Extending Practical Pre-Aggregation in On-Line Analytical Processing
Published 7 September 1999Open access
Torben Bach Pedersen, Christian S. Jensen, Curtis Dyreson
Citations76
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Algorithms are given that transform “irregular” dimension hierarchies and fact-dimension relationships, which often occur in real-world OLAP applications, into well-behaved structures that, when used by existing OLAP systems, enable practical pre-aggregation.
Abstract
On-Line Analytical Processing (OLAP) based on a dimensional view of data is being used increasingly for the purpose of analyzing very large amounts of data. To improve query performance, modern OLAP systems use a technique known as practical pre-aggregation, where select combinations of aggregate queries are materialized and re-used to compute other aggregates; full preaggregation, where all combinations of aggregates are materialized, is infeasible. However, this reuse of aggregates is contingent on the dimension hierarchies and the relationships between facts and dimensions satisfying stringent constraints, which severely limits the scope of practical preaggregation. This paper
Keywords
Computer Science
IEEE ExpertData mining and knowledge discovery: making sense out of data
4,643 Citations1996U.M. Feyyad
Find loads of the data mining and knowledge discovery making sense out of data book catalogues in this site as the choice of you visiting this page.
Data Mining and Knowledge DiscoveryData Cube: A Relational Aggregation Operator Generalizing Group-By, Cross-Tab, and Sub-Totals
2,134 Citations1997Jim Gray, Surajit Chaudhuri +6 more
This paper explains the cube and roll-up operators, shows how they fit in SQL, explains how users can define new aggregatefunctions for cubes, and discusses efficient techniques to compute the cube.
Research problems in data warehousing
635 Citations1995Jennifer Widom
This paper motivates the concept of a data warehouse, outlines a general data warehousing architecture, and proposes a number of technical issues arising from the architecture that are suitable topics for exploratory research.
John Wiley & Sons, Inc. eBooksThe data warehouse toolkit: practical techniques for building dimensional data warehouses
558 Citations1996Ralph Kimball
This definitive guide succinctly explains how to build a data warehouse by using actual case studies of existing data warehouses developed for specific types of business applications such as retail, manufacturing, banking, insurance, subcriptions and airline reservations.
Lecture notes in computer scienceSelection of views to materialize in a data warehouse
447 Citations1997Himanshu Gupta
Index selection for OLAP
425 Citations2002Himanshu Gupta, Venky Harinarayan +2 more
The authors give algorithms that automate the selection of summary tables and indexes, and present a family of algorithms of increasing time complexities, and prove strong performance bounds for them.
Very Large Data BasesMaterialized Views Selection in a Multidimensional Database
314 Citations1997Elena Baralis, Stefano Paraboschi +1 more
The technique is proposed reduces the soluticn space by considering only the relevant elements of the multidimensional lattice whose elements represent the solution space of the problem.
Algorithms for Materialized View Design in Data Warehousing Environment
311 Citations1997Jian Yang, Kamalakar Karlapalem +1 more
A heuristic algorithm is developed which can provide a feasible solution based on individual optimal query plans and map the materialized view design problem as O-l integer programming problem, whose solution can guarantee an optimal solution.
Aggregate-Query Processing in Data Warehousing Environments
297 Citations1995Ashish Gupta, Venky Harinarayan +1 more
Generalized projections are introduced, that capture aggregations, groupbys, duplicate-eliminating projections (distinct and duplicate-preserving projections in a common unified framework), and powerful query rewrite rules for aggregate queries are developed that unify and extend rewrite rules previously known in the literature.
Summarizability in OLAP and statistical data bases
292 Citations2002H. Lenz, Arie Shoshani
A framework for precisely specifying the context in which statistical objects are defined is introduced, which uses a three-step process to define normalized statistical objects.
Answering Queries with Aggregation Using Views
238 Citations1996Divesh Srivastava, Shaul Dar +2 more
It is shown that in the case where the query has grouping and aggregation but the views do not, a view is usable in answering a query only if there is an isomorphism between the view and a portion of the query.
Lecture notes in computer scienceSelection of Views to Materialize Under a Maintenance Cost Constraint
216 Citations1999Himanshu Gupta, Inderpal Singh Mumick
This article develops algorithms to select a set of views to materialize in a data warehouse in order to minimize the total query response time under the constraint of a given total view maintenance time and designs an A* heuristic, that delivers an optimal solution.
Data Warehouse Configuration
199 Citations1997Dimitri Theodoratos, Timos Sellis
A theoretical framework for dealing with the DW configuration problem by formulating it as a state space optimization problem and then solving it using an exhaustive incremental algorithm as well as a *Research supported by the European Commission under the ESPRIT Program LTR project No 22469 “DWQ: Foundations of Data Warehouse Quality”
Multidimensional data modeling for complex data
193 Citations1999Torben Bach Pedersen, Christian S. Jensen
The paper presents nine requirements to multidimensional data models, each of which is exemplified by a real world, clinical case study, and defines an extended multiddimensional data model, and an associated algebra, which address all nine requirements.
Storage Estimation for Multidimensional Aggregates in the Presence of Hierarchies
140 Citations1996Amit Shukla, Prasad Deshpande +2 more
Three strategies for estimating the storage blowup that will result from a proposed set of precomputations without actually computing them are proposed: one based on sampling, onebased on mathematical approximation, and one based upon probabilistic counting.
On-line warehouse view maintenance
130 Citations1997Dallan Quass, Jennifer Widom
This work describes an algorithm called 2VNL that allows warehouse maintenance transactions to run concurrently with readers, and describes how it can be implemented on top of a conventional relational DBMS using a query rewrite approach.
Lecture notes in computer scienceStorm: A statistical object representation model
121 Citations1990Maurizio Rafanelli, Arie Shoshani
This paper identifies a number of SO representational problems in current models and a methodology for their solution, and proposes a new "statistical object representation model", based on a graph representation.
The MIT Press eBooksMaintenance of Data Cubes and Summary Tables in a Warehouse
115 Citations1999Inderpal Singh Mumick, Dallan Quass +1 more
Data warehouses contain large amounts of information, often collected from a variety of independent sources, and decision-support functions in a warehouse, such as on-line analytical processing (OLAP),...
The MIT Press eBooksImplementing Data Cubes Efficiently
109 Citations1999Venky Harinarayan, Anand Rajaraman +1 more
Information Retrieval from an Incomplete Data Cube
47 Citations1996Curtis Dyreson
The design of an incomplete data cube is described, which includes an algorithm that removes redundant cubettes and an algorithm to increase the retrieval power of the federation through the creation of virtual cubettes.
On-Line Warehouse View Maintenance for Batch Updates
16 Citations1996Dallan Quass, Jennifer Widom
An algorithm called 2VNL is described that allows warehouse maintenance transactions to run concurrently with readers, logically maintaining two versions of the database, so that no locking is required and serializability is guaranteed.