Résumé de thèse
66 pages
English
Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres
66 pages
English
Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres

Description

Resum´ e´ de these`Ignasi Sau VallsDirecteurs: David Coudert, Jean-Claude BermondMascotte - Boreon´12-13 mars 2009Ignasi Sau Valls (Mascotte) Resum´ e´ de these` 12-13 mars 2009 1 / 28Plan de l’expose´ (et de la these` ..)Partie I: Groupage de traficPartie II: Sous-graphes avec contraintes sur le degre´Partie III: Autres problemes` (plus ou moins relies)´ConclusionsIgnasi Sau Valls (Mascotte) Resum´ e´ de these` 12-13 mars 2009 2 / 28Traffic GroomingIgnasi Sau Valls (Mascotte) Resum´ e´ de these` 12-13 mars 2009 3 / 28General ideaWDM networks (Wavelength Division Multiplexing)I 1 wavelength = up to 40 Gb/sI 1 fiber = hundreds of wavelengths = Tb/sIdea: Traffic grooming consists in grouping low-speed trafficflows into higher speed streams! we allocate the same wavelength to several low-speedrequests (TDM, Time Division Multiplexing)Objectives:I Efficient usage of bandwidthI Minimize network costIgnasi Sau Valls (Mascotte) Resum´ e´ de these` 12-13 mars 2009 4 / 28General ideaWDM networks (Wavelength Division Multiplexing)I 1 wavelength = up to 40 Gb/sI 1 fiber = hundreds of wavelengths = Tb/sIdea: Traffic grooming consists in grouping low-speed trafficflows into higher speed streams! we allocate the same wavelength to several low-speedrequests (TDM, Time Division Multiplexing)Objectives:I Efficient usage of bandwidthI Minimize network costIgnasi Sau Valls (Mascotte) Resum´ e´ de these` 12-13 mars 2009 4 / 28General ideaWDM networks (Wavelength ...

Informations

Publié par
Nombre de lectures 53
Langue English

Extrait

IguVSasinasaM(slla´R)ettochte`es21semue´ed0091/28-13mars2
R´esum´edeth`ese
Ignasi Sau Valls Directeurs:David Coudert, Jean-Claude Bermond
Mascotte-Bor´eon
12-13 mars 2009
ngIR)e´toetdeteus´mauVaasiSMasclls(
PartieII:Sous-graphesaveccontraintessurledegr´e
Partie I: Groupage de trafic
Conclusions
PartieIII:Autresprobl`emes(plusoumoinsreli´es)
2/09
Plan de l’expos ´ ( t de la the` se..) e e
28am3102srse`h-21e
Ignasi
Sau
Valls
(Mascotte)
Traffic
Grooming
Resume ´ ´
de
th`ese
12-13
mars
2009
3
/
28
m´ed´esute)Rscot(saMaVllSiuangsaI
Objectives: IEfficient usage of bandwidth IMinimize network cost
−→we allocate the same wavelength to several low-speed requests (TDM, Time Division Multiplexing)
General idea
Idea:Traffic groomingconsists in grouping low-speed traffic flows into higher speed streams
WDM networks (Wavelength Division Multiplexing) I1 wavelength = up to 40 Gb/s I1 fiber = hundreds of wavelengths = Tb/s
0920284/31-2sram`hte1ese
octtM(salaslaSVunasiIg12se3m-1s2ar9400´R)emuseede´e`ht2/8
Objectives: IEfficient usage of bandwidth IMinimize network cost
General idea
Idea:Traffic groomingconsists in grouping low-speed traffic flows into higher speed streams
−→we allocate the same wavelength to several low-speed requests (TDM, Time Division Multiplexing)
WDM networks (Wavelength Division Multiplexing) I1 wavelength = up to 40 Gb/s I1 fiber = hundreds of wavelengths = Tb/s
−→allocate the same wavelength to several low-speedwe requests (TDM, Time Division Multiplexing)
Objectives: IEfficient usage of bandwidth IMinimize network cost
WDM networks (Wavelength Division Multiplexing) I1 wavelength = up to 40 Gb/s I1 fiber = hundreds of wavelengths = Tb/s
Idea:Traffic groomingconsists in grouping low-speed traffic flows into higher speed streams
094/28
General idea
13mars20se`h-21e´mustedeteot´e)Rs(llscMauaaVsaSiIng
-13mars2th`ese12semue´edoctt)e´R
OADM(Optical Add/Drop Multiplexer)= adds-extracts a wavelength from a fiber ADM(Add/Drop Multiplexer)= add-extracts OC/STM (low speed signal) from a wavelength
we want to minimize the number of ADMs
−→
ADM and OADM
590082/VuaSisansaM(sllaIg
isanVuaSgIttcoR´e)lsalas(Mhte`es21semue´ed0095/28-13mars2
OADM(Optical Add/Drop Multiplexer)= adds-extracts a wavelength from a fiber ADM(Add/Drop Multiplexer)= add-extracts OC/STM (low speed signal) from a wavelength
we want to minimize the number of ADMs
−→
ADM and OADM
9600s2ar3m-112see`htede´muse´R)ecott(MasallsSauVanisgI2/8
Request(i,j): pair of nodes(i,j)that want to exchange (low-speed) traffic Grooming factor C:
Denitions
Loadof an arc on a wavelength: number of requests using this arc on this wavelength (C)
ValliSaugnasI/682
Denitions
Loadof an arc on a wavelength: number of requests using this arc on this wavelength (C)
Request(i,j): pair of nodes(i,j)that want to exchange (low-speed) traffic Grooming factor C:
amsr0290R)etuse´aM(stocse1es132-edm´h`et
20rsma132-e1esh`
Idea:UseADMs only at the endpoints of arequest (lightpaths)to save as many ADMs as possible
OADM(Optical Add/Drop Multiplexer)= adds-extracts a wavelength from a fiber ADM(Add/Drop Multiplexer)= add-extracts OC/STM (low speed signal) from a wavelength
ADM and OADM
82/790ngIlls(MascasiSauVaus´mdetetoetR)e´
  • Univers Univers
  • Ebooks Ebooks
  • Livres audio Livres audio
  • Presse Presse
  • Podcasts Podcasts
  • BD BD
  • Documents Documents