Much of the present P2P-IR literature is focused on distributed indexing structures. Within this paper, we present an approach based on the replication of peer data summaries via rumor spreading and multicast in a structured overlay.We will describe Rumorama, a P2P framework for similarity queries inspired by GlOSS and CORI and their P2P-adaptation, PlanetP. Rumorama achieves a hierarchization of PlanetP-like summary-based P2P-IR networks. In a Rumorama network, each peer views the network as a small PlanetP network with connections to peers that see other small PlanetP networks. One important aspect is that each peer can choose the size of the PlanetP network it wants to see according to its local processing power and bandwidth. Even in this adaptive environment, Rumorama manages to process a query such that the summary of each peer is considered exactly once in a network without churn. However, the actual number of peers to be contacted for a query is a small fraction of the total number of peers in the network.Within this article, we present the Rumorama base protocol, as well as experiments demonstrating the scalability and viability of the approach under churn.

The retrieval facilities of most peer-to-peer (P2P) systems are limited to queries based on a unique identifier or a small set of keywords. The techniques used for this purpose are hardly applicable for content based image retrieval (CBIR) in a P2P network. Furthermore, we will argue that the curse of dimensionality and the high communication overhead prevent the adaptation of multidimensional search trees or fast sequential scan techniques for P2P CBIR. In the present paper we will propose two compact data representations that can be distributed in a P2P network and used as the basis for a source selection. This allows for communicating with only a small fraction of all peers during query processing without deteriorating the result quality significantly. We will also present experimental results confirming our approach.

In peer-to-peer (P2P) networks, computers with equal rights form a logical (overlay) network in order to provide a common service that lies beyond the capacity of every single participant. Efficient similarity search is generally recognized as a frontier in research about P2P systems. One way to address it is using data source selection based approaches where peers summarize the data they contribute to the network, generating typically one summary per peer. When processing queries, these summaries are used to choose the peers (data sources) that are most likely to contribute to the query result. Only those data sources are contacted. There are two main contributions of this paper. We extend earlier work, adding a data source selection method for high-dimensional vector data, comparing different peer ranking schemes. More importantly, we present a method that uses progressive stepwise data exchange between peers to better each peer's summary and therefore improve the system's performance. 1

No abstract

In peer-to-peer (P2P) networks, computers with equal rights form a logical (overlay) network in order to provide a common service that lies beyond the capacity of every single participant. Efficient similarity search is generally recognized as a frontier in research about P2P systems. One way to address this issue is using data source selection based approaches where peers summarize the data they contribute to the network, generating typically one summary per peer. When processing queries, these summaries are used to choose the peers (data sources) that are most likely to contribute to the query result. Only those data sources are contacted.There are several contributions of this article. We extend earlier work, adding a data source selection method for high-dimensional vector data, comparing different peer ranking schemes. Furthermore, we present two methods that use progressive stepwise data exchange between peers to better each peer's summary and therefore improve the system's performance. We finally examine the effect of these data exchange methods with respect to load balancing.