Exam 2- Chapter 5 Flashcards by l l l l

Our IS world is divided into two major kinds of systems:

Our IS world is divided into two major kinds of systems:
Transactional (Line of Business) Systems
Reporting Systems

How well did you know this?

Not at all

Perfectly

We separate these because we want to:

IS world is divided into two major kinds of systems

We separate these because we want to:
Avoid contention in transactional systems
Integrate data from numerous other systems
Improve reporting data extraction performance
Provide a “single source of truth”

How well did you know this?

Not at all

Perfectly

The data warehouse is the

The data warehouse is the backbone of the reporting infrastructure

How well did you know this?

Not at all

Perfectly

Extract

[2]

Extract
Pulling data from a source
We may need to get data from multiple sources

How well did you know this?

Not at all

Perfectly

Transform



TransformModify data for standards and consistency

How well did you know this?

Not at all

Perfectly

Load



LoadPush the transformed data into a data mart or warehouse

How well did you know this?

Not at all

Perfectly

WHY TRANSFORM?

Translating / Mapping coded values
Calculating new values
Integrating from different sources
Aggregating data
Splitting / joining data
Pivoting data

How well did you know this?

Not at all

Perfectly

STAGING



STAGING
A holding area for data that has been extracted from a source
Used to pool data from multiple sources together

How well did you know this?

Not at all

Perfectly

DATA WAREHOUSE STRUCTURE
Facts



DATA WAREHOUSE STRUCTURE

FactsAnswer questions about WHAT happenedContains aggregable metrics

How well did you know this?

Not at all

Perfectly

DATA WAREHOUSE STRUCTURE
Dimensions



DATA WAREHOUSE STRUCTURE
Dimensions
Answer questions about when, for what, where, for whom, and by whom.
Used to slice through the facts in different ways

How well did you know this?

Not at all

Perfectly

DATA WAREHOUSE STRUCTURE



DATA WAREHOUSE STRUCTURE

The basic structure is called a “Star Schema”

How well did you know this?

Not at all

Perfectly

DATA WAREHOUSE STRUCTURE



DATA WAREHOUSE STRUCTURE

Each Star is a [Data Mart]

How well did you know this?

Not at all

Perfectly

DATA MART VS DATA WAREHOUSE
A data mart is a single star schema
     
    
    

DATA MART VS DATA WAREHOUSE
A data mart is a single star schema
Describes a single subject or activity
Usually has one fact table or set of related fact tables
Can share dimensions with other data marts

How well did you know this?

Not at all

Perfectly

DATA MART VS DATA WAREHOUSE
A data warehouse is



DATA MART VS DATA WAREHOUSE
A data warehouse is
Enterprise level
Aggregation of all data marts

How well did you know this?

Not at all

Perfectly

DATA MART VS DATA WAREHOUSE

Essentially, a data mart is a

DATA MART VS DATA WAREHOUSE

Essentially, a data mart is a subject-oriented segment of a data warehouse

How well did you know this?

Not at all

Perfectly

OLAP CONCEPTS

 _______ _____ Processing

OLAP CONCEPTS

Online Analytical Processing

OLAP CONCEPTS

 Data Cubes are

OLAP CONCEPTS

 Data Cubes are multidimensional structures separated from the data warehouse

OLAP CONCEPTS
Cubes can contain



OLAP CONCEPTS
Cubes can contain
Base data
Aggregations

OLAP CONCEPTS

 Always populated from the _____ ______

OLAP CONCEPTS

Always populated from the data warehouse

OLAP CONCEPTS

 Allows extraction of _____ and ______ to be faster

OLAP CONCEPTS

Allows extraction of data and aggregations to be faster

OLAP CONCEPTS

 Allows extraction of _____ and ______ to be faster

OLAP CONCEPTS

Allows extraction of data and aggregations to be faster

OPERATIONAL DATA STORE (ODS)

Used to

OPERATIONAL DATA STORE (ODS)

Used to monitor activities outside of the data warehouse

OPERATIONAL DATA STORE (ODS)

 ______ ____ than the data warehouse

OPERATIONAL DATA STORE (ODS)

Less latency than the data warehouse

OPERATIONAL DATA STORE (ODS)
Examples of values monitored can be:



OPERATIONAL DATA STORE (ODS)
Examples of values monitored can be:
KPI
Event flags

OPERATIONAL DATA STORE (ODS) | Often the source for _____

OPERATIONAL DATA STORE (ODS) | Often the source for Dashboards

OPERATIONAL DATA STORE (ODS) | ** Does NOT replace the _____ _______!!

OPERATIONAL DATA STORE (ODS) | ** Does NOT replace the data warehouse!!

THE METADATA REPOSITORY |  Metadata is “ __ __ __”

THE METADATA REPOSITORY |  Metadata is “data about data”

THE METADATA REPOSITORY |  Searchable _____ with _____ about ____

THE METADATA REPOSITORY |  Searchable repository with descriptions about data

THE METADATA REPOSITORY  Contains  

THE METADATA REPOSITORY  Contains  Documentation of data structures  Data dictionary with information about the content of the data structures

THE METADATA REPOSITORY  Examples:  

THE METADATA REPOSITORY  Examples: Revenue figures are stored in tables. How do we know if they are expressed in USD or some other currency? Does the revenue figure include sales taxes charged?

What is ETL?

Extract, Transform, Load