Mapping the Gnutella Network: Macroscopic Properties of Large-Scale Peer-to-Peer Systems

Abstract

Despite recent excitement generated by the peer-to-peer (P2P) paradigm and the surprisingly rapid deployment of some P2P applications, there are few quantitative evaluations of P2P systems behavior. The open architecture, achieved scale, and self-organizing structure of the Gnutella network make it an interesting P2P architecture to study. Like most other P2P applications, Gnutella builds, at the application level, a virtual network with its own routing mechanisms. The topology of this virtual network and the routing mechanisms used have a significant influence on application properties such as performance, reliability, and scalability. We have built a "crawler" to extract the topology of Gnutella's application level network. In this paper we analyze the topology graph and evaluate generated network traffic. Our two major findings are that: (1) although Gnutella is not a pure power-law network, its current configuration has the benefits and drawbacks of a power-law structure, and (2) the Gnutella virtual network topology does not match well the underlying Internet topology, hence leading to ineffective use of the physical networking infrastructure. These findings guide us to propose changes to the Gnutella protocol and implementations that may bring significant performance and scalability improvements. We believe that our findings as well as our measurement and analysis techniques have broad applicability to P2P systems and provide unique insights into P2P system design tradeoffs.

Exploring Small-World-Like Topologies Via SplitProber: Turning Power Laws into an Advantage in Unstructured Overlays

“…3. Utilization of the underlying network: a global property that is of interest to network service providers, measured by the metrics like the physical distance (latency) to the search results, and the mean stress, one of the most common definitions of the traffic load in overlay network [21].…”

Analysis of resource transfers in peer-to-peer file sharing applications using fluid models

“…The proposals presented in [23,24,28] studied traces of P2P applications such as Gnutella and Napster. These works focused on characterizing the overall P2P system, e.g., request patterns, traffic volume, traffic categorization and properties of shared online content as well as P2P structure and dynamics, e.g., connectivity and peer behaviors.…”

“…These works focused on characterizing the overall P2P system, e.g., request patterns, traffic volume, traffic categorization and properties of shared online content as well as P2P structure and dynamics, e.g., connectivity and peer behaviors. In particular the papers [23,24] presents a quantitative evaluations of P2P system behaviors and it is based on a "crawler" to extract topology of Gnutella application level network and brings to conclude that, although Gnutella is not a pure power-law network, its configuration has benefits and drawbacks of a power-law structure, e.g., resilience to random peer failures.…”

MuON: Epidemic based mutual anonymity in unstructured P2P networks

“…However, unstructured P2P networks pose significant challenges. For example Gnutella is known to consume high bandwidth [18]. Other studies [19] have shown that P2P systems exhibit high churn (changes in system membership); peers frequently leave/join the network and most peers are connected to the overlay for a short period of time.…”