Authentication Hierarchy in Distributed Deductive Databases Dorel S˘ avulea University of Craiova Department of Informatics Al.I. Cuza Street, No. 13, Craiova ROMANIA savulea@central.ucv.ro Nicolae Constantinescu University of Craiova Department of Informatics Al.I. Cuza Street, No. 13, Craiova ROMANIA nikyc@central.ucv.ro Abstract: Distributed deductive databases have become more and more popular in the past decade. Their impor- tance is mainly due to their low costs and the high level of protecting data. For such a database there must be secured all the remote database fragments and the infrastructure in order to provide a high security level and to avoid an impersonation. We propose an algorithm for providing such a security level with less resources. The algorithm is based on authenticating the rules using digital signature. Key–Words: Authentication; Distributed Deductive Databases; Digital Signature, Identity Prove, Deductive Database Fragmentation. 1 Introduction Distributed deductive database security systems have been an intensive field of research in the past decade. There have been added rules to the relational database systems for including deductive capabilities. A deduc- tive database system is a database system which can deduce or infer additional information based on rules and facts stored in the database [1]. In the mean time, a distributed database system is a collection of multi- ple, logically interrelated databases distributed over a computer network or over the Internet [2]. The process of distributing databases has a lot of advantages. They allow local autonomy of data so that the management is partitioned to different parts, being very important for voluminous databases [3]. More- over, such systems provide improved reliability due to replication of database and availability of data. It also provides fast access to local data and support for incremental growth. The performance is improved by splitting the queries of the database into subqueries enhancing query processing since the answers to each subquery are computed only against those portions of the database relevant to the subquery being generated in parallel. The process of evaluating these queries and subqueries independently improves the response time and reduces the resources cost. Using such a system is much cheaper due to the fact that it costs less to create a network of smaller computers with the power of a single large computer. The modularity is also a big advantage of the dis- tributed systems since they can be modified, added and removed from the distributed database without af- fecting other modules (systems) [6, 5]. The failure of a single module does not affect the others and nor the performance of the system. All the transactions of a distributed deductive system follow the A.C.I.D property: • Automaticity: the transaction takes place as whole or not at all; • Consistency: any changes to values in an in- stance are consistent with changes to other values in the same instance; a consistency constraint is a predicate on data which serves as a precondition, post-condition, and transformation condition on any transaction. • Isolation: each transaction sees a consistent database; the following degrees of isolation were originally described as degrees of consistency by Jim Gray [4]: • Durability: the ability of the system to recover committed transaction updates if either the sys- tem or the storage media fails. An authentication stage not only provides a high security level for the system due to the fact that the users are identified, but also it ensures integrity and non-repudiation for the rules or queries that user sends. We propose a method which signs each frag- ment of the query and their verification is made by different servers of the system. The servers are cho- sen according a dictionary of predicates. LATEST TRENDS on COMPUTERS (Volume I) ISSN: 1792-4251 238 ISBN: 978-960-474-201-1