Academic
Publications
Optimized information discovery using self-adapting indices over Distributed Hash Tables

Optimized information discovery using self-adapting indices over Distributed Hash Tables,10.1109/PCCC.2010.5682330,Faraz Memon,D. Tiebler,F. Dürr,K.

Optimized information discovery using self-adapting indices over Distributed Hash Tables  
BibTex | RIS | RefWorks Download
Distributed Hash Table (DHT)-based peer-to-peer information discovery systems have emerged as highly scalable systems for information storage and discovery in massively distributed networks. Originally DHTs supported only point queries. However, recently they have been extended to support more complex queries, such as multiattribute range (MAR) queries. Generally, the support for MAR queries over DHTs has been provided either by creating an individual index for each data attribute or by creating a single index using the combination of all data attributes. In contrast to these approaches, we propose to create and modify indices using the attribute combinations that dynamically appear in MAR queries in the system. In this paper, we present an adaptive information discovery system that adapts the set of indices according to the dynamic set of MAR queries in the system. The main contribution of this paper is a four-phase index adaptation process. Our evaluations show that the adaptive information discovery system continuously optimizes the overall system performance for MAR queries. Moreover, compared to a non-adaptive system, our system achieves several orders of magnitude improved performance.
Conference: Power Conversion Conference - PCCON , pp. 105-113, 2010
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.