Abstract: Periodic broadcast is a cost-effective solution for large-scale distribution of popular videos because this approach guarantees constant worst service latency, regardless of the number of video requests. An essential periodic broadcast method is the client-centric approach (CCA), which allows clients to use smaller receiving bandwidth to download broadcast data. An enhanced version, namely CCA++, was proposed to yield a shorter waiting time. This work further improves CCA++ in reducing client buffer requirements. The new scheme decreases the buffer requirements by as much as 52% when compared to CCA++. This study also provides an analytical evaluation to demonstrate the performance advantage, as compared with particular schemes.
Abstract: In this paper we present high performance
dynamically allocated multi-queue (DAMQ) buffer schemes for fault
tolerance systems on chip applications that require an interconnection
network. Two virtual channels shared the same buffer space. Fault
tolerant mechanisms for interconnection networks are becoming a
critical design issue for large massively parallel computers. It is also
important to high performance SoCs as the system complexity keeps
increasing rapidly. On the message switching layer, we make
improvement to boost system performance when there are faults
involved in the components communication. The proposed scheme is
when a node or a physical channel is deemed as faulty, the previous
hop node will terminate the buffer occupancy of messages destined
to the failed link. The buffer usage decisions are made at switching
layer without interactions with higher abstract layer, thus buffer
space will be released to messages destined to other healthy nodes
quickly. Therefore, the buffer space will be efficiently used in case
fault occurs at some nodes.