S.-H.G. Chan

Distributed servers architecture for networked video services

In an on-demand video system, the video repository generally has limited streaming capacities and may be far from the users. In order to achieve higher user capacity and lower network transmission cost, distributed servers architecture can be used, in which multiple local servers are placed close to user pools and, according to their local demands, dynamically cache the contents streamed from the repository. We study a number of caching schemes as applied in the local servers depending on whether the repository is able to multicast movie contents to the local servers or not, and whether the local servers can exchange their cached contents among themselves or not. Our caching schemes keep a circular buffer of data for the movie requested, and hence movies are partially cached. By adjusting the size of the buffer, such caching is able to achieve better tradeoff between network channels and local storage as compared to the traditional caching in which a movie is treated as an entity. For each caching scheme, we study the tradeoff between the local storage and the network channels, and address how the total cost of the system can be minimized by appropriately sizing the buffer. As compared to a number of traditional operations (request batching and multicasting, true-VOD, etc.), we show that distributed servers architecture is able to achieve much lower system cost to offer on-demand video services.

WIANI: wireless infrastructure and ad-hoc network integration

Wireless networks have been widely deployed in recent years to provide high-speed Internet access to mobile users. In traditional IEEE 802.11 wireless LANs, all users directly connect to an access point (AP) and all packets are forwarded by the AP. As a result, the coverage and capacity of the network is limited. If ad hoc mode is adopted in both the AP and mobile nodes, the one hop connections from AP can be extended to multiple hops. Such an architecture, termed WIANI (wireless infrastructure and ad-hoc network integration), is able to extend the network coverage beyond the coverage of APs. Furthermore, users may take advantage of the ad hoc connections to forward local data and hence alleviate the traffic load through the AP and increase the network capacity. We propose a dynamic load-balancing protocol for WIANI in which all APs and nodes operate in ad-hoc mode. Our protocol consists of two parts, a load-balancing zone forming algorithm and a weighted x-hop routing algorithm. Using simulation, we show that our protocol improves system throughput and reduces packet delivery delay.

Caching schemes for distributed video services

In an on-demand video system, the repository servers storing all the video contents generally have limited streaming capacities and may not be co-located with the users. To achieve higher user capacity and lower network transmission cost, the distributed servers architecture can be used. In such a system, multiple local servers are placed close to the user pools and, according to their local demands, the servers dynamically cache the contents streamed from the repository. We study in this paper a number of caching schemes by considering whether the local servers can exchange their cached contents among themselves or not. All the caching schemes keep a sliding window worth of data for each video being displayed; hence a video can be partially stored. We study the trade-off between the storage capacity required and the network channel used in each scheme, and address how the system cost can be minimised by appropriately sizing the window. We also show the cost advantage in using such a system over a system of request batching and multicasting in providing distributed video services.

Providing on-demand video services using request batching

In an on-demand video system, requests for a video file arriving within a period of time can be batched together and served with a single multicast stream; thereby reducing the bandwidth requirement compared with the unbatched case. We study various batching schemes to meet delay and profitability requirements-in the window-size based and moving-average schemes, maximum and average user delay respectively are guaranteed; in the batch-size based scheme, minimal profitability is maintained; in the adaptive scheme, both the delay and profitability can be balanced. We analyse and compare these schemes in terms of the delay experienced by the users, and their profitability (the number of users in a batch, the number of concurrent streams, etc.).

On achieving profit in providing near video-on-demand services

In near video-on-demand, requests for the same movie arriving within a period of time are grouped together (i.e., batched) and served with a single multicast stream; thereby reducing the bandwidth requirement compared with the unbatched case. We consider here that the use of a multicast channel comes with a cost. Since delayed users may cancel their requests while being batched, this uncollected revenue has to be appropriately balanced with the channel usage cost in order to achieve maximum profit. We address profit issues for a number of batching schemes. In the window-based schemes, users are batched for a fixed period of time before they are served; in the batch-size based scheme, users are served once a certain number of them are collected. We study the system profit in terms of user reneging behavior, movie popularity, and multicast channel cost. By combining the advantages of the window-based and the batch-size based schemes, we introduce a scheme which adapts to the fluctuating request rate to achieve high profit.

Threshold-based admission policies for video services

In a video system, a central server stores all the files of interest to a large number of geographically distributed users. In order to lower the network transmission cost, the service provider may set up local servers to cache the video contents dynamically according to their local demands; thus forming a distributed video system. To lower the transmission cost further, the local servers can batch users together before presenting their video requests to the central server. In this case, if a video request to the central server is accepted, the provider enjoys the income collected by the batch while incurring a certain network transmission cost to the local servers. Due to the limited capacity at the central server, if video requests were accepted indiscriminately, the server may run out of channels too soon, which subsequently leads to the rejection of more profitable batches (and hence low profit). In order to better utilize the server channels, we consider admission policies which accept requests according to a threshold depending on the number of available channels at that time. We demonstrate that such admission policies can substantially increase the system profit while still meeting a certain user loss objective.

Asymptotic performance of a buffered shufflenet with deflection routing

Throughput of a shufflenet with deflection routing under high load and low load is obtained as a function of the network and buffer sizes. The authors give general routing conditions which achieve high performance in a shufflenet. Using a routing algorithm similar to the algorithm considered, the throughput of a shufflenet with only one buffer can be increased by more than 45% compared with the shufflenet without any buffer, the so-called hot-potato case. The increase is general for a shufflenet of size ranging from as few as 24 nodes to more than 10,000 nodes. The increase is more significant when the network becomes larger. The authors note that a large number of routing algorithms proposed to be used in the shufflenet satisfy the general routing conditions mentioned. Using the routing algorithm the authors mention, a shufflenet with only two buffers can achieve performance comparable to the store-and-forward case. In previous studies of the shufflenet, the derivation of the important parameter-the probability of deflection of a packet in the network-is usually complicated. The authors have obtained a simple approximation of this parameter, which greatly simplifies the analysis of a shufflenet of any size and with any number of buffers. This enables to conclude that the performance of a shufflenet scales well with different network and buffer sizes if the routing algorithm is chosen properly. They finally verify the results with the simulations that have been done.

TCP performance with deflection routing in the Internet

Deflection routing has been well studied for optical networks with regular topologies. We propose using deflection routing in the Internet and study its TCP performance. In particular, we show that when the difference between the delay of the deflection path and the shortest path (i.e., the deflection cost) is in a certain range, deflection routing can make almost full use of the free bandwidth in the deflection path and hence achieve substantial throughput improvement. In the worst case when the deflection cost is large, deflection routing can achieve an aggregate throughput no less than that without deflection routing. We analyze the underlying mechanisms of these characteristics. In order to extend the usefulness of deflection routing, we also propose a deflection routing scheme with adaptive deflection point and show, via simulation, that this scheme can achieve very high throughput as long as the deflection cost is no larger than a certain value. We show that deflection routing is friendly to existing traffic in the deflection path. One possibly unfavorable requirement of deflection routing is that. it should be enabled for either all or none of the flows contending for the same outgoing (congested) link.

Photon Recycling for Threshold Reduction in Semiconductor Lasers

Photon recycling. or the re-absorption of spontaneously emitted photons by an active gain mediuni can, in principle, reduce a laser’s threshold to zero even when the number of cavity modes is large. We calculate the magnitude of this effect for the current VCSELD device structures and discuss appropriate conditions for effective recycling.