Politecnico di Torino | Servizi per la didattica

KEYWORD

ALCATEL LUCENT BELL LABS France - on the evolution of Content Distribution Networks (CDN)

azienda Thesis in external company estero Thesis abroad

keywords CONTENT CACHING, INTERNET TRAFFIC

Reference persons MARCO MELLIA

External reference persons Samir Ghamri-Doudane – Bell Labs, Alcatel-Lucent – samir.ghamri-doudane@alcatel-lucent.com Giovanna Carofiglio – Bell Labs, Alcatel-Lucent – giovanna.carofiglio@alcatel-lucent.com

Research Groups Alcatel Lucent Bell Labs - France, Telecommunication Networks Group

Thesis type MODELING AND EXPERIMENTAL

Description The Internet has evolved from a medium providing host-to-host data communication to a multi- tenant complex system centered on content distribution. This trend, together with the ever increasing growth of Internet traffic, called for the birth of content distribution network (CDN) architectures to serve content providers needs and user demand. Although these solutions responded to the challenge, they still show inefficiencies. For instance, the misuse of DNS [1] for request routing introduces additional latency, lack of reactivity and localization inaccuracy. Moreover, the connection-oriented nature of the HTTP/TCP transport model makes it hard to deal with both user and content mobility with consequent performance degradation.
The Information-Centric Networking (ICN) paradigm has been proposed to answer the above mentioned limitations by design (see [2] for a good survey). In ICN, the network layer is enriched with content-awareness, and content/service delivery is achieved via three main communication mechanisms: in-network caching, name-based request routing and pull-based connectionless transport. While a complete deployment of an ICN architecture requires upgrades to the current Internet protocol suite and network equipment functionalities, ICN communication mechanisms can already provide benefits if integrated at application layer within today’s content delivery solutions.
The internship will focus on proposing solutions to enable such integration. More precisely, the internship will pursue the following steps (and related objectives).
Analysis:
 Study current CDN solutions and outline their limitations and inefficiencies.
 Review ICN proposals and understand how ICN functionalities (e.g., [3, 4, 5]) can
overcome these limitations and optimize the content delivery process.
Design:
 Identify the relevant ICN mechanisms that can be integrated within a CDN at application layer.
 Propose solutions to enable this integration at minimal cost (impact on current CDN infrastructures).
Evaluation:
 Set up a test platform using available tools and infrastructures (lab resources).
 Assess the feasibility of the designed solutions and quantify the gains brought by ICN
mechanisms.

References:
[1] Paul Vixie, “What DNS Is Not”, In ACM Queue, 2009.
[2] B. Ahlgren, C. Dannewitz, C. Imbrenda, B. Ohlman, “A survey of Information-Centric Networking”, In Proc. Of Dagstuhl Seminar on Information-Centric Networking, 2011.
[3] D. Perino and M. Varvello, “A reality check for Content-Centric Networking”, In Proc. of ACM SIGCOMM workshop on Information Centric Networking, 2012.
[4] G. Carofiglio, M. Gallo, L. Muscariello, “ICP: Design and Evaluation of an Interest control Protocol for Content Centric Networks”, In Proc. of IEEE INFOCOM Workshop on emerging design choices in name oriented networking, 2012.
[5] G. Carofiglio, M. Gallo, L. Muscariello, M. Papalini, “Multipath congestion control in content- centric networks”, In Proc. of IEEE INFOCOM Workshop on emerging design choices in name oriented networking, 2013.