Xuping Tu

The Content Pollution in Peer-to-Peer Live Streaming Systems: Analysis and Implications

There has been significant progress in the development and deployment of peer-to-peer (P2P) live video streaming systems. However, there has been little study on the security aspect in such systems. Our prior experiences in Anysee exhibit that existing systems are largely vulnerable to intermediate attacks, in which the content pollution is a common attack that can significantly reduce the content availability, and consequently impair the playback quality. This paper carries out a formal analysis of content pollution and discusses its implications in P2P live video streaming systems. Specifically, we establish a probabilistic model to capture the progress of content pollution. We verify the model using a real implementation based on Anysee system; we evaluate the content pollution effect through extensive simulations. We demonstrate that (1) the number of polluted peers can grow exponentially, similar to random scanning worms. This is vital that with 1% polluters, the overall system can be compromised within minutes; (2) the effective bandwidth utilization can be sharply decreased due to the transmission of polluted packets; (3) Augmenting the number of polluters does not imply a faster progress of content pollution, in which the most influential factors are the peer degree and access bandwidth. We further examine several techniques and demonstrate that a hash-based signature scheme can be effective against the content pollution, in particular when being used during the initial phase.

Nearcast: A locality-aware P2P live streaming approach for distance education

Peer-to-peer (P2P) live video streaming has been widely used in distance education applications to deliver the captured video courses to a large number of online students. By allowing peers serving each other in the network, P2P technology overcomes many limitations in the traditional client-server paradigm to achieve user and bandwidth scalabilities. However, existing systems do not perform well when the number of online students increases, and the system performance degrades seriously. One of the reasons is that the construction of the peer overlay in existing P2P systems has not considered the underlying physical network topology and can cause serious topology mismatch between the P2P overlay network and the physical network. The topology mismatch problem brings great link stress (unnecessary traffic) in the Internet infrastructure and greatly degrades the system performance. In this article, we address this problem and propose a locality-aware P2P overlay construction method, called Nearcast, which builds an efficient overlay multicast tree by letting each peer node choose physically closer nodes as its logical children. We have conducted extensive simulations to evaluate the performance of Nearcast in comparison with the existing RTT and NICE protocols. Also, Nearcast has been deployed on a wide-area network testbed to delivery video coursed to about 7200 users distributed across 100 collages in 32 cities in China. The experimental results show that Nearcast leads to lower link stress and shorter end-to-end latencies compared with the RTT and NICE protocols.

A community-based trust model for P2P networks

Trust management is a key issue for P2P networks. Previous works build peer’s reputation just based on ratings of other individual peers. In this paper we present a novel trust construction approach, called CommunityTrust. The approach evaluates peers’ reputation taking other peers’ ratings into account as well as recommendations of some related communities and peers. Our preliminary simulation results show that CommunityTrust model significantly improves the trust computation accuracy of P2P e-Commerce systems.

An Efficient Data Scheduling Scheme for P2P Storage-Constrained IPTV System

In a mesh-based peer-to-peer live-streaming system, a data scheduler decides which segments and from where it should fetch to the local buffer of a peer. The scheduling strategy is critical when the local buffer is limited for caching segments, particularly for a storage-constrained Internet Protocol Television system such as a set-top box. The main objective is to deliver data to as many peers as possible in a timely manner, i.e., streaming data should be received before their playback deadlines. One popular approach used in existing systems is to let peers preferentially request data segments that are rarest and of the earliest playback deadlines among neighboring peers. However, it has been ignored that a fresh segment with a higher sequence identity but a later playback deadline may have a longer time to be shared with other peers. In this paper, we propose a novel data scheduling scheme, called ColorStream, which can achieve higher system throughput and shorter start-up delay. When deciding which segments to get, in addition to rarity and urgency, ColorStream also considers the freshness of segments because distributing fresher ones implies to be shared by more peers in the future. It categorizes the segments and peers by labeling them with colors and lets each peer request with preference the rare and urgent segments that have the same color as its own. This technique allows more fresh segments to be requested without abandoning rare and urgent segments and also balances the workload of data dissemination among peers. Simulations have been conducted to evaluate the performance of the proposed ColorStream scheme. The results show that, compared with existing schemes, ColorStream can greatly improve the performance under various conditions in terms of throughput and start-up delay.

Service organisation and information consistency for service grid

As the emergence of Open Grid Service Architecture (OGSA), service grid is seen as a natural progression from computational grid. Because of the diversity and the complexity of the applications, it is a great challenge for the research of service grid platform. Currently, the popular information service systems are basically used to provide registry, publication, and management for computing resources, data resources, and network resources. However, such systems could not meet the requirement of service grid. An information service model, HowU information service, is proposed for service management and organisation. The service is regarded as the basic unit of information publication. A novel service naming and meta-data description mechanism is provided in this model. A grid user is not concerned with the resource details of a service, but its existence and current serving capability. In this paper, focus is on the information consistency of HowU information service about how to guarantee service existence in wide area and service information validity in organisation area.

Design and deployment of locality-aware overlay multicast protocol for live streaming services

This paper presents the design and deployment of a locality-aware overlay multicast protocol called Anysee. The key idea of Anysee is to use the geometrical information of end hosts to construct the locality-aware overlay data delivery tree such that nearby users in the underlying network can be organized into nearby subtrees. The prototype of Anysee has been widely used in CERNET. Logging traces obtained from broadcasting 2004 Athens Olympic Games over 16 days have shown that the performance of Anysee, such as end-to-end delay and absolute data delivery delay, significantly outperforms that of randomly constructed overlay multicast.

TCMM: hybrid overlay strategy for P2P live streaming services

This paper proposes an application level multicast approach called Tree-Control-Mesh-Media (TCMM) to distribute live media streams to a large number of users efficiently. In TCMM, transmissions of media data are controlled by two independent relay protocols in a collaborative manner. One protocol here is used to help a peer to identify its neighbor peers using the location information while the other one is used to deliver of media stream among the peers. The two protocols organize all peers into two graphs with different topologies that the communications can benefit a lot from the hybrid control topology. We have studied the performance of TCMM approach using different simulation cases. The experimental results have shown that the broadcasting performance of TCMM can achieve that of a well constructed mesh network while it can adapt more dynamic and irregular network environment. We also see that the penalty of introducing two protocols is rarely low which implies the high scalability of TCMM.

CIPS: coordinated intrusion prevention system

In this paper, we present the design and implementation of Coordinated Intrusion Prevention System (CIPS), which includes Parallel Firewall (PFW), Flow Detection (FD) and Multiple Intrusion Detection System (MIDS) to against large-scale or coordinated intrusions. The PFW consists of several firewalls working in parallel mainly by means of packet filtering, state inspection, and SYN proxy. The FD and MIDS detect and analyze the flow at the same time. The former one uses artificial neural network to analyze network traffic and detect flow anomaly. The latter one adopts traditional techniques such as protocol flow analysis and content-based virus detection to detect and prevent conventional intrusions and virus. Taking load balancing into account, CIPS also has Flow Scheduler (FS) for dispatching packets to each parallel component evenly. In addition, there is a Console & Manager (CM) aiming to reduce redundant alerts and to provide a feedback mechanism by alert clustering and to recognize the potential correlation rules among coordinated intrusion through mining large amounts of alerts.

An Efficient Distributed Transactional Memory System

Transactional memory (TM) is a parallel programming concept which reduces challenges in parallel programming. Existing distributed transactional memory system consumes too much bandwidth and brings high latency. In this work, we present Transactional Memory System for Cluster (Clustm), a generalized and scalable distributed transactional memory system. Our system addresses several open issues posed by this domain, including transactional memory consistency protocol, cache consistency protocol, and the distribution strategy of the metadata of shared data across the cluster. Then, we evaluate our design with several workloads, and the results demonstrate outstanding performance.

The core degree based tag reduction on chip multiprocessor to balance energy saving and performance overhead

Tag reduction is an approach to save energy of the cache system in a processor. Our previous work described that it can save more energy on a Chip Multiprocessor (CMP) than on a single-core processor. In this paper, we further investigate the problem on balancing energy saving and performance overhead when tag reduction is used for the low power Chip Multiprocessor (CMP). We first introduce the core degree concept which is defined as the number of cores that tag reduction can use for each thread. We then propose a core degree based tag approach that is to optimize the core degree such that the best balance of energy and performance can be achieved. In particular, as the basis for such optimization, the theoretical upper bounds of the energy savings and performance overhead are decided as function of the core degree. In our experiments, we use a 16-core CMP for example. In order to obtain the energy consumption and performance overhead with various core degrees, we construct an experimental environment, which is based on the Linux operating system. With the experimental environment, benchmarks of SPEC CPU2006 are used to evaluate our core degree based tag reduction. Finally, the experimental results show that the most desired balance of energy saving and performance overhead is achieved when core degree is set to 6.