Gossip-based communication protocols are appealing in large-scale distributed applications such as information dissemination, aggregation, and overlay topology management. This paper factors out a fundamental mechanism at the heart of all these protocols: the peer-sampling service. In short, this service provides every node with peers to gossip with. We promote this service to the level of a first-class abstraction of a large-scale distributed system, similar to a name service being a first-class abstraction of a local-area system. We present a generic framework to implement a peer-sampling service in a decentralized manner by constructing and maintaining dynamic unstructured overlays through gossiping membership information itself. Our framework generalizes existing approaches and makes it easy to discover new ones. We use this framework to empirically explore and compare several implementations of the peer-sampling service. Through extensive simulation experiments we show that-although all protocols provide a good quality uniform random stream of peers to each node locally-traditional theoretical assumptions about the randomness of the unstructured overlays as a whole do not hold in any of the instances. We also show that different design decisions result in severe differences from the point of view of two crucial aspects: load balancing and fault tolerance. Our simulations are validated by means of a wide-area implementation.

No abstract

A lot of recent research on content-based P2P searching for file-sharing applications has focused on exploiting semantic relations between peers to facilitate searching. To the best of our knowledge, all methods proposed to date suggest reactive ways to seize peers' semantic relations. That is, they rely on the usage of the underlying search mechanism, and infer semantic relations based on the queries placed and the corresponding replies received. In this paper we follow a different approach, proposing a proactive method to build a semantic overlay. Our method is based on an epidemic protocol that clusters peers with similar content. It is worth noting that this peer clustering is done in a completely implicit way, that is, without requiring the user to specify his preferences or to characterize the content of files he shares.

A lot of recent work has dealt with improving performance of content searching in peer-to-peer file sharing systems. In this paper we attack this problem by modifying the overlay topology describing the peer relations in the system. More precisely, we create a semantic overlay, linking nodes that are "semantically close", by which we mean that they are interested in similar documents. This semantic overlay provides the primary search mechanism, while the initial peer-to-peer system provides the fail-over search mechanism. We focus on implicit approaches for discovering semantic proximity. We evaluate and compare three candidate methods, and review open questions.

Abstract. We propose PolderCast, a P2P topic-based Pub/Sub system that is (a) fault-tolerant and robust, (b) scalable w.r.t the number of nodes interested in a topic and number of topics that nodes are interested in, and (c) fast in terms of dissemination latency while (d) attaining a low communication overhead. This combination of properties is provided by an implementation that blends deterministic propagation over maintained rings with probabilistic dissemination following a limited number of random shortcuts. The rings are constructed and maintained using gossiping techniques. The random shortcuts are provided by two distinct peer-sampling services: Cyclon generates purely random links while Vicinity produces interest-induced random links. We analyze PolderCast and survey it in the context of existing approaches. We evaluate PolderCast experimentally using real-world workloads from Twitter and Facebook traces. We use widely renowned Scribe [5] as a baseline in a number of experiments. Robustness with respect to node churn is evaluated through traces from the Skype superpeer network. We show that the experimental results corroborate all of the above properties in settings of up to 10K nodes, 10K topics, and 5K topics per-node.

Overlay networks are central to the operation of large-scale decentralized applications, be it Internet-scale P2P systems deployed in the wild or cloud applications running in a controlled-albeit large-scale-environment. A number of custom solutions exist for individual applications, each employing a tailormade mechanism to build and maintain its specific structure. This paper addresses the role of randomness in developing and maintaining such structures. Taking VICINITY, a generic overlay management framework based on self-organization, we explore tradeoffs between deterministic and probabilistic decision-making for structuring overlays. We come to the conclusion that a pinch of randomness may even be needed in overlay construction, but also that much randomness or randomness alone is not good either.

Abstract. The newscast model is a general approach for communication in large agent-based distributed systems. The two basic servicesmembership management and information dissemination-are implemented by the same epidemic-style protocol. In this paper we present the newscast model and report on experiments using a Java implementation. The experiments involve communication in a large, wide-area cluster computer. By analysis of the outcome of the experiments we demonstrate that the system indeed shows the scalability and dependability properties predicted by our previous theoretical and simulation results.

SUMMARYMuch research on content-based P2P searching for file-sharing applications has focused on exploiting semantic relations between peers to facilitate searching. Current methods suggest reactive ways to manage semantic relations: they rely on the usage of the underlying search mechanism, and infer semantic relationships based on the queries placed and the corresponding replies received. In this paper we follow a different approach, proposing a proactive method to build a semantic overlay. Our method is based on an epidemic protocol that clusters peers with similar content. Peer clustering is done in a completely implicit way, that is, without requiring the user to specify preferences or to characterize the content of files being shared. In our approach, each node maintains a small list of semantically optimal peers. Our simulation studies show that such a list is highly effective when searching files. The construction of this list through gossiping is efficient and robust, even in the presence of changes in the network. Copyright