Qixiang Sun

Statistical identification of encrypted Web browsing traffic

Encryption is often proposed as a tool for protecting the privacy of World Wide Web browsing. However, encryption-particularly as typically implemented in, or in concert with popular Web browsers-does not hide all information about the encrypted plaintext. Specifically, HTTP object count and sizes are often revealed (or at least incompletely concealed). We investigate the identifiability of World Wide Web traffic based on this unconcealed information in a large sample of Web pages, and show that it suffices to identify a significant fraction of them quite reliably. We also suggest some possible countermeasures against the exposure of this kind of information and experimentally evaluate their effectiveness.

YAPPERS: a peer-to-peer lookup service over arbitrary topology

Existing peer-to-peer search networks generally fall into two categories: Gnutella-style systems that use arbitrary topology and rely on controlled flooding for search, and systems that explicitly build an underlying topology to efficiently support a distributed hash table (DHT). In this paper, we propose a hybrid scheme for building a peer-to-peer lookup service over arbitrary network topology. Specifically, for each node in the search network, we build a small DHT consisting of nearby nodes and then provide an intelligent search mechanism that can traverse all the small DHTs. Our hybrid approach can reduce the nodes contacted for a lookup by an order of magnitude compared to Gnutella, allows rapid searching of nearby nodes through quick fan-out, does not reorganize the underlying overlay, and isolates the effect of topology changes to small areas for better scalability and stability.

A gossip-based reliable multicast for large-scale high-throughput applications

Group-based reliable multicast is an important building block for distributed applications. For large systems, however, traditional approaches do not scale well due to centralized recovery mechanisms and excessive message overhead. In this paper, we present a reliable probabilistic multicast, rpbcast, that is a hybrid of the centralized and gossip-based approaches. In particular, rpbcast extends previous work by supporting high packet rates and many active senders. Rpbcast uses gossip as the primary retransmission mechanism and only contacts loggers if gossips fail. Large groups of active senders are supported using negative gossip that specifies those messages a receiver is missing instead of those messages it received. Moreover, we show that negative gossip allows pull based recovery and converges faster than push based recovery. Rpbcast also applies hashing techniques to reduce message overhead and approximate group membership for garbage collection. We describe the key features of rpbcast and present simulation results.

Adlib: a self-tuning index for dynamic peer-to-peer systems

Peer-to-peer (P2P) systems enable queries over a large database horizontally partitioned across a dynamic set of nodes. We devise a self-tuning index for such systems that can trade off index maintenance cost against query efficiency, in order to optimize the overall system cost. The index, Adlib, dynamically adapts itself to operate at the optimal trade-off point, even as the optimal configuration changes with nodes joining and leaving the system. We use experiments on realistic workloads to demonstrate that Adlib can reduce the overall system cost by a factor of four.

Partial lookup services

Lookup services are used in many Internet applications to translate a key (e.g., a file name) into an associated set of entries (e.g., the location of file copies). The key lookups can often be satisfied by returning just a few entries instead of the entire set. However, current implementations of lookup services do not take advantage of this usage pattern. In this paper, we formalize the notion of a partial lookup service that explicitly supports returning a subset of the entries per lookup. We present four schemes for building a partial lookup service, and propose various metrics for evaluating the schemes. We show that a partial lookup service may have significant advantages over conventional ones in terms of space usage, fairness, fault tolerance, and other factors.

Maximizing remote work in flooding-based peer-to-peer systems

In peer-to-peer (P2P) systems where individual peers must cooperate to process each others requests, a useful metric for evaluating the system is how many remote requests are serviced by each peer. In this paper, we apply this remote work metric to study the searching aspect of flooding-based P2P networks such as Gnutella. We study how to maximize the remote work (query) in the entire network by controlling the rate of query injection at each node. In particular, we provide a simple procedure for finding the optimal rate of query injection and prove its optimality. We also show that a simple prefer-high-TTL protocol in which each peer processes only queries with the highest time-to-live (TTL) is optimal.

Maximizing Remote Work in Flooding-Based Peer-to-Peer Systems

In peer-to-peer (P2P) systems where individual peers must cooperate to process each other’s requests, a useful metric for evaluating the system is how many remote requests are serviced by each peer. In this paper we apply this remote work metric to flooding-based P2P search networks such as Gnutella. We study how to maximize the remote work in the entire network by controlling the rate of query injection at each node. In particular, we provide a simple procedure for finding the optimal rate of query injection and prove its optimality. We also show that a simple prefer-high-TTL protocol in which each peer processes only queries with the highest time-to-live (TTL) is optimal.