Efficient lookup on unstructured topologies

Abstract

We present LMS, a protocol for efficient lookup on unstructured networks. Our protocol uses a virtual namespace without imposing specific topologies. It is more efficient than existing lookup protocols for unstructured networks, and thus is an attractive alternative for applications in which the topology cannot be structured as a Distributed Hash Table (DHT).We present analytic bounds for the worst-case performance of our protocol. Through detailed simulations (with up to 100,000 nodes), we show that the actual performance on realistic topologies is significantly better. We also show in both simulations and a complete implementation (which includes over five hundred nodes) that our protocol is inherently robust against multiple node failures and can adapt its replication strategy to optimize searches according to a specific heuristic. Moreover, the simulation demonstrates the resilience of LMS to high node turnover rates, and that it can easily adapt to orders of magnitude changes in network size. The overhead incurred by LMS is small, and its performance approaches that of DHTs on networks of similar size.

Rumor Riding: Anonymizing Unstructured Peer-to-Peer Systems

“…5. When taking the T to be O(logn), the sowers are evenly distributed guaranteed by the observations in [21]. Our simulation results in Section V show that carefully selecting parameters will lead to desired collision distances.…”

“…According to the observations in [19], random walk achieves statistical properties similar to independent sampling for every reasonable network. Studies in [11,21] show that if the walk length is sufficiently large, the final receivers of a random walk query are randomly distributed. We define collision distance, T, as the hop count of the shorter path that rumors walk from the initiator to the sower, as illustrated in Fig.…”

Exploiting parallelism in the design of peer-to-peer overlays

“…In a random walk, if an incoming query can not be locally matched, the request is forwarded to a randomly selected neighbor, excluding the neighbor from which the request was received. Systems using random walk include Gia [27] and LMS [28].…”

Distributed Construction Protocols of Probabilistic Degree-Weighted Peer-to-Peer Overlays

“…These so called unstructured P2P systems usually adopt blind routing methods (e.g. flooding and random walk) that peers randomly disseminate messages to nearby peers [8], [9]. In such systems, the number of messages a peer receives in many cases greatly dependes on the number of its incoming links (in-degree).…”