-
[show abstract]
[hide abstract]
ABSTRACT: Peer-to-Peer multikeyword searching requires distributed intersection/union operations across wide area networks, raising a large amount of traffic cost. Existing schemes commonly utilize Bloom Filters (BFs) encoding to effectively reduce the traffic cost during the intersection/union operations. In this paper, we address the problem of optimizing the settings of a BF. We show, through mathematical proof, that the optimal setting of BF in terms of traffic cost is determined by the statistical information of the involved inverted lists, not the minimized false positive rate as claimed by previous studies. Through numerical analysis, we demonstrate how to obtain optimal settings. To better evaluate the performance of this design, we conduct comprehensive simulations on TREC WT10G test collection and query logs of a major commercial web search engine. Results show that our design significantly reduces the search traffic and latency of the existing approaches.
IEEE Transactions on Knowledge and Data Engineering 05/2012; · 1.66 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Grid computing emerges as effective technologies to couple geographically distributed resources and solve large-scale computational
problems in wide area networks. The fault tolerance is a significant and complex issue in grid computing systems. Various
techniques have been investigated to detect and correct faults in distributed computing systems. Unreliable fault detection
is one of the most effective techniques. Globus as a grid middleware manages resources in a wide area network. The Globus
fault detection service uses the well-known techniques based on unreliable fault detectors to detect and report component
failures. However, more powerful techniques are required to detect and correct both system-level and application-level faults
in a grid system, and a convenient toolkit is also needed to maintain the consistency in the grid. A fault-tolerant grid platform
(FTGP) based on an unreliable fault detector and the Globus fault detection service is presented in this paper. The platform
offers effective strategies in such three aspects as grid key components, user tasks, and high-level applications.
Journal of Computer Science and Technology 04/2012; 18(4):423-433. · 0.56 Impact Factor
-
IEEE Trans. Parallel Distrib. Syst. 01/2012; 23:232-241.
-
IEEE Trans. Knowl. Data Eng. 01/2012; 24:692-706.
-
SCIENCE CHINA Information Sciences. 01/2011; 54:1340-1351.
-
IEEE Trans. Parallel Distrib. Syst. 01/2011; 22:1042-1055.
-
IEEE Trans. Computers. 01/2010; 59:969-980.
-
[show abstract]
[hide abstract]
ABSTRACT: By combining an unstructured protocol with a DHT-based index, hybrid Peer-to-Peer (P2P) improves search efficiency in terms of query recall and response time. The key challenge in hybrid search is to estimate the number of peers that can answer a given query. Existing approaches assume that such a number can be directly obtained by computing item popularity. In this work, we show that such an assumption is not always valid, and previous designs cannot distinguish whether items related to a query are distributed in many peers or are in a few peers. To address this issue, we propose QRank, a difficulty-aware hybrid search, which ranks queries by weighting keywords based on term frequency. Using rank values, QRank selects proper search strategies for queries. We conduct comprehensive trace-driven simulations to evaluate this design. Results show that QRank significantly improves the search quality as well as reducing system traffic cost compared with existing approaches.
IEEE Transactions on Parallel and Distributed Systems 02/2009; · 1.40 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Geographical hash table (GHT) has been widely used to provide energy efficiency for data-centric storage in wireless sensor networks. Such a mechanism, however, suffers from high communication cost when we apply multi-dimensional event search in the network. In this work, we present MDS, a flexible, complete, and efficient multi-dimensional search mechanism atop traditional GHT based data-centric storage architecture. MDS utilizes bloom filters to reduce the communication cost of in-network intersection and union operations for multi-dimensional queries in wireless sensor networks. This scheme can be easily extended to support multi-dimensional range queries. Our mathematical analysis indicates the optimal settings for the bloom filters that maximize the traffic savings according to the information popularities. We conduct comprehensive simulations to evaluate our design. Results show that MDS achieves significant performance improvement in terms of energy consumptions and thus improves the applicability of the multi-dimensional search over the GHT based data-centric storage in sensor networks.
Real-Time Systems Symposium, 2008; 01/2009
-
9th IEEE/ACM International Symposium on Cluster Computing and the Grid, CCGrid 2009, Shanghai, China, 18-21 May 2009; 01/2009
-
Proceedings of the 29th IEEE Real-Time Systems Symposium, RTSS 2008, Barcelona, Spain, 30 November - 3 December 2008; 01/2008
-
Network and Parallel Computing, IFIP International Conference, NPC 2008, Shanghai, China, October 18-20, 2008. Proceedings; 01/2008
-
Proceedings of the 17th International Conference on World Wide Web, WWW 2008, Beijing, China, April 21-25, 2008; 01/2008
-
[show abstract]
[hide abstract]
ABSTRACT: By combining an unstructured protocol with a DHT-based global index, hybrid Peer-to-Peer (P2P) improves search efficiency in terms of query recall and response time. The key challenge in hybrid search is to estimate the number of peers that can answer a given query. Existing approaches assume that such a number can be directly obtained by computing item popularity. In this work, we show that such an assumption is not always valid, and previous designs cannot distinguish whether items related to a query are distributed in many peers or are in a few peers. To address this issue, we propose QRank, a difficulty-aware hybrid search, which ranks queries by weighting keywords based on term frequency. Using rank values, QRank selects proper search strategies for queries. We conduct comprehensive trace-driven simulations to evaluate this design. Results show that QRank significantly improves the search quality as well as reducing system traffic cost compared with existing approaches.
Parallel Processing, 2007. ICPP 2007. International Conference on; 10/2007
-
[show abstract]
[hide abstract]
ABSTRACT: The emerging grids need an efficient replica location mechanism. In the experience of developing ChinaGrid supporting platform (CGSP), a grid middleware that builds a uniform platform supporting multiple grid-based applications, we meet a challenge of utilizing the properties of locality in replica location process to construct a practical and high performance replica location mechanism. The key of the solution to this challenge is to design an efficient replica location algorithm that meets above requirements. Some previous works have been done to build a replica location mechanism, but they are not suitable for replica location in a grid environment with multiple applications like ChinaGrid. In this paper, we present a novel peer-to-peer algorithm for replica location mechanism, Boundary Chord, which has the merits of locality awareness, self-organization, and load balancing. Simulation results show that the algorithm has better performance than other structured peer-to-peer solutions to the replica location problem.
Parallel Architectures,Algorithms and Networks, 2005. ISPAN 2005. Proceedings. 8th International Symposium on; 01/2006
-
Proceedings of the 15th international conference on World Wide Web, WWW 2006, Edinburgh, Scotland, UK, May 23-26, 2006; 01/2006
-
Proceedings of the 2006 ACM Symposium on Applied Computing (SAC), Dijon, France, April 23-27, 2006; 01/2006
-
20th International Conference on Advanced Information Networking and Applications (AINA 2006), 18-20 April 2006, Vienna, Austria; 01/2006
-
Advanced Web and Network Technologies, and Applications, APWeb 2006 International Workshops: XRA, IWSN, MEGA, and ICSE, Harbin, China, January 16-18, 2006, Proceedings; 01/2006
-
The Semantic Web - ASWC 2006, First Asian Semantic Web Conference, Beijing, China, September 3-7, 2006, Proceedings; 01/2006
