Index recommendation tool for optimized information discovery over distributed hash tables

Abstract

Peer-to-peer (P2P) networks allow for efficient information discovery in large-scale distributed systems. Although point queries are well supported by current P2P systems -in particular systems based on distributed hash tables (DHTs) -, providing efficient support for more complex queries remains a challenge. Our research focuses on the efficient support for multiattribute range (MAR) queries over DHT-based information discovery systems. Traditionally, 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 a set of indices over selected attribute combinations. In order to limit the overhead induced by index maintenance, the total number of created indices has to be limited. Thus, the resulting problem is to create a limited number of indices such that the overall system performance is optimal for MAR queries. In this paper, we propose an index recommendation tool that implements heuristic solutions to this NP-hard problem. Our evaluations show that these heuristics lead to a close-to-optimal system performance for MAR queries.

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

“…During the first phase, a workload of MAR queries is collected from the DHT network using uniform random sampling of peers. This workload is then used in the second phase for obtaining a new set of indices using the index recommendation tool [15]. During the third phase the cost and the benefit of installing a new set of indices is estimated.…”

“…Given a workload of MAR queries and a limit o for the maximum number of indices, the index recommendation tool recommends a close-to-optimal set of indices I r for the given workload. For a detailed description of the index recommendation tool and the index recommendation algorithms, see [15].…”

“…In [15], we presented a tool that assists the designer of a distributed application in defining a useful set of indices for the third type of DHT indexing approach. Given a limit for the maximum number of indices and a representative set of MAR queries (workload), our tool recommends a set of indices that produces close-to-optimal system performance for the workload within the given limit.…”