Glossary
This is the glossary of data warehousing terms. The top part of the glossary is an index. By clicking one of the links, you will be taken directly to that item. The bottom part is an alphabetical list of all the glossary items.
Click a letter to go directly to that letter's list of glossary items.
A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
Ad hoc report
An ad hoc report is a self-developed report by analysts or end-users. The term implies iterative development and exploration of the data. A desktop reporting software tool is frequently used for ad hoc reporting. An ad hoc report can also be nonrecurring, one time or random query or analysis. (back to top)
Aggregation
Also commonly called a summary, an aggregation is a collection of data calculated from detail transactions. An aggregation is usually a sum, count or average of the underlying detail, and often is calculated along several business dimensions , i.e. total sales by customer by product. (back to top)
Analytical report
A report that usually summarize or aggregates data to aid managers in identifying trends, analyzing volumes of data, or performing planning or forecasting. It is common for an analytical report to pull data from a warehouse or data mart, which may contain data from a variety of source systems. Examples of this type of report are an historical admissions trend analysis or a longitudinal study of graduation rates. (back to top)
Application
A computer program developed to provide a specific function or answer a specific business need. (back to top)
Application Account
An application account is an account that is directed toward a specific application that is used to periodically (daily, weekly, etc.) download data from the EDW and is owned by a specific individual known as the primary user. This person is responsible for use of this account. There may also be secondary users involved in the technical management of this account who share in the responsible of using it. The application account cannot be used to provide individuals with direct, shared access to the Data Warehouse. (back to top)
Architecture
A definition of the interconnection of computer components, network components or system components. (back to top)
Assumption
Factors that, for planning purposes, are considered to be true, real or certain. (back to top)
Banner
Banner is the Enterprise Resource Planning (ERP) system selected by the University of Illinois to integrate and manage its student, financial and human resources systems applications. (back to top)
BO
The acronym often used to refer to Business Objects, the commercial software product that provides an interface to Decision Support databases at the University of Illinois. See Business Objects. (back to top)
Business Driver
A problem in the business that can be solved by technology. (back to top)
Business Intelligence
An umbrella term for the processes and tools used for turning data into information that can be used in making decisions. Originally the term referred primarily to reporting, but today business intelligence also encompasses: ad hoc querying, online analytical processing, data mining, forecasting, and other decision support systems. Elements of a business intelligence system are: understanding customer needs and then gathering, analyzing and providing access to data. Used commercially, the term refers to software used to analyze and present data. At the University of Illinois, running a query against the Data Warehouse is an example of a business intelligence activity and EDDIE is an example of business intelligence software. (back to top)
Business Objects
Business Objects (BO) is a commercial software product licensed by the University of Illinois. It is collection of applications that provide query, reporting and analysis services. Business Objects provides an interface to Decision Support databases. See www.businessobjects.com.
(back to top)
Business Rules
Business rules are the logic applied to calculate or otherwise derive a value. They are based on reporting requirements and business practices. (back to top)
Cardinality
The specific type of relationship that exists between two tables. If a single record in the first table is related to only one record in the second table a one-to-one relationship exists. If a single record in the first table can be related to one or more records in the second table, a one-to-many relationship exists. Or, if a single record in the first table can be related to one or more records in the second table, and a single record in the second table can be related to one or more records in the first table, a many-to-many relationship exists. Hernandez, 1997. (back to top)
Change Request
A change request is a request for modifications to the scope, quality, time and/or cost of a project. (back to top)
Column
A means of implementing an item of data within a table or view. It can be character, date, number or other format, and be optional or mandatory. Oracle, 1998. (back to top)
Constraint
Restriction or boundary impacting overall capability, priority, and resource. The four primary project coordination constraints are scope, quality, time and resources. (back to top)
Data Dictionary
The data dictionary gives definitions for all of the tables in the Enterprise Data Warehouse and Data Marts as well as column names and descriptions, table keys, and sample valid values. The data dictionary is useful for finding out more about a specific table or column. (back to top)
Data Integrity
The requirement to provide data that is free of errors and omissions. (back to top)
Data Mart
A set of data designed and constructed for decision support purposes reflecting the design principles of a data warehouse provided to serve the needs of a homogenous user group. (back to top)
Data Product
A generic term for any of the various types of data structures that Decision Support provides in the Data Warehouse. For example: a data mart or a Universe is a data product. (back to top)
Data Warehouse
A data warehouse is one or more databases designed to support management, analysis, reporting and decision making in an organization. Data from the operational systems (such as Banner) are migrated to the data warehouse so that queries can be performed without disturbing the performance or the stability of the operational systems. The traditional features of a data warehouse are: subject-oriented, integrated across subject areas and time-oriented, i.e., there is a historical perspective to the data. A data warehouse generally contains both detailed data and may also contain summarized data. What is included in a data warehouse environment varies widely but is generally acknowledged to be the database(s), documentation about the data (metadata), a delivery/access method and a commitment to extract, transform and load (ETL) data from multiple source systems. (Adapted from http://www.pcmag.com/encyclopedia_term and Bill Inmon, 1996) (back to top)
Datawarehousing
The discipline of developing and maintaining a data warehouse. A data warehouse is more of a process than a structure, since data is cumulated on a regular basis and requires constant attention to provide useful data to customers. The data warehousing process can be complex and requires careful attention to customer needs and best practices to be successful. (back to top)
Database
A computer application that stores data and allows access and manipulation of the data. (back to top)
Decision Support
Decision Support (DS) is a customer service unit with a focus on data warehousing, business intelligence, and information management. Decision Support is part of the University Office for Planning and Budgeting. Its responsibilities include managing the University’s Data Warehouse, providing data access, helping staff create their own reports, and providing data education and general information about University data and reporting. Decision Support is the unit that manages Business Objects and EDDIE, one of the servers that delivers Standard Reports to users. See About Decision Support for more information. (back to top)
Deliverable
A report or product of one or more tasks that satisfy one or more objectives and must be delivered to satisfy requirements. (back to top)
Denormalized
A data storage design that allows repeating or redundant data in tables to simplify querying and reporting. (back to top)
Dependency
A relationship between tasks, such that one requires input from the other to begin. (back to top)
Detail Fact Table
A table used in a star schema to store the detail transaction level data. (back to top)
Dimension
A general category of data, such as time, product, or geography. (back to top)
Dimension Table
A table used in a star schema to store descriptive, hierarchical and metric information about an aspect of the business that is used for an analytical perspective. Common examples include product, customer, geography, and time. (back to top)
Disaster Recovery
The policies and plans used to restore a computer system from system failure. (back to top)
DS
The acronym often used to refer to Decision Support, the customer service unit with a focus on data warehousing, business intelligence, and information management. See Decision Support. (back to top)
EDDIE
The acronym used for Enterprise Data Delivery and Information Environment. EDDIE is the Business Objects server used to provide a variety of business intelligence services, including: query and analyze data, develop custom reports and distribute reports. EDDIE is managed by the Decision Support unit. (back to top)
EDW
The acronym often used to refer to the Enterprise Data Warehouse. See Enterprise Data Warehouse. (back to top)
Enterprise
An enterprise consists of all functional departments, people, and systems within an organization. In some cases, the enterprise can include partners--even vendors and customers. For example, the University of Illinois is an enterprise. (back to top)
Enterprise Architecture
A high level enterprise wide data warehouse framework that describes the subject areas, sources, business dimensions, metrics, business rules and semantics of an organization. It is used to identify shared sources, dimensions, metrics and semantics in an iterative data mart or
iterative subject area development methodology. (back to top)
Enterprise Data Warehouse
An Enterprise Data Warehouse (EDW) is a non-volatile data store containing historical, detailed data that spans a number of subject areas. This data store is fed by transactional data on a regular basis from a variety of data sources. In the eyes of the end-user, the EDW is a read-only environment. At the University of Illinois, the EDW is one component of the overall Data Warehouse. The Data Warehouse contains both the EDW and specialized Data Marts. (back to top)
Enterprise ID
The Enterprise ID and password are used to authenticate users at the University of Illinois via the Enterprise Application Service, a central provider of authentication, authorization, and session management services. This central service allows users of multiple enterprise applications on a variety of platforms at the University to use a common Enterprise ID and password. (back to top)
Enterprise Resource Planning
Enterprise Resource Planning (ERP) for University of Illinois purposes, is the name that describes a single vendor software suite that integrates Business and Finance, Human Resources and Payroll, and Student Systems applications for use across the enterprise. (back to top)
ERP
The acronym for Enterprise Resource Planning. See Enterprise Resource Planning. (back to top)
ETL
The abbreviation used for Extract Transform and Load. See Extract Transform and Load. (back to top)
Event Notice
A message sent out to University system users informing them of occurrences that might impact normal day-to-day operations. These messages include the following section of information: "OF INTEREST TO", "WHAT\WHEN", "IMPACT", "WHY", and "CONTACT". (back to top)
Executive Information System (EIS)
A computer application and specifically designed data sets used to provide answers and analysis to executive business management. (back to top)
Extract, Transform and Load
A set of software applications that can be programmed to read and copy data from source computer files into a data staging area for further work (extract), modify, clean or reorganize the data (transform) and move it into an Enterprise Data Warehouse in either a bulk or an incremental fashion (load). (back to top)
FAC
The acronym for Functional Area Coordinator, an employee position in Decision Support. See Functional Area Coordinator. (back to top)
FERPA
FERPA, or the Family Educational Rights and Privacy Act of 1974, is a part of the Educational Amendments of 1974, P.L. 93-380 that became effective November 19, 1974. A number of amendments were signed into law on December 31, 1974 and draft guidelines were issued by HEW on January 6, 1975. This act provides that federal funding will be withheld from any higher educational institution which denies a student the right to inspect and challenge the content of the student’s cumulative record. In addition, the act imposes substantial restrictions upon access by others to a student’s record without the student’s written consent. Any user of student data at the university must complete federally mandated FERPA training as outlined by University policy and administered by each campus OAR office. (back to top)
Field
A means of implementing an item of data within a file. It can be in character, date, number or other format, and be optional or mandatory. Oracle, 1998. (back to top)
Flat File