Abstract:
Multicast Video-on-Demand (VoD) systems are scalable and cheap-to-operate. In such
systems, a single stream is shared by a batch of common user requests. In this research,
we adapt multicast communication technique in an Enterprise Network where multimedia
data are stored in distributed servers. We consider Batching with Patching. Batching
incurs service latency but increases the possibility of larger group formation. Patching
eliminates this service latency and further improves the performance of multicast communication
in VoD systems. However, in conventional patching, each shortly requested
client requires a separate server stream to patch the missing portion. We consider a novel
patching scheme called Client-Assisted Patching which outperforms conventional patching.
Clients' buffer of a multicast group can be used to patch the missing portion of
the clients who will request the same movie shortly. The newly admitted client in turn
can also provide patching stream to the later client. This scheme significantly reduces the
server load without requiring larger client cache space than conventional patching scheme.
The conserved server bandwidth can be used to serve more client requests. Simulation experiments
show that our scheme is more scalable and cheap-to-operate than conventional
patching.
