LSP and Back up Path Setup in MPLS Networks
Based on Path Criticality Index
Ali Tizghadam, Alberto Leon-Garcia
Dept. of Electrical and Computer Engineering
University of Toronto
40 St. George St., Toronto, ON, M5S 2E4, Canada
{ali.tizghadam, alberto.leongarcia}@utoronto.ca
Abstract— This paper reports on a promising approach for
solving problems found when Multi Protocol Label Switching
(MPLS), soon to be a dominant protocol, is used in core network
systems. Difficulty is found largely in LSP routing and traffic
engineering approaches. While there are a number of online and
offline proposals to establish the LSPs but no one is a complete
solution considering all the aspects of routing plan from traffic
engineering point of view. Our research takes a viewpoint
inspired by the concept of “between-ness” from graph theory,
from which we introduce notions of link and path criticality
indexes. The basis of the work is finding the most critical paths
which are mathematically defined based on the algebra of
routing. We try to avoid running aggregated flows or
commodities on the most critical paths for the short term, and
plan increasing the bandwidth of the critical paths for future if
possible. This approach shows promise in simulations have run
on benchmark networks available from research literature.
Keywords- Flow Assignment; Graph Theory; MPLS; Quality of
Service (QoS); Routing; Traffic Engineering
I. INTRODUCTION
The infrastructure of the traditional service provider is
undergoing a fundamental transition from a telephone-service-
focused circuit-switching architecture to a multi-service packet-
switching architecture based on Internet Protocol (IP) and
enhancements that enable Quality of Service and traffic
engineering. IP transport networks that can transfer packets
according to differentiated levels of QoS, availability and price
are a key element to generating revenue through a rich offering
of services and applications.
In this shared infrastructure MPLS has a key role. In
addition to providing a transparent intermediate layer to hide
the complexity of the different data layer technologies from the
higher layers, MPLS provides a flexible routing mechanism by
assigning traffic flows to the end-to-end label switched paths
(LSP). MPLS can be used to engineer the traffic among
different paths of the network by intelligent matching of the
path capacities with flows to avoid network congestion.
An abundance of work has been done in the research
community and industry to address the routing and flow
assignment problem and traffic engineering issues in MPLS
networks [1], [2], [3]. The main goal of the research reported in
this paper is to look at the problem from another standpoint.
Motivated by the definition of “between-ness” from graph
theory [4], [5] we introduce the notion of link and path
criticality indexes. The essence of the approach is to identify
the most critical paths to avoid running the flows on these paths
and try to set up LSPs and back up LSPs on less critical paths
in the short-term as well as to plan to increase the bandwidth of
the paths with high criticality index when possible. Our
simulations on some benchmark networks discussed in the
research literature show that the approach is promising.
The paper organized as follows. Section II reviews the
state of the art in MPLS routing and flow assignment problems.
We describe the issues with existing methods and the
associated research challenges. In section III, we introduce the
mathematics and notions of algebraic routing based on [6] that
allow us, in section IV, propose our path-criticality index-based
routing scheme. Section V provides an extension to compute
maximally disjoint back up paths. Section VI provides
validation for our proposal. We assess our proposed method on
some benchmark networks and present encouraging results.
Finally we conclude with a discussion of open issues and future
work.
II. CURRENT STATE OF THE ART
The most popular algorithm used in the research
community for routing LSPs is the shortest path routing (SP).
In this method the path with the least total number of links
between source and destination (or the one with minimum
overall weight when the links are weighted) is chosen. If there
is more than one shortest path between source and destination,
the algorithm is flexible, and one can choose a shortest path at
random. Typically, the LSP setup algorithm keeps information
about the residual capacity of each link, and when a new
request comes, only the links with enough residual capacity are
considered by the shortest path algorithm. While the shortest
path algorithm enjoys the benefit of simplicity, it can cause
major problems to the network due to the lack of any load
balancing mechanism. Some links might become saturated
while some remain underutilized.
1-4244-0353-7/07/$25.00 ©2007 IEEE
This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the ICC 2007 proceedings.