MISE: MIddleboxes SElection for Multi-domain Service Function Chains

Cite this publication

​MISE: MIddleboxes SElection for Multi-domain Service Function Chains​
Silvestro, A.; Bifulco, R.; Schneider, F.; Fu, X.   & Kangasharju, J.​ (2017)
pp. 37​-42. ​CAN’17: ACM Cloud-Assisted Networking Workshop​, Incheon.
Korea​: Cloud-Assisted Networking Workshop. DOI: https://doi.org/10.1145/3155921.3155923 

Documents & Media

License

GRO License GRO License

Details

Authors
Silvestro, Alessio; Bifulco, Roberto; Schneider, Fabian; Fu, Xiaoming ; Kangasharju, Jussi
Abstract
In-network services, usually implemented through middle- boxes, play a key role in today’s Internet. They enhance end-to-end connections with the ability to provide content optimization, performance enhancement, privacy protection, etc. Multiple in-network services are often combined together to form a so-called Service Function Chain (SFC). Most SFC steering techniques are effective only when there is full visibil- ity and/or control of the underlying network infrastructure. However, when multiple independent parties are involved, the chain building process may be inefficient. We propose a Collaborative SFC Resolution method that provides optimized chain building, while requiring only min- imum information sharing among independent in-network service providers. We implemented a prototype using the Do- main Name System (DNS) in order to evaluate its impact on the middleboxes selection. Through an extensive evaluation on real topologies and simulated network graphs we prove that the collaborative middleboxes selection enables to find near-optimal solutions. More in detail, it shows up to 30% improvement, in terms of end-to-end network delay, when compared to solutions that do not share information and perform the middleboxes selection independently
Issue Date
2017
Publisher
Cloud-Assisted Networking Workshop
Conference
CAN’17: ACM Cloud-Assisted Networking Workshop
ISBN
978-1-4503-5423-3
Conference Place
Incheon
Event start
12
Event end
12
Language
English

Reference

Citations


Social Media