This paper describes the overall design and architecture of the Timber XML database system currently being implemented at the University of Michigan. The system is based upon a bulk algebra for manipulating trees, and natively stores XML. New access methods have been developed to evaluate queries in the XML context, and new cost estimation and query optimization techniques have also been developed. We present performance numbers to support some of our design decisions. We believe that the key intellectual contribution of this system is a comprehensive set-at-a-time query processing ability in a native XML store, with all the standard components of relational query processing, including algebraic rewriting and a cost-based optimizer.

Current data models like the NF 2 model and object-oriented models support set-valued attributes. Hence, it becomes possible to have join predicates based on set comparison. This paper introduces and evaluates two main memory algorithms to evaluate efficiently this kind of join. More specifically, we concentrate on subset predicates. 1 Introduction Since the invention of relational database systems, tremendous effort has been undertaken in order to develop efficient join algorithms. Starting from a simple nested-loop join algorithm, the first improvement was the introduction of the merge join [1]. Later, the hash join [2, 7] and its improvements [20, 23, 28, 39] became alternatives to the merge join. For overviews see [27, 37] and for a comparison between the sort-merge and hash joins see [13, 14]. A lot of effort has also been spent on parallelizing join algorithms based on sorting [10, 25, 26, 34] and hashing [6, 12, 36]. Another important research area is the development of inde...

Indexing a class hierarchy, in order to efficiently search or update the objects of a class according to a (range of) value(s) of an attribute, impacts OODB performance heavily. For this indexing problem, most systems use the class hierarchy index (CH) technique of [15] implemented using B + -trees. Other techniques, such as those of [14,18,30], can lead to improved average-case performance but involve the implementation of new data-structures, which are not as well-understood parts of database technology as B + -trees. As a special form of external dynamic two-dimensional range searching, this OODB indexing problem is solvable within reasonable worst-case bounds [12]. Based on this insight, we have developed a technique, called indexing by class-division (CD), which we believe can be used as a practical alternative to CH. We present an optimized implementation and experimental validation of CD's average-case performance. The main advantages of the CD technique are: (1) CD...

Time of creation is one of the predominant (often implicit) clustering strategies found not only in Data Warehouse systems: line items are created together with their correspond-ing order, objects are created together with their subparts and so on. The newly created data is then appended to the existing data. We present a new join algorithm, called Diag-Join, which exploits time-of-creation cluster-ing. If we are able to take advantage of time-of-creation clustering, then the performance evaluation reveals the superiority of Diag-Join over standard join algorithms like block-wise nested-loop join, GRACE hash join, and index nested-loop join. We also present an analyti-cal cost model for Diag-Join. 1

\Gamma An important feature to be considered in the design of a Multimedia Database Systems (MMDBS) is content-based retrieval of images. Spatial features represent the spatial relationships among objects in an image. The salient objects (interesting objects) can be organized in an object hierarchy, based on object-oriented concepts. This paper proposes a new indexing scheme, called 2d-h trees, for content-based retrieval of images. This scheme organizes the representations of the spatial relationships among objects in images and the hierarchical relationships among objects efficiently for query optimization. Our performance analysis indicates that the 2D-h-tree is an efficient index scheme for content-based retrieval of images. 1 Introduction Current hardware technology enables us to acquire, store, manipulate and transmit various types of data such as image, audio, and video. This has stimulated a great deal of interest in Multimedia Database Systems (MMDBS). An important feature t...

Object-oriented databases (OODB) have received considerable attention in recent years. Their performance is a critical factor hindering their current use. Several indexing schemes have been proposed in the literature for enhancing OODB performance and they are briefly reviewed here. In this paper a new and uniform indexing scheme is proposed. This scheme is based on a single Btree and combines both the hierarchical and nested indexing schemes. The uniformity of this scheme enables compact and optimized code for dealing with a large range of queries on the one hand, and flexibility in adding and removing indexed paths on the other hand. The performance of the scheme is discussed and an extensive experimental analysis for the class-hierarchy case is presented. The results show the advantages of the scheme for small range, clustered sets queries. Copyright c fl1997 Elsevier Science Ltd Key words: Object-Oriented Databases, Indexing, Performance, B-Trees 1. INTRODUCTION Object-...

The efficient execution of queries in object-oriented databases requires the design of specific indexing techniques, to efficiently deal with predicates against nested attributes or against class inheritance hierarchies. Indexing techniques so far proposed can be classified into three groups: inheritance indexing techniques, whose goal is to support queries along inheritance hierarchies; aggregation indexing techniques, dealing with the efficient evaluation of nested predicates; integrated techniques. The aim of this paper is to analyze two techniques providing an integrated support, the path index and the nested-inherited index, with respect to traditional techniques, such as the multi-index and the inherited multi-index. The analysis is performed assuming that multi-valued attributes, as well as instances with null attribute values, are present in the database. For this purpose, the paper first presents the considered techniques. An extension of the path index, firstly defined in [6]...

This paper describes SciMMDB, a scientific multimedia database system designed for Polymer Science experiment data applications. SciMMDB maintains multimedia data of type video, text and picture (as well as other standard data types), and allows users to (1) form multimedia presentation streams about experiments, (2) define a presentation which organizes a number of presentation streams for playout purposes, and (3) control the playout of a presentation. To form video, text or picture streams, a presentation stream construction language is described. Queries in this language express spatial and temporal relationships involving content objects; and, efficient evaluation of such queries are facilitated by indexing structures. SciMMDB uses IB+trees for queries involving temporal intervals. In SciMMDB, presentation streams are formed into a presentation using the notion of presentation graphs. During the playout of a presentation, each presentation stream in the presentation is played out...

An indexing structure called triple-node hierarchy is proposed for enhancing query processing in object-oriented database systems. The proposed structure provides efficient support for object references along an aggregation hierarchy by maintaining direct mapping between objects of interested pairs of classes. The intermediate classes along the object path are maintained separately for update purpose. We show that the proposed structure can achieve better performance compared to the previously known methods. The superior performance is also demonstrated by a set of simulations based on a cost model that we have developed. With some modification, the proposed structure can also provide fast support for object navigation in an integration of aggregation and inheritance hierarchies. Our results show that the extended triple-node hierarchy performs better than the best previous method known to us. keywords: object-oriented databases, indexing, aggregation, inheritance, performance analysis...

Since the introduction of object base management systems (OBMS), many query optimization techniques tailored for object query languages have been proposed. They adapt known optimization techniques to the OBMS context, exploit special object-oriented features, or give solutions to problems specific to querying objects. Nonetheless, one of the most prominent features of object models --- namely class hierarchies --- have so far not been exploited for query optimization. The current paper proposes new optimization techniques for queries referring to classes integrated into a class hierarchy. The techniques are generic in the sense that we do not give a set of algebraic equivalences the optimizer has to apply, but instead try to provide the reader with a general understanding of how to exploit class hierarchies for query optimization purposes. We give general descriptions of the techniques as well as illustrating examples. Besides yielding considerable savings in terms of execution time, t...