Fig 1 - uploaded by Seif Haridi
Content may be subject to copyright.
Source publication
This paper presents gradienTv, a distributed, market-based approach to live streaming. In gradienTv, multiple streaming trees
are constructed using a market-based approach, such that nodes with increasing upload bandwidth are located closer to the
media source at the roots of the trees. Market-based approaches, however, exhibit slow convergence pro...
Context in source publication
Context 1
... Gradient overlay is a class of P2P overlays that arrange nodes using a local utility function at each node, such that nodes are ordered in descending utility values away from a core of the highest utility nodes [16,17]. As can be seen in Figure 1, the highest utility nodes (darkest colour) are found at the core of the Gradient, and nodes with decreasing utility values (lighter grays) are found at increasing distance from the centre. ...
Similar publications
Peer-to-peer (P2P) systems are used by millions of users every day. In many scenarios, it is desirable for the users from different P2P systems to communicate and exchange content resources with each other. This requires co-operation between the P2P systems, which is often difficult or impossible, due to the two following reasons. First, we have th...
Current Internet applications are still mainly bound to their transport layer connections. This prevents many features such as end-to-end security and mobility from functioning smoothly in a dynamic network. In this paper, we propose a novel architecture for decoupling communication from their supporting devices. This enforces the complete separati...
Citations
... Recently gradient topology comparing to other mentioned adaptive methods has been used in many applications [30][31][32][33] (Fig. 2). That is why it is very important to pay attention to the issue. ...
Peer-to-peer network is organized on top of another network as an overlay network. Super peer network is one of the peer-to-peer networks. A super peer, in a super peer based network, is a peer that has more responsibility than other peers have and is responsible for some of the tasks of network management. Since different peers vary in terms of capabilities, selecting a super peer is a challenge in super peer based networks. Gradient topology is of the networks based on super peers. Existing adaptive algorithms, which have been proposed to select super peer in gradient topology, are not aware of delays among the peers. In this paper, the proposed algorithm being aware of the delay among super peers, using learning automata, which is a reinforcement model of learning, selects the new super peers in an adaptive manner. According to the simulation results, the proposed algorithm with respect to the average end-to-end delay in community of super peers, and error in the super peer selection, has better performance than existing algorithms.
... The current technology of e-learning takes benefit from P2P network to boost up the service of e-learning to its users as potential solution for the bottleneck problem is assisting the P2P streaming network by a cloud computing infrastructure to guarantee a minimum level of QoS.Peer-to-peer (P2P) live streaming is becoming an increasingly popular technology, with a large number of academic [1], [2], [3], [5], [16] and commercial [17], [19] products being designed and deployed. Therefore, P2P networks which integrated with cloud were developed by researchers to achieved efficient usage of the network resources and also improved user experience in terms of bandwidth, delay and jitter. ...
Cloud computing has been introduced as a solution to several problems of the traditional web-based e-learning system, such as a limited storage, a high infrastructure maintenance costs and a low interoperability among the component of web-based e-learning system. However, the cloud computing system performance may deteriorate with increasing number of users and become worse when many users access the video streaming from the cloud system. This is due to the centralized architecture of the cloud computing that can generate network traffic congestion and bottleneck in the cloud servers. Peer-to-peer (P2P) architecture has been proposed to overcome this problem. Using P2P, all nodes in the cloud system can act as servers as well as clients at the same time for reducing the congestion and bottleneck of the system. Currently, there are few reviews on P2P video streaming although intensive studies have been done on the development of the system. With this condition, identifying and understanding the development of P2P video streaming will be expensive, time consuming and physically exhausting. The objective of this paper is to review the latest development of P2P video streaming based on cloud computing. A narrative review method has been used as the methodology for investigating the P2P video streaming’ articles from 2009 to 2014. The outcomes of this research shows that 90% of cloud based e-learning integrate with P2P when it's involving streaming video.
... Supernodes are also used to work around connectivity issues [12] by either relaying NAT traversal messages or even actual content when no direct connectivity can be established [1]. In P2P live streaming, supernodes with high upload capacity are tasked to retrieve the streaming data directly from the source of the stream and distribute it to a large number of neighbors [13] In this work, we tackle the supernode selection problem, that is the decision of which peers, among all nodes in the overlay, should become supernodes. This effort is motivated by the needs of a commercial peer-assisted live streaming platform called Hive Streaming [19] 1 , which utilizes supernodes to efficiently deliver content to the viewers. ...
In peer-to-peer applications deployed on the Inter-net, it is common to assign greater responsibility to supernodes, which are usually peers with high computational power, large amount of memory, or high network bandwidth capacity. In this paper, we describe a practical solution to the problem of supernode selection, that is the process of discovering the best peers in the network by some application-specific metric. We provide a distributed heuristic that allows to identify the best K nodes in the P2P overlay, by taking into consideration the realities of actual deployments, such as the presence of NATs. Our approach consists of an epidemic protocol which does not require new connections to be established, but rather relies on established connections, such as the ones provided by a NAT-resilient peer sampling framework. We support our claims with a thorough evaluation of our solution in simulation and in a real deployment on thousands of consumer machines.
... However , in overlay networks, nodes have only a relatively small partial view of the system, so preference functions are based only on local state and the state of the node's neighbors. Examples of existing overlay networks that construct their topologies using gossiping and preference functions include Spotify, that preferentially connects nodes with similar music play-lists [8] , Sepidar, that preferentially connects P2P livestreaming nodes with similar upload bandwidth capacity [9], and T-Man, a framework that provides a generic preference function for building such overlays [10]. To the best of our knowledge, there has been no analysis of the convergence properties of such information-carrying gossip-generated topologies built using preference functions. ...
... GLive uses an approximate auction algorithm to match nodes that are willing and able to share the stream with one another. GLive extends our previous work on treebased live-streaming, gradienTv [13] and Sepidar [9], to mesh-based live-streaming. Nodes want to establish connections to other nodes that are as close as possible to the source. ...
... The desired affect of our auction algorithm is that the source will upload to nodes who contribute the most upload bandwidth, who will, in turn, upload to nodes who contribute the next highest amount of bandwidth, and so on until the topology is fully constructed. More details on our approximate assignment algorithm can be found in [9]. One of the main problems with the lack of global information about nodes' upload bandwidths is that it affects the rate of convergence of auction algorithm. ...
In this paper, we investigate the topology convergence problem for the gossip-based Gradient overlay network. In an overlay network where each node has a local utility value, a Gradient overlay network is characterized by the properties that each node has a set of neighbors containing higher utility values, such that paths of increasing utilities emerge in the network topology. The Gradient overlay network is built using gossiping and a preference function that samples from nodes using a uniform random peer sampling service. We analyze it using tools from matrix analysis, and we prove both the necessary and sufficient conditions for convergence to a complete gradient structure, as well as estimating the convergence time. Finally, we show in simulations the potential of the Gradient overlay, by building a more efficient live-streaming peer-to-peer (P2P) system than one built using uniform random peer sampling.
... In GLive, we divide the media stream into a sequence of blocks, and each node pulls the blocks of the stream from a set of nodes called parents. Nodes use a distributed market model, first introduced in [16], to choose parents from among the nodes in the system. A major problem with market-based approaches that select parents from random nodes is that they exhibit slow convergence properties. ...
... Finally, we evaluate the convergence of our market model when the node samples are taken from the Gradient overlay compared to a random overlay. Our work is an extension of our previous work on multiple-tree live streaming [17, 16], and the contributions of this paper include: ...
... Many different overlay network topologies have been used for data delivery in P2P media streaming systems, but the two most widely used approaches are multiple-tree [16, 3, 13, 17] and mesh-based overlays [8, 6, 5] . Multipletree overlay networks use push-based content delivery over multiple tree-shaped overlays with the media source as a root of all trees. ...
Peer-to-Peer (P2P) live video streaming over the Internet is becoming increasingly popular, but it is still plagued by problems of high playback latency and intermittent playback streams. This paper presents GLive, a distributed market-based solution that builds a mesh overlay for P2P live streaming. The mesh overlay is constructed such that (i) nodes with increasing upload bandwidth are located closer to the media source, and (ii) nodes with similar upload bandwidth become neighbours. We introduce a market-based approach that matches nodes willing and able to share the stream with one another. However, market-based approaches converge slowly on random overlay networks, and we improve the rate of convergence by adapting our market-based algorithm to exploit the clustering of nodes with similar upload bandwidths in our mesh overlay. We address the problem of free-riding through nodes preferentially uploading more of the stream to the best up loaders. We compare GLive with our previous tree-based streaming protocol, Sepidar, and New Cools treaming in simulation, and our results show significantly improved playback continuity and playback latency.
... However, the hole punching service can be used by clients of the PSS to establish a direct connection with a sampled private node. NAT hole punching is typically required by applications such as video-on-demand [2] and live streaming [22, 23], where relaying would introduce too much overhead on public nodes. A private node may have several redundant partners. ...
... Finally, we show that if 80% of the nodes are private, and when 50% of the nodes suddenly fail, more than 92% of nodes stay connected. In future work, we will integrate our existing P2P applications with Gozar, such as our work on video streaming [22, 23], and evaluate their behaviour on the open Internet. ...
Gossip-based peer sampling protocols have been widely used as a building block for many large-scale distributed applications.
However, Network Address Translation gateways (NATs) cause most existing gossiping protocols to break down, as nodes cannot
establish direct connections to nodes behind NATs (private nodes). In addition, most of the existing NAT traversal algorithms
for establishing connectivity to private nodes rely on third party servers running at a well-known, public IP addresses. In
this paper, we present Gozar, a gossip-based peer sampling service that: (i) provides uniform random samples in the presence of NATs, and (ii) enables
direct connectivity to sampled nodes using a fully distributed NAT traversal service, where connection messages require only
a single hop to connect to private nodes. We show in simulation that Gozar preserves the randomness properties of a gossip-based
peer sampling service. We show the robustness of Gozar when a large fraction of nodes reside behind NATs and also in catastrophic
failure scenarios. For example, if 80% of nodes are behind NATs, and 80% of the nodes fail, more than 92% of the remaining
nodes stay connected. In addition, we compare Gozar with existing NAT-friendly gossip-based peer sampling services, Nylon
and ARRG. We show that Gozar is the only system that supports one-hop NAT traversal, and its overhead is roughly half of Nylon’s.
... However, the current peer sampling services are not quality-aware as they do not take peer capacity into account when choosing neighbors. The idea of locating higher capacity peers closer to the source than lower capacity peers is similar to the idea of gradient overlays, recently presented in [19], [20]. There are several main differences between these studies and ours. ...
... Payberah et. al. present gradienTv [20] using gradient overlays for live P2P streaming. GradienTv is a multiple tree based system for single-layer P2P streaming, whereas our protocol is made for unstructured layered P2P streaming. ...
With the standardization of SVC, the scalable extension of H.264/AVC, layered peer-to-peer streaming has attracted more and more attention as it offers adaptability to network fluctuations and heterogeneous end users. Although overlay construction is important for system performance, not much effort has been spent on unstructured overlay construction for layered peer-to-peer streaming. Related work more concentrates on layered streaming algorithms, and assumes that a list of peers for data exchange, called neighbors, is provided by traditional membership management protocols, e.g., SCAMP. Our previous studies have demonstrated that a random overlay is not good enough for layered peer-to-peer streaming. In this paper, we propose a new membership management protocol, based on peer sampling services, for layered peer-to-peer streaming. The protocol is quality-aware as it constructs the overlay so that (1) high capacity peers will be located at good positions in the overlay, e.g., close to the server, and (2) peers having similar capacity are likely to connect to each other. Both features are necessary to maximize bandwidth utilization of peers and to mitigate layer bottlenecks.With implementation of the protocol in PeerSim, we evaluate important graph properties of the overlay, constructed by the proposed protocol, to understand how it evolves during the streaming session with peer churn. Evaluation results show that the overlay is (1) scalable: it is stable with different sizes, from hundreds up to 10000 peers; and (2) robust: the good features are maintained or recovered fast under high peer churn rates, and it only becomes disconnected when more than 86% of the peers are removed from the network.
... Peer sampling services (PSS) have been widely used in large scale distributed applications, such as information dissemination [29], aggregation [30], and overlay topology management [31][32][33][34]. The main purpose of a PSS is to provide the particpating nodes with uniformly random sample of the nodes in the system. ...
... Gossiping algorithms in this case are usually biased towards finding neighbours that have interesting characteristics, e.g. peers with similar playback point and available upload capacity [35] or that are geographically closer to the requester [23]. Upon the discovery of peers, the partnership service is used to establish temporary peering connections with a subset of peers which is considered suitable for the receipt of the stream. ...
... In this dissertation, we present our P2P live media streaming solution in the form of three systems: gradienTv [1], Sepidar [2], and GLive [3]. In gradienTv and Sepidar, we build multiple approximately minimal height overlay trees for content delivery, whereas, in GLive, we build a mesh overlay, such that the average path length between nodes and the media source is approximately minimum. ...
... Therefore, meanwhile the child node, whose father failed, finds an appropriate parent for that stripe, the node misses the content of that stripe. Sepidar [2], gradienTv [1], Orchard [13], ChunkySpread [14], and CoopNet [15] are the solutions in this class. ...
... Many algorithms are proposed based on this model, e.g., NewCoolstreaming [24], DONet/-Coolstreaming [18], PULSE [20] and [21] use a gossip-generated random overlay network to search for the supplying nodes. We use the gossip-generated Gradient overlay [5] for node discovery in gradineTv [1], Sepidar [2], and GLive [3]. In the gossip-based method, each node periodically sends its data availability information to its neighbours, a partial view of nodes in the system, to enable them find appropriate suppliers, who possess data they are looking for. ...
