ACM SIGMOD Record

Published by Association for Computing Machinery
Online ISSN: 0163-5808
Publications
Article
The database community is exploring more and more multidisciplinary avenues: Data semantics overlaps with ontology management; reasoning tasks venture into the domain of artificial intelligence; and data stream management and information retrieval shake hands, e.g., when processing Web click-streams. These new research avenues become evident, for example, in the topics that doctoral students choose for their dissertations. This paper surveys the emerging multidisciplinary research by doctoral students in database systems and related areas. It is based on the PIKM 2010, which is the 3rd Ph.D. workshop at the International Conference on Information and Knowledge Management (CIKM). The topics addressed include ontology development, data streams, natural language processing, medical databases, green energy, cloud computing, and exploratory search. In addition to core ideas from the workshop, we list some open research questions in these multidisciplinary areas.
 
Article
Complex queries are becoming commonplace, with the growing use of decision support systems. These complex queries often have a lot of common sub-expressions, either within a single query, or across multiple such queries run as a batch. Multi-query optimization aims at exploiting common sub-expressions to reduce evaluation cost. Multi-query optimization has hither-to been viewed as impractical, since earlier algorithms were exhaustive, and explore a doubly exponential search space. In this paper we demonstrate that multi-query optimization using heuristics is practical, and provides significant benefits. We propose three cost-based heuristic algorithms: Volcano-SH and Volcano-RU, which are based on simple modifications to the Volcano search strategy, and a greedy heuristic. Our greedy heuristic incorporates novel optimizations that improve efficiency greatly. Our algorithms are designed to be easily added to existing optimizers. We present a performance study comparing the algorithms, using workloads consisting of queries from the TPC-D benchmark. The study shows that our algorithms provide significant benefits over traditional optimization, at a very acceptable overhead in optimization time.
 
Article
Evaluating the relational join is one of the central algorithmic and most well-studied problems in database systems. A staggering number of variants have been considered including Block-Nested loop join, Hash-Join, Grace, Sort-merge for discussions of more modern issues). Commercial database engines use finely tuned join heuristics that take into account a wide variety of factors including the selectivity of various predicates, memory, IO, etc. In spite of this study of join queries, the textbook description of join processing is suboptimal. This survey describes recent results on join algorithms that have provable worst-case optimality runtime guarantees. We survey recent work and provide a simpler and unified description of these algorithms that we hope is useful for theory-minded readers, algorithm designers, and systems implementors.
 
Article
Good database design is crucial to obtain a sound, consistent database, and - in turn - good database design methodologies are the best way to achieve the right design. These methodologies are taught to most Computer Science undergraduates, as part of any Introduction to Database class. They can be considered part of the "canon", and indeed, the overall approach to database design has been unchanged for years. Moreover, none of the major database research assessments identify database design as a strategic research direction. Should we conclude that database design is a solved problem? Our thesis is that database design remains a critical unsolved problem. Hence, it should be the subject of more research. Our starting point is the observation that traditional database design is not used in practice - and if it were used it would result in designs that are not well adapted to current environments. In short, database design has failed to keep up with the times. In this paper, we put forth arguments to support our viewpoint, analyze the root causes of this situation and suggest some avenues of research.
 
4 summarizes the third level RAID characteristics and Figure 1.3 compares the sector layout and check disks for levels 2 and 3. 
Article
Increasing performance of CPUs and memories will be squandered if not matched by a similar performance increase in I/O. While the capacity of Single Large Expensive Disks (SLED) has grown rapidly, the performance improvement of SLED has been modest. Redundant Arrays of Inexpensive Disks (RAID), based on the magnetic disk technology developed for personal computers, offers an attractive alternative to SLED, promising improvements of an order of magnitude in performance, reliability, power consumption, and scalability. This paper introduces five levels of RAIDs, giving their relative cost/performance, and compares RAID to an IBM 3380 and a Fujitsu Super Eagle.
 
Article
Data warehouses are the core of decision support sys- tems, which nowadays are used by all kind of enter- prises in the entire world. Although many studies have been conducted on the need of decision support systems (DSSs) for small businesses, most of them adopt ex- isting solutions and approaches, which are appropriate for large-scaled enterprises, but are inadequate for small and middle-sized enterprises. Small enterprises require cheap, lightweight architec- tures and tools (hardware and software) providing on- line data analysis. In order to ensure these features, we review web-based business intelligence approaches. For real-time analysis, the traditional OLAP architecture is cumbersome and storage-costly; therefore, we also re- view in-memory processing. Consequently, this paper discusses the existing approa- ches and tools working in main memory and/or with web interfaces (including freeware tools), relevant for small and middle-sized enterprises in decision making.
 
Article
This is a thought piece on data-intensive science requirements for databases and science centers. It argues that peta-scale datasets will be housed by science centers that provide substantial storage and processing for scientists who access the data via smart notebooks. Next-generation science instruments and simulations will generate these peta-scale datasets. The need to publish and share data and the need for generic analysis and visualization tools will finally create a convergence on common metadata standards. Database systems will be judged by their support of these metadata standards and by their ability to manage and access peta-scale datasets. The procedural stream-of-bytes-file-centric approach to data analysis is both too cumbersome and too serial for such large datasets. Non-procedural query and analysis of schematized self-describing data is both easier to use and allows much more parallelism.
 
Article
Simple economic and performance arguments suggest appropriate lifetimes for main memory pages and suggest optimal page sizes. The fundamental tradeoffs are the prices and bandwidths of RAMs and disks. The analysis indicates that with today's technology, five minutes is a good lifetime for randomly accessed pages, one minute is a good lifetime for two-pass sequentially accessed pages, and 16 KB is a good size for index pages. These rules-of-thumb change in predictable ways as technology ratios change. They also motivate the importance of the new Kaps, Maps, Scans, and $/Kaps, $/Maps, $/TBscan metrics.
 
Article
Databases are increasingly being used to store multimedia objects such as maps, images, audio, and video. Storage and retrieval of these objects is accomplished using multidimensional index structures such as R*-trees and SS-trees. As dimensionality increases, query performance in these index structures degrades. This phenomenon, generally referred to as the dimensionality curse, can be circumvented by reducing the dimensionality of the data. Such a reduction is, however, accompanied by a loss of precision of query results. Current techniques such as QBIC use SVD transform-based dimensionality reduction to ensure high query precision. The drawback of this approach is that SVD is expensive to compute and, therefore, not readily applicable to dynamic databases. In this paper, we propose novel techniques for performing SVD-based dimensionality reduction in dynamic databases. When the data distribution changes considerably so as to degrade query precision, we recompute the SVD transform and incorporate it in the existing index structure. For recomputing the SVD-transform, we propose a novel technique that uses aggregate data from the existing index rather than the entire data. This technique reduces the SVD-computation time without compromising query precision. We then explore efficient ways to incorporate the recomputed SVD-transform in the existing index structure. These techniques reduce the computation time by a factor of 20 in experiments on color and texture image vectors. The error due to approximate computation of SVD is less than 10%.
 
Article
Users and developers are tapping into big, complex entity graphs for numerous applications. It is challenging to select entity graphs for a particular need, given abundant datasets from many sources and the oftentimes scarce information available for them. We propose methods to automatically produce preview tables for entity graphs, for compact presentation of important entity types and relationships. The preview tables assist users in attaining a quick and rough preview of the data. They can be shown in a limited display space for a user to browse and explore, before she decides to spend time and resources to fetch and investigate the complete dataset. We formulate several optimization problems that look for previews with the highest scores according to intuitive goodness measures, under various constraints on preview size and distance between preview tables. The optimization problem under distance constraint is NP-hard. We design a dynamic-programming algorithm and an Apriori-style algorithm for finding optimal previews. The experiments and user studies on Freebase demonstrated both the scoring measures' accuracy and the discovery algorithms' efficiency
 
Article
Random sampling is a standard technique for constructing (approximate) histograms for query optimization. However, any real implementation in commercial products requires solving the hard problem of determining "How much sampling is enough?" We address this critical question in the context of equi-height histograms used in many commercial products, including Microsoft SQL Server. We introduce a conservative error metric capturing the intuition that for an approximate histogram to have low error, the error must be small in all regions of the histogram. We then present a result establishing an optimal bound on the amount of sampling required for pre-specified error bounds. We also describe an adaptive page sampling algorithm which achieves greater efficiency by using all values in a sampled page but adjusts the amount of sampling depending on clustering of values in pages. Next, we establish that the problem of estimating the number of distinct values is provably difficult, but propose ...
 
Article
This paper presents a benchmark that concisely captures the data base requirements of a collection of Earth Scientists working in the SEQUOIA 2000 project on various aspects of global change research. This benchmark has the novel characteristic that it uses real data sets and real queries that are representative of Earth Science tasks. Because it appears that Earth Science problems are typical of the problems of engineering and scientific DBMS users, we claim that this benchmark represents the needs of this more general community. Also included in the paper are benchmark results for four example DBMSs, ARCINFO, GRASS, IPW and POSTGRES. 1. INTRODUCTION There have been numerous benchmarks oriented toward DBMS performance in a variety of application areas. Perhaps the most famous one, TP1 [ANON85] is oriented toward business data processing, and has spawned a collection of derivative benchmarks, the most recent being TPC-A, TPC-B and TPC-C. These benchmarks represent the typical needs o...
 
Article
This paper reports on the 1997 KRDB workshop [2] that took place in conjunction with VLDB'97 in Athens, Greece, and whose main topic was accessing heterogeneous information. The KRDB workshop series started in 1994 as an offspring of the Esprit project on Computational Logic (CompuLog). CompuLog had the goal to establish logic programming on a broader basis for systems development by adding software engineering ideas (modules, program transformation), a better understanding of semantics, esp. negation, and knowledge representation techniques for knowledge base management. Roughly speaking, CompuLog tried to marry the static aspects of a system with the dynamic part. The static part, consisting of schema and integrity constraints, can be analyzed with knowledge representation techniques. Logic programming contributes expressive languages for querying. In CompuLog, expressive description logics (DL) have been developed in which rich schemata can be modeled. DLs are predicative languages having only unary predicates (set membership) and binary predicates (relationships between elements). The traditional purpose of a DL is to find out whether one DL expression is subsumed by another. Reasoning with DL's can be made fruitful for checking, e.g., the consistency of schemas or the containment of views [3]. The CompuLog experience showed that there is considerable potential of crossfertilization between knowledge representation and databases. Since then, KRDB took place every year. By assembling program committees with researchers from the two fields we wanted to highlight that KRDB is a forum for exchange and assure that contributions are of interest to each side. Now, after four workshops, it is time to present the topics to a larger audience in both communities. In the remain...
 
Article
this report ([1] to [19]), and the proceedings are available electronically.
 
Article
this report, we review the presentations given at the DMDW workshop and present some open problems which we believe should be addressed by future research and whose solution could contribute to make data warehouse research more relevant to the practice.
 
Article
In data-integration systems, the queries supported by a mediator are affected by the queryprocessing limitations of the sources being integrated. Existing mediation systems employ a variety of mechanisms to describe the query-processing capabilities of sources. However, these systems do not compute the capabilities of the mediators based on the capabilities of the sources they integrate. In this paper, we propose a framework to capture a rich variety of queryprocessing capabilities and present algorithms to compute the set of mediator-supported queries based on the capability limitations of its sources. Our algorithms take into consideration a variety of query-processing techniques employed by mediators to enhance the set of supported queries, and they yield concise capability descriptions. By computing mediator query capabilities and representing them in the same way as those of data sources, we enable mediators to be used by other mediators, and we also make it easier for end users to know in advance which mediator queries are feasible.
 
Article
Introduction The design and tuning of new access methods (AMs) for nontraditional data types and application areas has always been more of a black art than a rigorous discipline. The designer can only rely on intuition to come up with an effective design; its evaluation and profiling require tedious instrumentation of complex AM code and a host of hand-written scripts. To address these issues, we developed amdb, a visual AM "debugging " tool to support the AM design and implementation process. It is based on the GiST (Generalized Search Tree, [HNP95]) framework for AM construction, which offers the designer an abstracted view of a tree-structured AM and factors out the mechanical aspects of an AM implementation, such as tree traversal, concurrency control and recovery. Amdb is a visual analysis, debugging and profiling tool for AMs that are written as extensions of libgist, a public-domain stand-alone C++ implementation of GiSTs. 2 System Features Amdb was dev
 
Article
Information applications are increasingly required to be distributed among numerous remote sites through both wireless and wired links. Traditional models of distributed computing are inadequate to overcome the communication barrier this generates and to support the development of complex applications. In this paper, we advocate an approach based on agents. Agents are software modules that encapsulate data and code, cooperate to solve complicated tasks, and run at remote sites with minimum interaction with the user. We define an agent-based framework for accessing mobile heterogeneous databases. We then investigate concurrency control and recovery issues and outline possible solutions. Agent-based computing advances database transaction and control flow management concepts and remote programming techniques. Keywords: agents, mobile computing, workflow, multidatabases, transactions, concurrency control, recovery 1 Introduction With the rapid development of networking, computing appli...
 
Article
Object-oriented database systems (OODBs) need efficient support for manipulation of complex objects. In particular, support of queries involving evaluations of set predicates is often required in handling complex objects. In this paper, we propose a scheme to apply signature file techniques, which were originally invented for text retrieval, to the support of set value accesses, and quantitatively evaluate their potential capabilities. Two signature file organizations, the sequential signature file and the bitsliced signature file, are considered and their performance is compared with that of the nested index for queries involving the set inclusion operator (`). We develop a detailed cost model and present analytical results clarifying their retrieval, storage, and update costs. Our analysis shows that the bitsliced signature file is a very promising set access facility in OODBs. 1 INTRODUCTION Advanced database application areas, such as computer aided design, office automation, and...
 
Article
The importance of multidimensional index structures to numerous emerging database applications is well established. However, before these index structures can be supported as access methods (AMs) in a "commercial-strength" database management system (DBMS), efficient techniques to provide transactional access to data via the index structure must be developed. Concurrent accesses to data via index structures introduce the problem of protecting ranges specified in the retrieval from phantom insertions and deletions (the phantom problem). This paper presents a dynamic granular locking approach to phantom protection in Generalized Search Trees (GiSTs), an index structure supporting an extensible set of queries and data types. The granular locking technique offers a high degree of concurrency and has a low lock overhead. Our experiments show that the granular locking technique (1) scales well under various system loads and (2) similar to the B-tree case, provides a significantly more effici...
 
Article
Although several access control policies can be devised for controlling access to information, all existing authorization models, and the corresponding enforcement mechanisms, are based on a specific policy (usually the closed policy). As a consequence, although different policy choices are possible in theory, in practice only a specific policy can be actually applied within a given system. However, protection requirements within a system can vary dramatically, and no single policy may simultaneously satisfy them all. In this paper we present a flexible authorization manager (FAM) that can enforce multiple access control policies within a single, unified system. FAM is based on a language through which users can specify authorizations and access control policies to be applied in controlling execution of specific actions on given objects. We formally define the language and properties required to hold on the security specifications and prove that this language can express all security specifications. Furthermore, we show that all programs expressed in this language (called FAM/CAM-programs) are also guaranteed to be consistent (i.e., no conflicting access decisions occur) and CAM-programs are complete (i.e., every access is either authorized or denied). We then illustrate how several well-known protection policies proposed in the literature can be expressed in the FAM/CAM language and how users can customize the access control by specifying their own policies. The result is an access control mechanism which is flexible, since different access control policies can all coexist in the same data system, and extensible, since it can be augmented with any new policy a specific application or user may require.
 
Article
The Time-Split B-tree is an integrated index structure for a versioned timestamped database. It gradually migrates data from a current database to an historical database, records migrating when nodes split. Records valid at the split time are placed in both an historical node and a current node. This implies some redundancy. Using both analysis and simulation, we characterize the amount of redundancy, the space utilization, and the record addition (insert or update) performance for a spectrum of different rates of insertion versus update. Three splitting policies are studied which alter the conditions under which either time splits or key space splits are performed. 1. INTRODUCTION A growing area of interest in the database community is in the support of multiversioned data [LoSa, AhSn, JeMR, Ston]. Multiversioned data, when updated, results in a new version of the data being created. Because the old version is retained, several versions of a record can exist, each appropriate to some...
 
Article
Recently, a number of extensions to the traditional transaction model have been proposed to support new information-intensive applications such as CAD/CAM and software development. However, these extended models capture only a subset of interactions that can be found in such applications, and represent only some of the points within the spectrum of interactions possible in competitive and cooperative environments. ACTA is a formalizable framework developed for characterizing the whole spectrum of interactions. The ACTA framework is not yet another transaction model, but is intended to unify the existing models. ACTA allows for specifying the structure and the behavior of transactions as well as for reasoning about the concurrency and recovery properties of the transactions. In ACTA, the semantics of interactions are expressed in terms of transactions' effects on the commit and abort of other transactions and on objects' state and concurrency status (i.e., synchronization state). Its ability to capture the semantics of previously proposed transaction models is indicative of its generality. The reasoning capabilities of this framework have also been tested by using the framework to study the properties of a new model that is derived by combining two existing transaction models.
 
Article
This paper describes testing of rule conditions and execution of rule actions in the Ariel active DBMS. The Ariel rule system is tightly coupled with query and update processing. Ariel rules can have conditions based on a mix of patterns, events, and transitions. For testing rule conditions, Ariel makes use of a discrimination network composed of a special data structure for testing single-relation selection conditions efficiently, and a modified version of the TREAT algorithm, called A-TREAT, for testing join conditions. The key modification to TREAT (which could also be used in the Rete algorithm) is the use of virtual ff-memory nodes which save storage since they contain only the predicate associated with the memory node instead of copies of data matching the predicate. The rule-action executor in Ariel binds the data matching a rule's condition to the action of the rule at rule fire time, and executes the rule action using the query processor. 1 Introduction Designers of database...
 
MANET Architecture SMHClient
Article
Mobile Ad-hoc Networks (MANET) is an emerging area of research. Most current work is centered on routing issues. This paper discusses the issues associated with data communication with MANET database systems. While data push and data pull methods have been previously addressed in mobile networks, the proposed methods do not handle the unique requirements associated with MANET. Unlike traditional mobile networks, all nodes within the MANET are mobile and battery powered. Existing wireless algorithms and protocols are insufficient primarily because they do not consider the mobility and power requirements of both clients and servers. This paper will present some of the critical tasks facing this research.
 
Article
Ad hoc querying is difficult on very large datasets, since it is usually not possible to have the entire dataset on disk. While compression can be used to decrease the size of the dataset, compressed data is notoriously difficult to index or access. In this paper we consider a very large dataset comprising multiple distinct time sequences. Each point in the sequence is a numerical value. We show how to compress such a dataset into a format that supports ad hoc querying, provided that a small error can be tolerated when the data is uncompressed. Experiments on large, real world datasets ( AT&T customer calling patterns) show that the proposed method achieves an average of less than 5% error in any data value after compressing to a mere 2.5% of the original space ( i.e. , a 40:1 compression ratio), with these numbers not very sensitive to dataset size. Experiments on aggregate queries achieved a 0.5% reconstruction error with a space requirement under 2%.
 
Article
Caching approximate values instead of exact values presents an opportunity for performance gains in exchange for decreased precision. To maximize the performance improvement, cached approximations must be of appropriate precision: approximations that are too precise easily become invalid, requiring frequent refreshing, while overly imprecise approximations are likely to be useless to applications, which must then bypass the cache. We present a parameterized algorithm for adjusting the precision of cached approximations adaptively to achieve the best performance as data values, precision requirements, or workload vary. We consider interval approximations to numeric values but our ideas can be extended to other kinds of data and approximations. Our algorithm strictly generalizes previous adaptive caching algorithms for exact copies: we can set parameters to require that all approximations be exact, in which case our algorithm dynamically chooses whether or not to cache each data value. We have implemented our algorithm and tested it on synthetic and real-world data. A number of experimental results are reported, showing the effectiveness of our algorithm at maximizing performance, and also showing that in the special case of exact caching our algorithm performs as well as previous algorithms. In cases where bounded imprecision is acceptable, our algorithm easily outperforms previous algorithms for exact caching.
 
Article
Aggregation and duplicate removal are common in SQL queries. However, in the parallel query processing literature, aggregate processing has received surprisingly little attention; furthermore, for each of the traditional parallel aggregation algorithms, there is a range of grouping selectivities where the algorithm performs poorly. In this work, we propose new algorithms that dynamically adapt, at query evaluation time, in response to observed grouping selectivities. Performance analysis via analytical modeling and an implementation on a workstation-cluster shows that the proposed algorithms are able to perform well for all grouping selectivities. Finally, we study the effect of data skew and show that for certain data sets the proposed algorithms can even outperform the best of traditional approaches. 1 Introduction SQL queries are replete with aggregate and duplicate elimination operations. One measure of the perceived importance of aggregation is that in the proposed TPCD benchmark...
 
Article
A spatial distance join is a relatively new type of operation introduced for spatial and multimedia database applications. Additional requirements for ranking and stopping cardinality are often combined with the spatial distance join in on-line query processing or internet searchenvironments. These requirements pose new challenges as well as opportunities for more ecient processing of spatial distance join queries. In this paper, we rst present an ecient #-distance join algorithm that uses spatial indexes such as R-trees. Bidirectional node expansion and plane-sweeping techniques are used for fast pruning of distant pairs, and the planesweeping is further optimized bynovel strategies for selectingasweeping axis and direction. Furthermore, we propose adaptivemulti-stage algorithms for #-distance join and incremental distance join operations. Our performance study shows that the proposed adaptive multi-stage algorithms outperform previous work by up to an order of magnitude for both #-d...
 
Architecture of the Tukwila information integration system.
Even counting the cost of repeated materialization, interleaved planning and execution runs faster than a fully pipelined, static plan.
Article
Query processing in data integration occurs over networkbound, autonomous data sources. This requires extensions to traditional optimization and execution techniques for three reasons: there is an absence of quality statistics about the data, data transfer rates are unpredictable and bursty, and slow or unavailable data sources can often be replaced by overlapping or mirrored sources. This paper presents the Tukwila data integration system, designed to support adaptivity at its core using a two-pronged approach. Interleaved planning and execution with partial optimization allows Tukwila to quickly recover from decisions based on inaccurate estimates. During execution, Tukwila uses adaptive query operators such as the double pipelined hash join, which produces answers quickly, and the dynamic collector, which robustly and efficiently computes unions across overlapping data sources. We demonstrate that the Tukwila architecture extends previous innovations in adaptive execution (such as...
 
Phases of the OO7 Test Application.
Example of the OO7 Database Structure. the test database ranges from approximately 3.7 to 7.9 megabytes in size. This range allows us to run the numerous simulations required to understand the repeatability of our results. We have also experimented with applications running on a database up to 17 megabytes in size and have observed behavior consistent with the results reported in Section 4. Figure 3 is a depiction of the structure of the OO7 database as it appears at some point during the execution of an example trace. The top level of the database is a tree hierarchy that leads to a number of composite part objects. Composite parts are composed of atomic parts and their connections, which are subordinate to the composite part. These atomic parts are highly interconnected, with an average connectivity of three. The connections highlighted in the gure, together with their associated atomic part and connection objects, form an object cluster that can be detached from the rest of the graph by overwriting just six pointers.  
EEectiveness of SAIO Policy as a Function of the Requested I/O Percentage.  
EEectiveness of SAGA Policy as a Function of the Requested Garbage Percentage.  
Article
A fundamental problem in automating object database storage reclamation is determining how often to perform garbage collection. We show that the choice of collection rate can have a significant impact on application performance and that the "best" rate depends on the dynamic behavior of the application, tempered by the particular performance goals of the user. We describe two semi-automatic, selfadaptive policies for controlling collection rate that we have developed to address the problem. Using tracedriven simulations, we evaluate the performance of the policies on a test database application that demonstrates two distinct reclustering behaviors. Our results show that the policies are effective at achieving user-specified levels of I/O operations and database garbage percentage. We also investigate the sensitivity of the policies over a range of object connectivities. The evaluation demonstrates that semi-automatic, self-adaptive policies are a practical means for flexibly controllin...
 
Article
Recently we have proposed an adaptive, random sampling algorithm for general query size estimation. In earlier work we analyzed the asymptotic efficiency and accuracy of the algorithm; in this paper we investigate its practicality as applied to selects and joins. First, we extend our previous analysis to provide significantly improved bounds on the amount of sampling necessary for a given level of accuracy. Next, we provide "sanity bounds" to deal with queries for which the underlying data is extremely skewed or the query result is very small. Finally, we report on the performance of the estimation algorithm as implemented in a host language on a commercial relational system. The results are encouraging, even with this loose coupling between the estimation algorithm and the DBMS. 1 Introduction Estimates of query result size are useful in query optimization, as a means of determining the feasibility of queries, and as a quick way of answering queries for which the size of the answer i...
 
Article
In large federated and shared-nothing databases, resources can exhibit widely fluctuating characteristics. Assumptions made at the time a query is submitted will rarely hold throughout the duration of query processing. As a result, traditional static query optimization and execution techniques are ineffective in these environments. In this paper we introduce a query processing mechanism called an eddy , which continuously reorders operators in a query plan as it runs. We characterize the moments of symmetry during which pipelined joins can be easily reordered, and the synchronization barriers that require inputs from different sources to be coordinated. By combining eddies with appropriate join algorithms, we merge the optimization and execution phases of query processing, allowing each tuple to have a flexible ordering of the query operators. This flexibility is controlled by a combination of fluid dynamics and a simple learning algorithm. Our initial implementation demonstrates promising results, with eddies performing nearly as well as a static optimizer/executor in static scenarios, and providing dramatic improvements in dynamic execution environments.
 
Article
Several unifications that the application development process has long needed are now occurring, due to developments in object technologies and standards. These will (gradually) change the way data intensive applications are developed, reduce databases' prominence in this process, and change data administration's goals and participants. At the same time, the database community needs to ensure that its experiences are leveraged and its concerns are met within the new methodologies and toolsets. We discuss these issues, and illustrate how they apply to a portion of the Department of Defense (DOD). We also examine things that object technology won't accomplish, and identify research problems whose solution would enable further progress. 1. Introduction Object technology will have evolutionary and revolutionary effects on data administration. Evolutionary changes include new descriptive constructs (e.g., user-defined datatypes) and new reuse mechanisms such as inheritance. This note focu...
 
Shared-Memory Architecture
Shared-Disk Architecture
Article
In a 'shared-nothing' parallel computer, each processor has its own memory and disks and processors communicate by passing messages through an interconnect. Many academic researchers, and some vendors, assert that shared-nothingness is the 'consensus' architecture for parallel DBMSs. This alleged consensus is used as a justification for simulation models, algorithms, research prototypes and even marketing campaigns. We argue that shared-nothingness is no longer the consensus hardware architecture and that hardware resource sharing is a poor basis for categorising parallel DBMS software architectures if one wishes to compare the performance characteristics of parallel DBMS products.
 
Article
this paper, I will summarize these directions and provide the reader with pointers to other publications for further information.
 
Article
s Service (CAS) registry number (7439-97-6), "raw" CAS number (dashes removed---7439976), and common name ("quicksilver "). Within an ontology, each class has one or more slots, where each slot has a conceptual domain name with a canonical value domain over which all agents communicate when referring to that slot. This mapping problem has several manifestations, which we relate roughly in the order they have been addressed in the literature (insofar as they are addressed at all): ffl Traditional: Mapping between schemas can take place by imposing a view on that information and/or defining functions to translate the data from one value domain to another, as is currently done in relational databases, e.g., [9]. This type of mapping provides little support for semantics, but rather relies on the structure of the data. ffl Ontology-based: A common ontology is defined with well-specified semantics for the concepts it describes. Mapping between a schema and an ontology is done on a semantic...
 
Article
Introduction The amount of services and deployed software agents in the most famous offspring of the Internet, the World Wide Web, is exponentially increasing. In addition, the Internet is an open environment, where information sources, communication links and agents themselves may appear and disappear unpredictably. Thus, an effective, automated search and selection of relevant services or agents is essential for human users and agents as well. We distinguish three general agent categories in the Cyberspace, service providers, service requester, and middle agents. Service providers provide some type of service, such as finding information, or performing some particular domain specific problem solving. Requester agents need provider agents to perform some service for them. Agents that help locate others are called middle agents[2]. Matchmaking is the process of finding an appropriate provider for a requester thr
 
The Data Cube lattice
Article
The Relational On-Line Analytical Processing (ROLAP) is emerging as the dominant approach in data warehousing with decision support applications. In order to enhance query performance, the ROLAP approach relies on selecting and materializing in summary tables appropriate subsets of aggregate views which are then engaged in speeding up OLAP queries. However, a straight forward relational storage implementation of materialized ROLAP views is immensely wasteful on storage and incredibly inadequate on query performance and incremental update speed. In this paper we propose the use of Cubetrees, a collection of packed and compressed R-trees, as an alternative storage and index organization for ROLAP views and provide an efficient algorithm for mapping an arbitrary set of OLAP views to a collection of Cubetrees that achieve excellent performance. Compared to a conventional (relational) storage organization of materialized OLAP views, Cubetrees offer at least a 2-1 storage reduction, a 10-1 b...
 
Article
Finding approximate answers to multi-dimensional range queries over real valued attributes has significant applications in data exploration and database query optimization. In this paper we consider the following problem: given a table of d attributes whose domain is the real numbers, and a query that specifies a range in each dimension, find a good approximation of the number of records in the table that satisfy the query. We present a new histogram technique that is designed to approximate the density of multi-dimensional datasets with real attributes. Our technique finds buckets of variable size, and allows the buckets to overlap. Overlapping buckets allow more efficient approximation of the density. The size of the cells is based on the local density of the data. This technique leads to a faster and more compact approximation of the data distribution. We also show how to generalize kernel density estimators, and how to apply them on the multi-dimensional query approximation problem. Finally, we compare the accuracy of the proposed techniques with existing techniques using real and synthetic datasets.
 
Article
Computing multiple related group-bys and aggregates is one of the core operations of On-Line Analytical Processing (OLAP) applications. Recently, Gray et al. [GBLP95] proposed the "Cube" operator, which computes group-by aggregations over all possible subsets of the specified dimensions. The rapid acceptance of the importance of this operator has led to a variant of the Cube being proposed for the SQL standard. Several efficient algorithms for Relational OLAP (ROLAP) have been developed to compute the Cube. However, to our knowledge there is nothing in the literature on how to compute the Cube for Multidimensional OLAP (MOLAP) systems, which store their data in sparse arrays rather than in tables. In this paper, we present a MOLAP algorithm to compute the Cube, and compare it to a leading ROLAP algorithm. The comparison between the two is interesting, since although they are computing the same function, one is value-based (the ROLAP algorithm) whereas the other is position-based (the M...
 
Error tree for N = 8. may be involved in the approximation of several of the 2 d 0 values, so the CPU time complexity is O(2 d 0 d 0 k), whereas in our new approach the CPU time complexity for the standard implementation is only O(d 0 k) with a very small constant factor of proportionality (roughly 2) in terms of the number of arithmetic operations. (A more complicated approach yields an O(d 0 k)-time algorithm for the former case, but only if the logarithm transform is not used.) To fully understand the on-line query processing algorithm, we examine the relationship between wavelet coeecients and a range-sum query by using the error tree structure introduced in MVW98]. The error tree is built based upon the wavelet transform procedure. Figure 1 is the error tree for the example in Section 5.1. Each internal node is associated with a wavelet coeecient value, and each leaf is associated with an original signal value. (For purposes of exposition, the wavelet coeecients are unnormalized, but in the implementation the values are normalized and the algorithm is modiied appropriately.) Internal nodes and leaves are labeled sepa
Article
Computing multidimensional aggregates in high dimensions is a performance bottleneck for many OLAP applications. Obtaining the exact answer to an aggregation query can be prohibitively expensive in terms of time and/or storage space in a data warehouse environment. It is advantageous to have fast, approximate answers to OLAP aggregation queries. In this paper, we present a novel method that provides approximate answers to high-dimensional OLAP aggregation queries in massive sparse data sets in a time-efficient and space-efficient manner. We construct a compact data cube , which is an approximate and space-efficient representation of the underlying multidimensional array, based upon a multiresolution wavelet decomposition. In the on-line phase, each aggregation query can generally be answered using the compact data cube in one I/O or a smalll number of I/Os, depending upon the desired accuracy. We present two I/O-efficient algorithms to construct the compact data cube for the important case of sparse high-dimensional arrays , which often arise in practice. The traditional histogram methods are infeasible for the massive high-dimensional data sets in OLAP applications. Previously developed wavelet techniques are efficient only for dense data. Our on-line query processing algorithm is very fast and capable of refining answers as the user demands more accuracy. Experiments on real data show that our method provides significantly more accurate results for typical OLAP aggregation queries than other efficient approximation techniques such as random sampling.
 
Article
Aggregation in traditional database systems is performed in batch mode: a query is submitted, the system processes a large volume of data over a long period of time, and, eventually, the final answer is returned. This archaic approach is frustrating to users and has been abandoned in most other areas of computing. In this paper we propose a new online aggregation interface that permits users to both observe the progress of their aggregation queries and control execution on the fly. After outlining usability and performance requirements for a system supporting online aggregation, we present a suite of techniques that extend a database system to meet these requirements. These include methods for returning the output in random order, for providing control over the relative rate at which different aggregates are computed, and for computing running confidence intervals. Finally, we report on an initial implementation of online aggregation in postgres. 1 Introduction Aggregation is an incre...
 
Article
An important feature of many advanced active database prototypes is support for rules triggered by complex patterns of events. Their composite event languages provide powerful primitives for event-based temporal reasoning. In fact, with one important exception, their expressive power matches and surpasses that of sophisticated languages offered by Time Series Management Systems (TSMS), which have been extensively used for temporal data analysis and knowledge discovery. This exception pertains to temporal aggregation, for which, current active database systems offer only minimal support, if any. In this paper, we introduce the language TREPL, which addresses this problem. The TREPL prototype, under development at UCLA, offers primitives for temporal aggregation that exceed the capabilities of state-of-the-art composite event languages, and are comparable to those of TSMS languages. TREPL also demonstrates a rigorous and general approach to the definition of composite event language semantics. The meaning of a TREPL rule is formally defined by mapping it into a set of Datalog 1 S rules, whose logic-based semantics characterizes the behavior of the original rule. This approach handles naturally temporal aggregates, including user-defined ones, and is also applicable to other composite event languages, such as ODE, Snoop and SAMOS.
 
Article
This paper surveys the area of biological and genomic sources integration, which has recently become a major focus of the data integration research field. The challenges that an integration system for biological sources must face are due to several factors such as the variety and amount of data available, the representational heterogeneity of the data in the different sources, and the autonomy and differing capabilities of the sources. This survey describes the main integration approaches that have been adopted. They include warehouse integration, mediator-based integration, and navigational integration. Then we look at the four major existing integration systems that have been developed for the biological domain: SRS, BioKleisli, TAMBIS, and DiscoveryLink. After analyzing these systems and mentioning a few others, we identify the pros and cons of the current approaches and systems and discuss what an integration system for biologists ought to be.
 
Article
The need to automatically extract and classify the contents of multimedia data archives such as images, video, and text documents has led to significant work on similarity based retrieval of data. To date, most work in this area has focused on the creation of index structures for similarity based retrieval. There is very little work on developing formalisms for querying multimedia databases that support similarity based computations and optimizing such queries, even though it is well known that feature extraction and identification algorithms in media data are very expensive. We introduce a similarity algebra that brings together relational operators and results of multiple similarity implementations in a uniform language. The algebra can be used to specify complex queries that combine different interpretations of similarity values and multiple algorithms for computing these values. We prove equivalence and containment relationships between similarity algebra expressions and develop qu...
 
Article
Rule-based optimizers and optimizer generators use rules to specify query transformations. Rules act directly on query representations, which typically are based on query algebras. But most algebras complicate rule formulation, and rules over these algebras must often resort to calling to externally defined bodies of code. Code makes rules difficult to formulate, prove correct and reason about, and therefore compromises the effectiveness of rule-based systems. In this paper we present KOLA; a combinator-based algebra designed to simplify rule formulation. KOLA is not a user language, and KOLA's variable-free queries are difficult for humans to read. But KOLA is an effective internal algebra because its combinatorstyle makes queries manipulable and structurally revealing. As a result, rules over KOLA queries are easily expressed without the need for supplemental code. We illustrate this point, first by showing some transformations that despite their simplicity, require head and body ro...
 
Film Finder 1]: a sample DQI. The large square is the starreld display. The widgets on the left, bottom, and right are for querying. This example does not contain an explicit preview bar or chart.
A sample range slider. By moving the arrows, the user speciies the range, which is represented by the white rectangle. The numbers above the arrows give the current range. The numbers on the far ends of the range slider are the extreme values that the attribute can take.
Article
Dynamic query interfaces (DQIs) are a recently developed form of database access that provides continuous realtime feedback to the user during the query formulation process. Previous work shows that DQIs are an elegant and powerful interface to small databases. Unfortunately, when applied to large databases, previous DQI algorithms slow to a crawl. We present a new approach to DQI algorithms that works well with large databases. Keywords Data Structure, Algorithm, Database, User Interface, Information Visualization, Direct Manipulation, Dynamic Query. 1 Our Innovations We propose a new approach to dynamic query interface (DQI) algorithms that can handle much larger databases than previous implementations. Our approach gains its efficiency from the following innovations: Active subset We define an "active" subset of the database, of limited size, which we store in main memory. (While in principle the size of main memory may seem like a severe limitation, in practice DQI algorithms se...
 
Article
In a main memory database (MMDB), the pri- mary copy of the database may be stored in a volatile memory When a crash occurs, a reload of the database from archive memory to main memory must be performed. It is essential that an efficient reload scheme be used to ensure that the expectations of high performance database systems are met. This implies that the overall performance measures of any potential reload algorithm should not be measured simply by reload time, but by its impact on overall system performance. This paper presents four different reload algorithms that aim at fast response time of transactions and high throughput of the overall system. Simulation studies compaxing the algorithms indicate that the best overall approach is one based on frequency of access.
 
Article
Two new spatial join operations, distance join and distance semijoin, are introduced where the join output is ordered by the distance between the spatial attribute values of the joined tuples. Incremental algorithms are presented for computing these operations, which can be used in a pipelined fashion, thereby obviating the need to wait for their completion when only a few tuples are needed. The algorithms can be used with a large class of hierarchical spatial data structures and arbitrary spatial data types in any dimensions. In addition, any distance metric may be employed. A performance study using Rtrees shows that the incremental algorithms outperform non-incremental approaches by an order of magnitude if only a small part of the result is needed, while the penalty, if any, for the incremental processing is modest if the entire join result is required. 1
 
Top-cited authors
Arun Swami
  • Stanford University
Rakesh Agrawal
Tomasz Imielinski
  • Rutgers, The State University of New Jersey
Joerg Sander
  • University of Alberta
Raghu Ramakrishnan