Skip navigation
Please use this identifier to cite or link to this item: https://repository.esi-sba.dz/jspui/handle/123456789/86
Full metadata record
DC FieldValueLanguage
dc.contributor.authorDRICI, OUssama-
dc.contributor.authorHADJAME, MOhamed-
dc.date.accessioned2022-04-03T11:32:09Z-
dc.date.available2022-04-03T11:32:09Z-
dc.date.issued2021-
dc.identifier.urihttps://repository.esi-sba.dz/jspui/handle/123456789/86-
dc.descriptionM. Zouaoui Chakib Encadreur M. Benslimane Sidi Mohamed Co-Encadreuren_US
dc.description.abstractIn MPTCP, a congestion control algorithm (CCA) is an important factor that governs data flow through multiple subflows (SFs). There are two major difficulties related with MPTCP CCAs. First, MPTCP flows must have an advantage over single-path flows; sec ond, MPTCP flows should be fair, implying that SFs sharing a similar bottleneck should use a bandwidth that is comparable to that of a single-path flow. Several MPTCP CCAs have been created; however, none of them have achieved these challenges in all scenarios. Google has announced the bottleneck bandwidth and round-trip-time (BBR), the new CCA for single-path TCP that achieves high throughput with minimal delay through the use of a network model. In this work, we presented a modification and optimization of BBR that addresses the basic challenges of MPTCP by using the notion of network modeling in BBR.*** Dans le protocole MPTCP, un algorithme de contrˆole de congestion (CCA) est un fac teur important qui r´egule le flux de donn´ees `a travers plusieurs sous-flux (SFs). Il existe deux difficult´es majeures li´ees aux CCA MPTCP. Premi`erement, les flux MPTCP doivent avoir un avantage sur les flux `a chemin unique ; deuxi`emement, les flux MPTCP doivent ˆetre ´equitables, ce qui implique que les SF partageant un goulot d’´etranglement similaire doivent utiliser une bande passante comparable `a celle d’un flux `a chemin unique. Plusieurs CCA pour MPTCP ont ´et´e cr´e´es, mais aucun d’entre eux n’a r´eussi `a relever ces d´efis dans tous les sc´enarios. Google a annonc´e la bande passante et le temps de parcours du goulot d’´etranglement (BBR), une nouvelle ACC pour TCP `a chemin unique qui permet d’atteindre un d´ebit ´elev´e avec un retard minimal grˆace `a l’utilisation d’un mod`ele de r´eseau. Dans ce travail, nous avons pr´esent´e une modification et une optimisation de BBR qui r´epond aux d´efis fondamentaux de MPTCP en utilisant la notion de mod´elisation de r´eseau dans BBR.en_US
dc.language.isoenen_US
dc.subjectMultipathen_US
dc.subjectMPTCPen_US
dc.subjectTCPen_US
dc.subjectCongestion Controlen_US
dc.subjectBBRen_US
dc.titleOptimization of BBR in MPTCPen_US
dc.typeThesisen_US
Appears in Collections:Ingénieur

Files in This Item:
File Description SizeFormat 
EngineerVf.pdf141,09 kBAdobe PDFView/Open
Show simple item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.