MPLS TE (Traffic Engineering) : Introduction and Objectives

3.1 Introduction

Simply put, traffic engineering (TE) is about manipulating your traffic to fit your network resources. Many of the MPLS benefits discussed above relate to the traffic engineering aspect of MPLS, but TE is by no means an MPLS specific thing. TE can be performed by simply tweaking the IP metrics on interfaces, which of course is unsuitable for huge networks.

Although, IP-over-ATM model does provide traffic engineering through the use of explicitly routed private virtual channels (PVCs), but the IP-over-ATM model itself throws up problems like: mapping between two distinct architectures that require the definition and maintenance of separate topologies, address spaces, routing protocols, signaling protocols, and resource allocation schemes. MPLS TE alleviates many of these problems, and attempts to bring in the best from the connection-oriented TE techniques and merge them with IP routing.

3.2 MPLS TE (Traffic Engineering) Objectives

More formally speaking, the key traffic engineering objectives are:

1.Minimizing congestion:

Congestion occurs either when network resources are insufficient or inadequate to accommodate offered load or if traffic streams are inefficiently mapped onto available resources; causing subsets of network resources to become over-utilized while others remain underutilized. [2]

2.Reliable network operations:

Adequate capacity for service restoration must be available keeping in mind multiple failure scenarios, and at the same time, there must be mechanisms to efficiently and speedily reroute traffic through the redundant capacity. On recovering from the faults, reoptimization may be necessary to include the restored capacity.

3.Quality of Service requirements:

In a multiclass service environment, where traffic streams with different service requirements contend with each other, the role of traffic engineering becomes more decisive. In such scenarios, traffic engineering has to provision resources selectively for various classes of streams, judiciously sharing the network resources, giving preferential treatment to some service classes.