Advanced Database Applications Research Articles

Towards the Maturity of Object-Relational Database Technology: Promises and Reality

Object-relational technology provides a significant increase in scalability and flexibility over the traditional relational databases. The additional object-relational features are particularly satisfying for advanced database applications that relational database systems have experienced difficulties. The key factor to the success of object-relational database systems is their performance. This paper aims to review the promises of Object-Relational database systems, examine the reality, and how their promises may be fulfilled through unification with the relational technology. To investigate the performance implications of using object-relational relative to relational technology, the query-oriented BUCKY benchmark has been previously applied to an early object-relational database system, i.e., Illustra 97. This paper presents the results obtained from implementing and running the BUCKY benchmark on Oracle 10g. The results acquired from the work described in this paper are compared with the results obtained in BUCKY benchmark. This study throws light on the functionality of object-relational databases, where object-relational technology has made improvements but some limitations are identified as well. In general, the performance of relational supersedes that of object-relational database system.

A Framework for Object-Oriented Data Mining Based on Higher-Order Logic Programming

Data mining discovers knowledge and useful information from large amounts of data stored in databases. With the increasing popularity of object-oriented database system in advanced database applications, it is significantly important to study the data mining methods for object-oriented database. This paper proposes that higher-order logic programming languages and techniques is very suitable for object-oriented data mining, and presents a framework for object-oriented data mining based on higher-order logic programming. Such a framework is inductive logic programming which adopts higher-order logic programming language Escher as knowledge representation formalism. In addition, Escher is a generalization of the attribute-value representation, thus many higher-order logic learners under this framework can be upgraded directly from corresponding propositional learners.

An Object-Oriented Modeling and Implementation of Spatio-Temporal Knowledge Discovery System

Spatiotemporal databases deal with changes to spatial objects with time. The Spatiotemporal knowledge discovery process involves the integration of such database systems with data warehousing, data mining and online-analytical processing technologies. The applications in this domain will process spatial, temporal and attribute data elements and use spatio-temporal relationships among these elements. These applications deal with the evolution of spatial objects and changes in their topological relationships, associations along with time. These advanced database applications require storing, management and processing of complex spatiotemporal data. In this paper we consider the classification and modeling requirements for spatiotemporal applications, System modeling, preprocessing spatiotemporal data, discovering spatiotemporal topological relationships and extension of apriori algorithm for mining spatiotemporal frequent predicates. The design of an appropriate system that can capture spatiotemporal features of datasets is discussed. Prototype implementation of the system is carried out on top of open source object relational spatial database management system called postgresql and postgis. The algorithms are experimented on historical cadastral datasets which are created using OpenJump. The results that are visualized using OpenJump software are presented.

Persistent Semi-Dynamic Ordered Partition Index

Similarity search is a popular paradigm in advanced database applications. In content based image retrieval (CBIR) for example, images are transformed into feature vectors, which are then used for similarity search via k-nearest-neighbor (k-NN) queries in the feature vector space. Clustering by building a disk resident index is one method to speed up the processing of k-NN queries. In the case of high-dimensional feature vectors the dimensionality curse results in a high degree of overlap among the minimum bounding rectangles of the index, which results in most pages of the index being accessed. This is especially detrimental to performance, since disk positioning time for random disk accesses is slow and improving only at a rate of 8% annually. We propose an alternative solution to indexing high-dimensional data, which takes advantage of increasing main memory sizes and the 40% annual improvement in disk transfer rates, More specifically we make the Ordered-Partition---OP-tree, which is a main memory resident index, persistent by writing it onto disk. We investigate the optimization of OP-tree parameters and compare its performance with the sequential scan method with and without Karhunen--Loeve transformation. We use serialization to compact the dynamically allocated nodes of the OP-tree in main memory, which form a linked list, into a contiguous area. The index can then be saved on disk as a single file and loaded into main memory by a single transfer. The original OP-tree is static, so we propose several methods to support the insertion of new points dynamically. We compare these methods from the viewpoints of time and space efficiency. We also study the effect of incrementally building the index with and without applying the Karhunen-Loeve transformation. We compare the processing time of k-NN queries on persistent OP-trees and SR-trees to demonstrate the viability of the proposed method. We use one synthetic and three real world datasets in our experiments.

Efficient index-based KNN join processing for high-dimensional data

In many advanced database applications (e.g., multimedia databases), data objects are transformed into high-dimensional points and manipulated in high-dimensional space. One of the most important but costly operations is the similarity join that combines similar points from multiple datasets. In this paper, we examine the problem of processing K- nearest neighbor similarity join ( KNN join). KNN join between two datasets, R and S, returns for each point in R its K most similar points in S. We propose a new index-based KNN join approach using the iDistance as the underlying index structure. We first present its basic algorithm and then propose two different enhancements. In the first enhancement, we optimize the original KNN join algorithm by using approximation bounding cubes. In the second enhancement, we exploit the reduced dimensions of data space. We conducted an extensive experimental study using both synthetic and real datasets, and the results verify the performance advantage of our schemes over existing KNN join algorithms.

Report on the 7 th ACM International Workshop on Web Information and Data Management (WIDM 2005)

The 7th ACM International Workshop on Web Information and Data Management (WIDM 2005) was held at the Hilton Bremen Hotel in Bremen (Germany), on November 5, 2005, in conjunction with the 14th ACM International Conference on Information and Knowledge Management (CIKM 2005). The main objective of the workshop was to bring together researchers, industrial practitioners, and developers to discuss how Web information can be extracted, stored, analyzed, and processed to provide useful knowledge to the end users for various advanced database and Web applications. WIDM 2005 was sponsored by ACM SIGIR and by GI (Gesellshaft für Informatik) and was in cooperation with ACM SIGMOD.

Report on the 6th ACM international workshop on web information and data management (WIDM 2004) held at CIKM 2004

The 6th ACM International Workshop on Web Information and Data Management (WIDM 2004) was held at the Hyatt Arlington Hotel, in Washington DC, on November 12-13, 2004, in conjunction with the 13th ACM International Conference on Information and Knowledge Management (CIKM 2004). The main objective of the workshop was to bring together researchers, industrial practitioners, and developers to discuss how Web information can be extracted, stored, analyzed, and processed to provide useful knowledge to the end users for various advanced database and Web applications. As for previous years, WIDM 2004 was sponsored by ACM SIGIR.

Alternative Approaches to Configuring Computing Labs

Introduction Within a matter of a few decades the computing function within the business, as well as non-business, environment has evolved from a support function to an essential function. One would be hard pressed to name an industry that does not rely on computer systems to support its operations and decision-making. This reliance has resulted in increased demand for personnel with computer science and management information systems backgrounds. While large colleges and universities already have had such programs in place for quite some time, many small U.S. and international colleges and universities are just undertaking such endeavors. Introducing such programs mandates that the institutions must put in place the necessary computing labs to support the programs being introduced. However, in creating such labs, the institutions usually encounter significant obstacles, such as budgetary constraints, lack of physical space to house the labs, conflicting needs of different courses, and rapid obsolescence of technology (Andleman et al., 2003; Belles & Miller, 2002; Dubose, 2000; Jones, 2000; Kumar, 2003; Lovgren, 2000; Walter, 1993; Wilson, 2002). The Management Information Systems (MIS) program at the University of Houston-Clear Lake (UHCL) faced the same daunting challenges when the program was introduced in late 90s. This paper presents the alternate computing lab configurations that were evolved to provide a cost-effective, flexible and versatile lab support for the MIS program within budgetary and space constraints. Computing Lab Support Challenges to UHCL/MIS Program University of Houston Clear Lake (UHCL) comprises four schools and offers graduate and undergraduate programs to more than 7000 students. The School of Business is the largest school in the university and includes a management information systems program, introduced in the mid 90s, as a major discipline. Since its inception, one of the challenges the UHCL/MIS department has been coping with is the lack of adequate dedicated computing labs. The problem resulted from budgetary constraints and lack of sufficient physical space to house the computing labs. A new building that is currently under construction will alleviate this problem and house four computing labs to support the MIS program. In the meantime, however, the MIS faculty and university computing personnel have had to devise some pragmatic solutions to provide computing lab support to the program in limited space and within budgetary constraints. The following section presents the three primary categories of graduate MIS courses at UHCL and describes the computing resources required by each category for the students to complete course assignments, exercises, projects, etc. These computing requirements provided the catalyst for different lab configurations. MIS Course Categories and their Computing Requirements The UHCL/MIS graduate courses can be grouped into the following three broad categories: (1) general computing courses, (2) advanced computing courses, and (3) special projects. General Computing Courses This category is comprised of courses that introduce students to a specific skill, technology or subject area. Typically, these courses don't have a pre-requisite in the same skill, technology or subject area; however, by themselves, directly or indirectly, they form pre-requisites to one or more courses. This category includes courses such as Introduction to Management Information Systems, Introduction to Computer Networks, Fundamentals of Database Design and Development, and Introduction to Business Applications Programming. Advanced Computing Courses The courses in this category include System Administration, Network Security, Advanced Database Applications, Database Administration, Advanced Business Applications Programming, and E-Commerce Applications. One of the distinguishing characteristics of these courses is that the students in these courses generally require administrative access and privileges on the workstations and servers to work on course assignments and projects. …

Flexible robust programming in distributed object systems

Distributed applications that access persistent objects must maintain object state consistency even when failures are encountered during the manipulation of such objects. The basic transaction model, which has been implemented by several systems to ensure consistent executions of distributed applications, is not flexible enough to meet the requirements of many complex distributed applications. This has also been recognized for advanced database applications and, as a result, extended transaction models have been developed. We argue that distributed applications that manipulate long-lived data can benefit from such transaction models. We take an approach which views the various transaction models as policies for building robust applications. Thus, we advocate that the system implement several transaction models. A robust application can be programmed in such a system using a combination of several transaction models to meet its consistency requirements. We use applications from the domain of computer-supported cooperative work to motivate such an approach. We also develop a set of system-level mechanisms which can be used to implement multiple transaction models in a uniform manner. These mechanisms are used to implement nested, split, and cooperating transaction models. A prototype system that has been implemented is described to demonstrate the feasibility of this approach.

A retrieval technique for high-dimensional data and partially specified queries

While the persistent data of many advanced database applications, such as OLAP and scientific studies, are characterized by very high dimensionality, typical queries posed on these data appeal to a small number of relevant dimensions. Unfortunately, the multi-dimensional access methods designed for high-dimensional data perform rather poorly for these partially specified queries. The retrieval technique proposed in this paper uses a combination of two complementary measures to support efficient processing of partial queries over high-dimensional data. First, an elaborate storage organization, called the inverted space, allows the system administrator to control the size of individual indexes in order to avoid the negative impact of extremely high data dimensionality on the retrieval performance. Second, a new indexing structure, which is designed to support the inverted-space storage organization, enables efficient query processing in projected spaces with moderate dimensionality. This indexing mechanism is a general-purpose point access method that effectively attacks the limitations of KDB-trees in spaces with many dimensions, while preserving the simplicity and relatively good performance of the later structure in low-dimensional spaces. The analytical and experimental results show that the new indexing scheme outperforms two other variants of KDB-trees investigated in the paper for both fully and partially specified queries.

Report on the third International Workshop on Web Information and Data Management (WIDM'2001)

The third International Workshop on Web Information and Data Management (WIDM'2001) was held at Doubletree Hotel, Buckhead in Atlanta on November 9, 2001. The workshop was sponsored by ACM SIGIR, ACM SIGMIS and E-Book Systems. Like it predecessors, WIDM'2001 was held in conjunction with the International Conference on Information and Knowledge Management (CIKM'2001).The objective of the workshop was to bring together researchers, industrial practitioners, and developers to study how the web information can be extracted, stored, analysed and processed to provide useful knowledge to the end users for various advanced database applications. The workshop received 34 submission of research papers, out of which 11 were accepted.

Effects of shaping characteristics on the performance of nested transactions

The nested transaction model was introduced to satisfy the requirements of advanced database applications. Moreover, it is currently the basic transaction model for new database applications like workflow systems and new database systems like mobile databases and object-relational databases. Though there are several performance evaluation studies of different concurrency control mechanisms in nested transactions, the effects of transaction parameters on the overall system performance have not received any attention. In this paper, we study the effects of transactions characteristics on system performance. We developed a detailed simulation model and conducted several experiments to measure the impact of transactions characteristics on the performance. First, the effect of the number of leaves on the performance of nested transactions is investigated under different shaping parameters. Also, effects of the depth of the transaction tree on the system performance are investigated.

Typed query languages for databases containing queries

This paper introduces and studies the relational meta algebra, a statically typed extension of the relational algebra to allow for meta programming in databases. In this meta algebra one can manipulate database relations involving not only stored data values (as in classical relational databases) but also stored relational algebra expressions. Topics discussed include modeling of advanced database applications involving “procedural data” ; desirability as well as limitations of a strict typing discipline in this context; equivalence with a first-order calculus; and global expressive power and non-redundancy of the proposed formalism.

Semantic serializability: A correctness criterion for processing transactions in advanced database applications

Serializability requires that the execution of each transaction must give the illusion to be an atomic action, i.e., the execution of a set of transactions must appear to be a serial one. This requirement, however, is too strong and unnecessarily restricts concurrency among transactions, when semantic information is available for the transaction processing mechanism. In this paper, a new correctness criterion for concurrent execution of database transactions, denoted semantic serializability, is proposed. Semantic serializability is based on the use of semantic information about database objects (and not about transactions). The main idea of our proposal is to provide different atomicity views for each transaction and, for this reason, to allow interleavings among transactions which are nonserializable, but which preserve database consistency. We develop two concurrency control protocols, which are based on semantic serializability. One protocol is based on a locking-mechanism and the other one uses a non-locking approach. Our proposal is suitable to a wide variety of advanced database applications, such as CAx, MDBS, GIS and WFMS.

Implementing ECA rules in an active database

Traditional database systems are designed to manage large volumes of data, but rule support is generally not a strong feature. On the other hand, expert systems have the deductive capability to manage rule processing. A coupling between these two systems to support advanced database applications results in what is termed an active database. An approach to implementing active database systems is to represent knowledge as Event–Condition–Action (ECA) rules. ECA rules determine when and how to react to different kinds of events. This paper focuses on the different type of events supported by an active database CLOSE. A set of primitive and composite events is presented together with examples of their application. Event detection in this system is also discussed.

Generalization-based data mining in object-oriented databases using an object cube model

Data mining is the discovery of knowledge and useful information from the large amounts of data stored in databases. With the increasing popularity of object-oriented database systems in advanced database applications, it is important to study the data mining methods for object-oriented databases because mining knowledge from such databases may improve understanding, organization, and utilization of the data stored there. In this paper, issues on generalization-based data mining in object-oriented databases are investigated in three aspects: (1) generalization of complex objects, (2) class-based generalization, and (3) extraction of different kinds of rules. An object cube model is proposed for class-based generalization, on-line analytical processing, and data mining. The study shows that (i) a set of sophisticated generalization operators can be constructed for generalization of complex data objects, (ii) a dimension-based class generalization mechanism can be developed for object cube construction, and (iii) sophisticated rule formation methods can be developed for extraction of different kinds of knowledge from data, including characteristic rules, discriminant rules, association rules, and classification rules. Furthermore, the application of such discovered knowledge may substantially enhance the power and flexibility of browsing databases, organizing databases and querying data and knowledge in object-oriented databases.

OSAM*.KBMS/P: a parallel, active, object-oriented knowledge base server

An active object-oriented knowledge base server can provide many desirable features for supporting a wide spectrum of advanced and complex database applications. Knowledge rules, which are used to define a variety of database tasks to be performed automatically on the occurrence of some events, often need much more sophisticated rule-specification and control mechanisms than the traditional priority-based mechanism to specify the control structural relationships and parallel execution properties among rules. The underlying object-oriented (OO) knowledge representation model must provide a means to model the structural relationships among data entities and the control structures among rules in a uniform fashion. The transaction execution model must provide a means to incorporate the execution of structured rules in a transaction framework. Also, a parallel implementation of an active knowledge base server is essential to achieve the needed efficiency in processing nested transactions and rules. In this paper, we present the architecture, implementation, and performance of a parallel active OO knowledge base server, which has the following features. First, the server is developed based on an extended OO knowledge representation model that models rules as objects and their control structural relationships as association types. This is analogous to the modeling of entities as objects and their structural relationships as association types. Thus, entities and rules, and their structures, can be uniformly modeled. Second, the server uses a graph-based transaction model that can naturally incorporate the control semantics of structured rules and guarantee the serializable execution of rules as subtransactions. Thus, the rule-execution model is uniformly integrated with that of transactions. Third, it uses an asynchronous parallel execution model to process the graph-based transactions and structured rules. This server, named OSAM*.KBMS/P, has been implemented on a shared-nothing multiprocessor system (nCUBE2) to verify and evaluate the proposed knowledge-representation model, graph-based transaction model, and asynchronous parallel-execution model.

The database and information system research group at the University of ULM

The University of Ulm was founded in 1967 with focus on medicine and natural sciences. In 1989 the University established two new faculties: Engineering Sciences and Computer Science. This enlargement took place within the framework of the so-called Science City Ulm . In a joint effort, the State of Baden-Württemberg, industrial companies, the University, and the City of Ulm successfully established a research and development infrastructure at or nearby the university campus consisting of the university's research labs, university-related research institutes like the Research Institute for Applied Knowledge Processing (FAW), and industrial research and development labs, especially a large research center of Daimler-Benz AG. Today, the Faculty of Computer Science consists of seven divisions (called 'departments'), each of which equipped with two professor positions: Theoretical Computer Science Artificial Intelligence Distributed Systems Databases and Information Systems Software Technology and Compiler Construction Computer Structures Neural Information Processing. The Dept. of Databases and Information Systems (DBIS) became operational at the beginning of 1990 when Peter Dadam joined the faculty. He came from the IBM Heidelberg Science Center (HDSC) where he managed the research department for Advanced Information Management (AIM). At the HDSC he was working on advanced database technology and applications and contributed to the development of the AIM-P system (see [1]). The second professor position was first occupied by Marc Scholl, who belonged to the DBIS department from 1992 to 1994. In 1996 Wolfgang Klas joined the DBIS department as second professor. He came from the GMD Institute for Integrated Publication and Information Systems (IPSI) where he managed the research division Distributed Multimedia Information Systems and was working on advanced object-oriented database systems technology, interoperable database systems, and multimedia information systems. At present, the DBIS team consists of the teaching and research assistants Thomas Bauer, Susanne Boll, Christian Heinlein, Clemens Hensinger, Erich Müller, Manfred Reichert, Birgit Schultheiβ, the system engineer Rudi Seifert, the secretary Christiane Köppl, and the doctoral students Thomas Beuter and Anita Krämer. In the following, we concentrate on the research and development work performed previously and presently in the research groups of Peter Dadam and of Wolfgang Klas. For references to Marc Scholl's work please visit http://www.informatik.uni-konstanz.de/dbis.

Relative Serializability: An Approach for Relaxing the Atomicity of Transactions

Serializability is too strong a correctness criterion and unnecessarily restricts concurrency. We use the semantic information of a transaction to provide different atomicity views of the transaction to other transactions. The proposed approach improves concurrency and allows interleavings among transactions which are nonserializable, but which nonetheless preserve the consistency of the database and are acceptable to the users. We develop a graph-based tool whose acyclicity is both a necessary and sufficient condition for the correctness of an execution. Our theory encompasses earlier proposals that incorporate semantic information of transactions. Furthermore, it is the first approach that provides an efficient graph-based tool for recognizing correct schedules without imposing any restrictions on the application domain. Our approach is widely applicable to many advanced database applications such as systems with long-lived transactions and collaborative environments.

Concurrency control in an object-oriented data base system

Object-oriented data bases are fast gaining in popularity, especially with advanced applications like computer aided design (CAD) and multimedia data bases (MMDB). The modeling techniques required by these applications cannot be met by conventional data base systems. The semantic richness of the object-oriented model facilitates the modeling of the advanced data base applications. These applications are characterized by long-duration cooperating transactions. Unlike the conventional data bases, serializability can no longer be the correctness criterion. This article describes our on-going research in the area of transaction management for object-oriented data bases. We have previously defined a new transaction model for object-oriented data bases that takes into consideration the unique requirements of the advanced applications. This article describes a new concurrency control protocol that satisfies the correctness criterion for a concurrent execution of transactions in an object-oriented data base environment.