Field demonstration of multi-domain software-defined transport networking with multi-controller collaboration for data center interconnection [Invited]

Software-defined networking (SDN) with optical transport techniques enables network operators and data center (DC) operators to provide their resources dynamically on user's demands while minimizing operating and capital expenditure. Current multi-domain networking techniques mainly rely on the...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Journal of optical communications and networking Ročník 7; číslo 2; s. A301 - A308
Hlavní autori: Yu, Yiming, Zhang, Jie, Zhao, Yongli, Lin, Yi, Han, Jianrui, Zheng, Haomian, Cui, Yadi, Xiao, Minglu, Li, Hui, Peng, Yang, Ji, Yuefeng, Yang, Haifeng
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: IEEE 01.02.2015
Predmet:
ISSN:1943-0620, 1943-0639
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Software-defined networking (SDN) with optical transport techniques enables network operators and data center (DC) operators to provide their resources dynamically on user's demands while minimizing operating and capital expenditure. Current multi-domain networking techniques mainly rely on the path-computing element, while SDN has shown considerable potential for this inevitable issue. The first field demonstration of multi-domain software-defined transport networking (SDTN) without global topology information is detailed in this paper. A multi-controller collaboration framework and three schemes, a controller-driven scheme (ConDS), a cloud-driven scheme (ClDS), and a ClDS with dynamic optimization (ClDS-DO), for datacenter interconnection based on SDTN are demonstrated via field networks. We also extend the OpenFlow protocol and design the Java-Script Object Notation application programming interface to support this framework. Multi-domain lightpaths are automatically provided with limited signaling latency. The blocking performances of the proposed schemes are estimated, and the ClDS-DO has the best performance approaching the optimal boundary.
ISSN:1943-0620
1943-0639
DOI:10.1364/JOCN.7.00A301