Thesis (MSc)--Stellenbosch University, 2008. / ENGLISH ABSTRACT: Traffic Engineering (TE) is intended to be used in next generation IP networks to optimize
the usage of network resources by effecting QoS agreements between the traffic offered
to the network and the available network resources. TE is currently performed by the
IP community using three methods including (1) IGP TE using connectionless routing
optimization (2) MPLS TE using connection-oriented routing optimization and (3) Hybrid
TE combining IGP TE with MPLS TE. MPLS has won the battle of the core of the Internet
and is making its way into metro, access and even some private networks. However,
emerging provider practices are revealing the relevance of using IGP TE in hybrid TE
models where IGP TE is combined with MPLS TE to optimize IP routing. This is done by
either optimizing IGP routing while setting a few number of MPLS tunnels in the network
or optimizing the management of MPLS tunnels to allow growth for the IGP traffic or
optimizing both IGP and MPLS routing in a hybrid IGP+MPLS setting.
The focus of this thesis is on IGP TE using heuristic algorithms borrowed from the computational
intelligence research field. We present four classes of algorithms for Maximum
Link Utilization (MLU) minimization. These include Genetic Algorithm (GA), Gene Expression
Programming (GEP), Ant Colony Optimization (ACO), and Simulated Annealing
(SA). We use these algorithms to compute a set of optimal link weights to achieve IGP
TE in different settings where a set of test networks representing Europe, USA, Africa and
China are used. Using NS simulation, we compare the performance of these algorithms
on the test networks with various traffic profiles. / AFRIKAANSE OPSOMMING: Verkeersingenieurswese (VI) is aangedui vir gebruik in volgende generasie IP netwerke vir
die gebruiksoptimering van netwerkbronne deur die daarstelling van kwaliteit van diens
ooreenkomste tussen die verkeersaanbod vir die netwerk en die beskikbare netwerkbronne.
VI word huidiglik algemeen bewerkstellig deur drie metodes, insluitend (1) IGP VI gebruikmakend
van verbindingslose roete-optimering, (2) MPLS VI gebruikmakend van verbindingsvaste
roete-optimering en (3) hibriede VI wat IGP VI en MPLS VI kombineer. MPLS
is die mees algemene, en word ook aangewend in metro, toegang en selfs sommige privaatnetwerke.
Nuwe verskaffer-praktyke toon egter die relevansie van die gebruik van IGP VI
in hibriede VI modelle, waar IGP VI gekombineer word met MPLS VI om IP roetering te
optimeer. Dit word gedoen deur `of optimering van IGP roetering terwyl ’n paar MPLS
tonnels in die netwerk gestel word, `of optimering van die bestuur van MPLS tonnels om
toe te laat vir groei in die IGP verkeer `of die optimering van beide IGP en MPLS roetering
in ’n hibriede IGP en MPLS situasie.
Die fokus van hierdie tesis is op IGP VI gebruikmakend van heuristieke algoritmes wat
ontleen word vanuit die berekeningsintelligensie navorsingsveld. Ons beskou vier klasse van
algoritmes vir Maksimum Verbindingsgebruik (MVG) minimering. Dit sluit in genetiese
algoritmes, geen-uitdrukkingsprogrammering, mierkoloniemaksimering and gesimuleerde
temperoptimering. Ons gebruik hierdie algoritmes om ’n versameling optimale verbindingsgewigte
te bereken om IGP VI te bereik in verskillende situasies, waar ’n versameling
toetsnetwerke gebruik is wat Europa, VSA, Afrika en China verteenwoordig. Gebruikmakende
van NS simulasie, vergelyk ons die werkverrigting van hierdie algoritmes op die
toetsnetwerke, met verskillende verkeersprofiele.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/20877 |
| Date | 03 1900 |
| Creators | Wang, Hong Feng |
| Contributors | Bagula, Antoine B., Krzesinski, A. E., Stellenbosch University. Faculty of Science. Dept. of Mathematical Sciences. Institute for Applied Computer Science. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | xii, 93 leaves : ill., maps |
| Rights | Stellenbosch University |
Page generated in 0.0147 seconds