traffic engineering is that it does not take into account the available bandwidth on individual links. For the purpose of traffic engineering, constraint based routing is used to route traffic trunk[2], which is defined as a collection of individual transmission control protocol (TCP), or user datagram protocol (UDP) flows, called "microflows" that share two common properties.

The **first** property is that all microflows are forwarded along the same common path.

The **second** property is that they all share the same class of service. By routing at the granularity of traffic trunks, traffic trunks have better scaling properties than routing at the granularity of individual microflows with respect to the amount of forwarding state and the volume of control traffic.

In a sense, IP networks manage themselves. A host using the Transmission Control Protocol (TCP) adjusts its sending rate according to the available bandwidth on the path to the receiver. If the network topology should change, routers react to changes and calculate new paths to the destination. This has made the TCP/IP [3] Internet a robust communication network. But robustness does not implicate that the network runs efficiently. The interior gateway protocols used today like OSPF and ISIS compute the shortest way to the destination and routers forward traffic according to the routing tables build from those calculations. This means that traffic from different sources passing through a router with the same destination will be

Traffic Engineering 249

In most cases, the calculation of the optimum routing of virtual circuits was done off-line by a network management platform; advanced networks (offering Frame Relay or ATM switched virtual circuits) also offered real-time on-demand establishment of virtual circuits.

The end-to-end hop-by-hop path throughout the network that satisfied the contractual

Internet and most IP-based services, including IP-based virtual private networks (VPNs) implemented with MPLS VPN, IPsec or Layer 2 transport protocol (L2TP), follow a

The traffic contract specifies ingress and egress bandwidth for each site, not site-to-site

Every IP packet is routed through the network independently, and every router in the

Once merged, all packets toward the same destination take the same path (whereas

 Layer 2 switched networks assume that the bandwidth is expensive and try to optimize its usage, resulting in complex circuit setup mechanisms and expensive switching

IP networks assume that the bandwidth is "free" and focus on low-cost, high-speed

The significant difference between the cost-per-switched-megabit of Layer 2 network (for example, ATM) and routed (IP) network has forced nearly all service providers to build next-generation networks exclusively on IP. Even in modern fiber-optics networks, however, bandwidth is not totally free, and there are always scenarios where you could use free resources of an underutilized link to ease the pressure on an overloaded path. Effectively, you would need traffic engineering capabilities in routed IP networks, but they are simply not available in the traditional hop-by-hop, destination-only routing model that

Various approaches (including creative designs, as well as new technologies) have been tried to bring the traffic engineering capabilities to IP-based networks. We can group them

 The network core uses Layer 2 switched technology (ATM or Frame Relay) that has inherent traffic engineering capabilities. Virtual circuits are then established between

IP routing tricks are used to modify the operation of IP routing protocols, resulting in

Deployment of IP-based virtual circuit technologies, including IP-over-IP tunnels and

adjustments to the path the packets are taking through the network.

multiple virtual circuits toward the same site could traverse different links).

Simplified to the extreme, the two paradigms could be expressed as follows:

requirements (and, if needed, met other criteria) was computed.

A virtual circuit was established along the computed path.

path makes independent next-hop decisions.

switching of a high volume of traffic.

However, the process was always the same: The free network capacity was examined.

completely different service model:

traffic requirements.

methods.

most IP networks use.

roughly into these categories:

edge routers as needed.

MPLS traffic engineering.

aggregated and sent through the same path. Therefore a link may be congested despite the presence of under-utilized link in the network. And delay sensitive traffic like voice-over-IP calls may travel over a path with high propagation delay because this is the shortest path while a low latency path is available.

Fig. 1. Traffic Engineering

As illustrated in the above figure 1 the shortest path from router 1 to 5 is the path (1-3-5). All traffic passing through router 1 with destination router 5 (or another router with router 5 in the shortest path) will travel through this shortest path if the shortest path algorithm is used for forwarding in this network. Although there is an alternative path (1-2-4-5) available that could be used to distribute traffic more evenly in the network.

Traffic engineering is the process of controlling how traffic flows through a network to optimize resource utilization and network performance [4]. Traffic engineering is basically concerned with two problems that occur from routing protocols that only use the shortest path as constraint when they construct a routing table.

The shortest paths from different sources overlap at some links, causing congestion on those links. The traffic from a source to a destination exceeds the capacity of the shortest path, while a longer path between these two routers is under-utilized.

MPLS can be used as a traffic engineering tool to direct traffic in a network in a more efficient way then original IP shortest path routing. MPLS can be used to control which paths traffic travels through the network and therefore a more efficient use of the network resources can be achieved. Paths in the network can be reserved for traffic that is sensitive, and links and router that is more secure and not known to fail can be used for this kind of traffic.
