Current video conferencing design based on client-server or peer-to-peer architecture do not scale to large number of users. ICN addresses this through network-based multicast built on applications semantics. Considering the pull-based content primitive, scalability is also required to learn the current state of the media names (text, audio, video) of participants considering adhoc join/leave behavior due to variable number of participants, or situations involving mobility. We propose a solution based on an overlay of ICN-based service edge routers (iSER). iSERs being CCN routers allow content multicasting which allows data plane scalability. In [1], we proposed an architecture to sync names using services hosted on the iSER, scalable to large number of conference participants. This paper extends [1] for video conferencing, focussing on a conference participant video production and ingestion, employing (iSER) services to sync conference state among participants, and a controller infrastructure to realize it as a service. We evaluate our design with experiments over a prototype system and share initial results considering several parameters that affect the real-time quality of a video conferencing session in our implementation.
[1]
Yue Lu,et al.
Measurement Study of Multi-party Video Conferencing
,
2010,
Networking.
[2]
Asit Chakraborti,et al.
Demo overview: multi-party conference over virtual service edge router (vser) platform
,
2014,
ICN '14.
[3]
Pablo César,et al.
Enabling Composition-Based Video-Conferencing for the Home
,
2011,
IEEE Transactions on Multimedia.
[4]
Alexander Afanasyev,et al.
Chronos : Serverless Multi-User Chat Over NDN
,
2012
.
[5]
Asit Chakraborti,et al.
Towards software defined ICN based edge-cloud services
,
2013,
2013 IEEE 2nd International Conference on Cloud Networking (CloudNet).
[6]
Van Jacobson,et al.
Networking named content
,
2009,
CoNEXT '09.
[7]
Sakir Sezer,et al.
Queen ' s University Belfast-Research Portal Are We Ready for SDN ? Implementation Challenges for Software-Defined Networks
,
2016
.