The popularity of peer-to-peer multimedia file sharing applications such as Gnutella and Napster has created a flurry of recent research activity into peer-to-peer architectures. We believe that the proper evaluation of a peer-to-peer system must take into account the characteristics of the peers that choose to participate. Surprisingly, however, few of the peer-to-peer architectures currently being developed are evaluated with respect to such considerations. In this paper, we remedy this situation by performing a detailed measurement study of the two popular peer-to-peer file sharing systems, namely Napster and Gnutella. In particular, our measurement study seeks to precisely characterize the population of end-user hosts that participate in these two systems. This characterization includes the bottleneck bandwidths between these hosts and the Internet at large, IP-level latencies to send packets to these hosts, how often hosts connect and disconnect from the system, how many les hosts share and download, the degree of cooperation between the hosts, and several correlations between these characteristics. Our measurements show that there is significant heterogeneity and lack of cooperation across peers participating in these systems.

Abstract not found

Nearly all proposed DHTs have echoes -- either explicit or implicit -- of parallel interconnect networks such as butterfly, torus, hypercube, and de Bruijn graphs. However, unlike interconnection networks, DHTs define an overlay graph for all network sizes, and allow the overlay graph to evolve as nodes join and leave as participants. Most of the well-cited DHT designs obscured two basic concerns in DHT design: the choice of topology (the links and routes in "ideal" cases), and of interconnect "emulation" strategy (how they deal with dynamism and a sparsely filled identifier space). In this paper, we separate these concerns into two formal frameworks. We first observe several idealized DHT topologies from an algebraic standpoint, and discuss the utility of such an approach in creating and understanding new DHT topologies. We then examine several emulation schemes and consider their applicability to algebraic graphs. Given these pieces, we demonstrate a promising new DHT design that emerges from this separation of concerns: IHOP (Internet Hashing Over Pancake graphs), a Cayley graph of the symmetric group emulated with a scheme of Abraham et al.

Due to the growth of unresponsive UDP traffic in the Internet, it becomes increasingly important for ISPs to amply shape the traffic that leaves their network. Ideally, flows should be forced to be TCP-friendly; to this end, knowledge about certain end-to-end path characteristics is needed. We present a suitable mechanism (the Burst-PiggyBack (BPB) technique) that obtains the necessary information at a device that is located close to the sender without requiring any changes at the communicating peers or in routers. The method's hypothesis is that an injected probe packet at the end of a burst is treated similar to the burst. The results acquired from simulations showed strong correlations between the loss rate and the RTT of bursts and probe packets.