Enhancing P2P live streaming performance by balancing description distribution and available forwarding bandwidth in P2P streaming network

Lin, Chow-Sing

doi:10.1002/dac.1173

Cited by 14 publications

(10 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given these values, this roughly amounts to 12per cent of the total of the overlay nodes joining and leaving every minute. These parameters are chosen as such because they are typical of P2P live video streaming overlays as shown in . As each dynamic simulation lasts for 2h, this distribution of the session lengths produces a lot of churn.…”

Section: Discussionmentioning

confidence: 99%

Dynamic greedy routing in overlay networks using virtual coordinates from the hyperbolic plane

Magoni

Cassagnes

2015

Trans. Emerging Tel. Tech.

View full text Add to dashboard Cite

Greedy routing algorithms based on virtual coordinates have attracted considerable interest in recent years. Those based on coordinates taken from the hyperbolic plane have interesting theoretical scalability properties. However, their scalability and reliability are yet to be ensured when applied to large‐scale dynamic networks. In this paper, we propose a scalable and reliable solution for creating and managing dynamic overlay networks where nodes have hyperbolic coordinates. In this context, our solution provides a greedy routing algorithm based on the hyperbolic distance. To cope with network dynamics, we have defined two methods for avoiding temporary local minima and one method for maintaining the greedy embedding over time. Through analysis, we evaluate the complexity costs of our solution. Through simulations, we assess the scalability of our solution on static networks and its reliability on dynamic networks. Results show that using our solution based on hyperbolic geometry provides scalability and reliability to both addressing and routing tasks in dynamic overlay networks. Copyright © 2015 John Wiley & Sons, Ltd.

show abstract

Section: Discussionmentioning

confidence: 99%

Dynamic greedy routing in overlay networks using virtual coordinates from the hyperbolic plane

Magoni

Cassagnes

2015

Trans. Emerging Tel. Tech.

View full text Add to dashboard Cite

show abstract

“…Calculation of node's utility and resorting the bid will be carried out after each processing round. The time complexity of Merge sort is O ( k ⋅ log 2 k ) . In our algorithm, we sort the bids of K nodes by Merge sort.…”

Section: Vickrey–clarke–groves‐based Scheme For Replica Allocationmentioning

confidence: 99%

“…Let the adjusting step size be h, therefore, processing time for one node will be completed within 100= h. Calculation of node's utility and resorting the bid will be carried out after each processing round. The time complexity of Merge sort is O.k log 2 k/ [24,25]. In our algorithm, we sort the bids of k nodes by Merge sort.…”

Section: Proofmentioning

confidence: 99%

A group strategy‐proof incentive approach for eliminating selfish behaviors in peer‐to‐peer file allocation

Zuo

Zhang

2013

Int J Communication

View full text Add to dashboard Cite

In this paper, we formulate and solve the selfish allocation problem by using game theory, which is different from the previously studied researches in three ways that make it more accurately reflective of real world peer-to-peer (P2P) allocation: (i) we treat the nodes as strategic agents and treat the replica allocation as a deliberate auction where node is incentivized to give his true quality of service for obtaining the replica; (ii) our mechanism computes node utility for all possible replica destination and payments for those destination nodes, and the best appropriate node can be selected as the final placement destination; and (iii) we show how to carry out our scheme with a distributed algorithm that is a straightforward extension to P2P allocation method and causes an overhead in convergence time. Our design and analysis of a strategy proof, feasible, Vickrey-Clarke-Groves-based auction scheme provides a new, promising direction in distributed algorithmic mechanism design, which has heretofore been focused mainly on P2P application. Copyright

show abstract

“…Hence, all the IP Multicast functionalities are shifted into the application layer [9]. The Application Layer Multicast (ALM) protocols, suffer from the delay stretch, bandwidth, throughput, reliability, group management, network management and security [10], [11], [12]. ALM creates an overlay network rather than an IP layer network [13].…”

Section: Introductionmentioning

confidence: 99%

POSET: A high-performance application layer multicast protocol design and implementation

Ruso

Chellappan

2014

China Commun.

View full text Add to dashboard Cite

83topology of the network is needed. Multicast was initially introduced by the Stephen Deering [1]. A lot of research work has been done on multicast routing protocols [2,3,4,5,6,7]. All the protocols discussed in the existing work have been focused on extending the multicast routing protocols. Usually, these existing routing protocols construct a multicast spanning tree from a source to all the group members. The nodes (switches or routers) in the network replicate packets to reach multiple receivers, such that the messages are sent only once. The most popular IP multicast protocols are the IGMP and PIM. The IGMP is an internet group management protocol. It involves two devices; the IGMP host and the IGMP router. There are three basic functions, viz, JOIN, LEAVE and QUERY. It uses two types of IGMP messages, Membership Report and Membership Query. It uses two reserved multicast addresses, 224.0.0.1 and 224.0.0.22 [3]. PIM is a Protocol Independent Multicast routing that runs over an existing unicast infrastructure. PIM uses two modes of operations viz; (i) PIM-DM (dense mode) and (ii) PIM-SM (sparse mode). The PIM-DM differs from the PIM-SM in two essential ways. One in which, there is no periodic join transmitted, only explicitly triggered prunes and grafts; and the other, in which there is no RP (Rendezvous Point). The PIM-sparse mode [4], assumes that nodes are likely to be located far away from each other. The available Abstract: Many research issues have been raised in Application Layer Multicasting (ALM), such as group management, security, integrity of data, link stress, link stretch, load balancing, fault tolerance and scalability, because of the shifting of the multicast protocol from the IP layer to the application layer. To address these issues many protocols have evolved by changing their topology structure. In this paper, the POSET protocol stack is proposed, which consists of three layers, such as communication control, POSET cube, and content distribution. The novelty of this paper is the lattice-based data distribution with POSET cube architecture. The results have been compared with the existing NICE and Narada protocols. The experimental results show that the proposed POSET protocol improves throughput between 7.14% and 40% and decreases the delay between 7.69% and 25%, compared to the existing NICE protocol.

show abstract

Enhancing P2P live streaming performance by balancing description distribution and available forwarding bandwidth in P2P streaming network

Cited by 14 publications

References 28 publications

Dynamic greedy routing in overlay networks using virtual coordinates from the hyperbolic plane

Dynamic greedy routing in overlay networks using virtual coordinates from the hyperbolic plane

A group strategy‐proof incentive approach for eliminating selfish behaviors in peer‐to‐peer file allocation

POSET: A high-performance application layer multicast protocol design and implementation

Contact Info

Product

Resources

About