Arie Segev

The representation of a temporal data model in the relational environment

In previous work, we introduced a data model and a query language for temporal data. The model was designed independently of any existing data model rather than an extension of one. This approach provided an insight into the special requirements for handling temporal data. In this paper, we discuss the implications of supporting such a model in the relational database environment. We show that the concept of a temporal relation needs to be introduced, and define a temporal normal form for it. We discuss several options for the representation of our models constructs in the relational context, and explain why we chose a particular representation. We also suggest the concept of a temporal relation family as a virtual view to treat multiple temporal relations together.

Component-based e-commerce: assessment of current practices and future directions

Component-based e-commerce technology is a recent trend towards resolving the e-commerce challenge at both system and application levels. Instead of delivering a system as a prepacked monolith system containing any conceivable feature, component-based systems consist of a lightweight kernel to which new features can be added in the form of components. In order to identify the central problems in component-based e-commerce and ways to deal with them, we investigate prototypes, technologies, and frameworks that will transcend the current state of the practice in Internet commerce. In this paper, we first discuss the current practices and trends in component-based electronic commerce based on the International Workshop on Component-based Electronic Commerce. Then, we investigate a number of research issues and future directions in component-based development for electronic commerce.

Optimization of join operations in horizontally partitioned database systems

This paper analyzes the problem of joining two horizontally partitioned relations in a distributed database system. Two types of semijoin strategies are introduced, local and remote. Local semijoins are performed at the site of the restricted relation (or fragment), and remote semijoins can be performed at an arbitrary site. A mathematical model of a semijoin strategy for the case of remote semijoins is developed, and lower bounding and heuristic procedures are proposed. The results of computational experiments are reported. The experiments include an analysis of the heuristics performance relative to the lower bounds, sensitivity analysis, and error analysis. These results reveal a good performance of the heuristic procedures, and demonstrate the benefit of using semijoin operations to reduce the size of fragments prior to their transmission. The algorithms for the case of remote semijoins were found to be superior to the algorithms for the case of local semijoins. In addition, we found that the estimation accuracy of the selectivity factors has a significant effect on the incurred communication cost.

A framework for query optimization in temporal databases

We investigate issues pertaining to query processing of temporal databases in a relational environment. Tuple-versioning of relations is the adopted method of temporal data representation. New operators are necessary in order to exploit the richer semantics of temporal queries. We define four types of temporal joins— theta-join, time intersection, time union and the event-join. Factors that affect processing strategies are discussed, especially the problem of estimating data selectivity for various temporal operations. Strategies for implementing the temporal equijoin operator is evaluated.

TOOSQL - A Temporal Object-Oriented Query Language

This paper proposes a temporal, object-oriented query language called TOOSQL which has an SQL-like syntax and compares it with other recent developments in the area. TOOSQL was designed to satisfy the criteria for temporal languages developed in previous work on temporal relational models and to extend this criteria to a language for an object-oriented model which has several inherent features that make it a more attractive basis for a temporal data model. TOOSQL differs from previous work and concurrent efforts in its support for corrections, schema updates, tune-sequence operations and user-defined operations. A grammar is developed as a means of formalizing TOOSQL. Finally, we conclude with a discussion of open issues.

Extended update functionality in temporal databases

This paper presents an extended update functionality in temporal databases. In temporal databases, the information is associated with several time dimensions that designate the validity of the information in the application domain as well as the database domain. The complexity of information, coupled with the fact that historical data is being kept in the database, requires the use of an update model that provides the user with high-level abstractions. In this paper we provide an enhanced schema language and an enhanced collection of update operation types that help the system designer and the user to cope with the added complexities of such a model. One of the major issues dealt with in this paper is the situation of simultaneous values of a single data item; this situation occurs when multiple values, valid at the same time, were assigned to a data item at different times over the database history. Unlike the fixed semantics in conventional and existing temporal database models, we provide a flexible mechanism to handle simultaneous values which also distinguishes between regular modifications and error corrections. The extended update functionality is part of an update model that is currently being implemented in a prototype for a simulation project in a hospitals training center. Issues related to the implementation of this functionality in various data models are discussed. In particular, a mapping of the basic primitive operation types to TSQL2, and suggestions for its augmentation are provided.

TOOA: A Temporal Object-Oriented Algebra

In this paper, we present a temporal, object-oriented algebra which serves as a formal basis for the query language of a temporal, object-oriented data model. Our algebra is a superset of the relational algebra in that it provides support for manipulating temporal objects, temporal types, type hierarchies and class lattices, multiple time-lines, and correction sequences in addition to supporting the five relational algebra operators. Graphs are used as the visual representations of both the schema and the object instances. The algebra provides constructs to modify and manipulate the schema graph and its extension, the object graph. The algebra operates on a collection or collections of objects and returns a collection of objects. This algebra is a first step in providing a formal foundation for query processing and optimizing in a temporal, object-oriented data model.

TALE: A Temporal Active Language and Execution Model

Complex applications in domains such as decision support systems and real time systems require a functionality that is achieved by combining the active and temporal database technologies. In this paper we present TALE, a Temporal Active Language and Execution model. TALE is a temporal active database programming language, combined with an execution model that enables a correct and efficient processing of operations. As such, TALE is a step in accommodating software engineering challenges in modern information systems. TALE primitives are presented using examples and an EBNF. The run-time control mechanism of the model is introduced and TALE properties, namely active and temporal capabilities, and reflective programming capabilities are discussed.

Selectivity estimation of temporal data manipulations

Abstract Temporal relations possess several characteristics that distinguish them from conventional snapshot relations. First, for each instance of the surrogate (entity) there is a set of time-ordered tuples. Second, surrogate instances may arrive and depart in some time-dependent manner. Third, the surrogate instance may arrive and depart more than once, thus creating gaps (null values) within its history. Lastly, the value of the temporal attribute may also be time-dependent. Conventional methods of estimation are incapable of providing good approximations of the cost of various temporal operations, even for those involving selections on a single relation. The problem is more acute in the case of join operations, because selectivities on time interval intersections have to be estimated. We propose a practical, yet theoretically sound model to characterize the changes of temporal relations. From this model, estimates of the cardinalities of various unary and binary operations are derived. Simulation results show that the proposed estimates are both robust and superior to conventional estimates.

Optimizing Voting-Type Algorithms for Replicated Data

The main objectives of data replication are improved availability and reduced communications cost for queries. Maintaining the various copies consistent, however, increases the communications cost incurred by updates. For a given degree of replication, the choice of a specific concurrency control algorithm can have a significant impact on the total communications cost. In this paper we present various models for analyzing and understanding the trade-offs between the potentially opposing objectives of maximum resiliency and minimum communications cost in the context of the quorum consensus class of algorithms. It is argued that an optimal vote assignment is one which meets given resiliency goals and yet incurs the least communications cost compared with all other alternative assignments. A mathematical model for vote assignment is developed, and optimal algorithms are presented. It is demonstrated that significant cost savings can be realized from these approaches.