This paper addresses the problem of building scalable semantic overlay networks. Our approach follows the principle of data independence by separating a logical layer, the semantic overlay for managing and mapping data and metadata schemas, from a physical layer consisting of a structured peer-to-peer overlay network for e#cient routing of messages. The physical layer is used to implement various functions at the logical layer, including attribute-based search, schema management and schema mapping management. The separation of a physical from a logical layer allows us to process logical operations in the semantic overlay using di#erent physical execution strategies. In particular we identify iterative and recursive strategies for the traversal of semantic overlay networks as two important alternatives. At the logical layer we support semantic interoperability through schema inheritance and semantic gossiping. Thus our system provides a complete solution to the implementation of semantic overlay networks supporting both scalability and interoperability.

this paper was to answer the question, "How robust are P2P search networks?" A good proportion of the paper first addresses the supporting question, "How do P2P search networks work?" This foundation is important given the pace and breadth of research. Where it is available, we emphasize work on robustness and flush out promising lines of investigation

Structured peer-to-peer (P2P) systems are considered as the next generation application backbone on the Internet. An important problem of these systems is load balancing in the presence of non-uniform data distributions. In this paper we propose a completely decentralized mechanism that in parallel addresses a local and a global load balancing problem: (1) balancing the storage load uniformly among peers participating in the network and (2) uniformly replicating different data items in the network while optimally exploiting existing storage capacity. Our approach is based on the P-Grid P2P system which is our variant of a structured P2P network. Problem (1) is solved by directly adapting the search structure to the data distribution. This may result in an unbalanced search structure, but we will show that the expected search cost in P-Grid in number of messages remains logarithmic under all circumstances.

Among the open problems in P2P systems, support for non-trivial search predicates, standardized query languages, distributed query processing, query load balancing, and quality of query results have been identified as some of the most relevant issues. This paper describes how range queries as an important non-trivial search predicate can be supported in a structured overlay network that provides O(log n) search complexity on top of a trie abstraction. We provide analytical results that show that the proposed approach is efficient, supports arbitrary granularity of ranges, and demonstrate that its algorithmic complexity in terms of messages is independent of the size of the queried ranges and only depends on the size of the result set. In contrast to other systems which provide evaluation results only through simulations, we validate the theoretical analysis of the algorithms with large-scale experiments on the PlanetLab infrastructure using a fully-fledged implementation of our approach.

In this paper we present and evaluate uncoordinated on-line algorithms for simultaneous storage and replication load-balancing in DHT-based peer-to-peer systems. We compare our approach with the classical balls into bins model, and point out the similarities but also the differences which call for new loadbalancing mechanisms specifically targeted at P2P systems. Some of the peculiarities of P2P systems, which make our problem even more challenging are that both the network membership and the data indexed in the network is dynamic, there is neither global coordination nor global information to rely on, and the load-balancing mechanism ideally should not compromise the structural properties and thus the search efficiency of the DHT, while preserving the semantic information of the data (e.g., lexicographic ordering to enable range searches).

Abstract. We investigate the search cost in terms of number of messages generated for routing queries in tree-based P2P structured systems including binary and k-ary tree structures with different arities and different degrees of imbalance in the tree shape. This work is motivated by the fact that k-ary balanced tree access structures can greatly reduce the number of hops for searching compared to the binary trees. We study to what extent the same fact is true when the tree-like structures for access in P2P environments are unbalanced. Another important issue related to P2P environments is how to build these structures in a self-organizing way. We propose a mechanism for constructing k-ary tree based decentralized access structure in a self-organizing way and based on local interactions only. The ability to search efficiently also on unbalanced k-ary trees opens interesting opportunities for load balancing as has been shown in earlier work on P-Grid, our approach to structured P2P systems. 1

GridVine is a semantic overlay infrastructure based on a peer-to-peer (P2P) access structure. Built following the principle of data independence, it separates a logical layer — in which data, schemas, and schema mappings are managed — from a physical layer consisting of a structured P2P network supporting decentralized indexing, key load-balancing, and efficient routing. The system is decentralized, yet fosters semantic interoperability through pair-wise schema mappings and query reformulation. GridVine’s heterogeneous but semantically related information sources can be queried transparently using iterative query reformulation. The authors discuss a reference implementation of the system and several mechanisms for resolving queries collaboratively.

Abstract not found

1 Introduction to P-Grid P-Grid[1] is a structured peer-to-peer (P2P) system based on a distributed hash table (DHT). An illustration of a P-Grid tree can be seen in Figure 1. Each peer has a path, a binary string that gives the position of a peer in the tree (paths are