Moss L.T., Atre S. Business intelligence roadmap: The complete project lifecycle for decision-support applications
Подождите немного. Документ загружается.
[ Team LiB ]
Deliverables Resulting from These Activities
Prototype charter
This document is similar to a project charter because it represents an
agreement between the business sponsor and the IT staff regarding the
prototyping activities for the BI project. It contains the following sections:
- Primary purpose for the prototype
- Prototype objectives
- Prototype participants
- Data to be used for the prototype
- Hardware and software platforms to be used
- Measures of success
- Application interface agreement
1.
Completed prototype
The main deliverable from this step is a completed prototype. This can be a
demo, a few mocked-up screens, or a partially functioning BI application.
2.
Revised application requirements document
During prototyping, you may discover new requirements or decide to change or
drop some of the original ones. Reflect these changes in the application
requirements document.
3.
Skills survey matrix
This matrix indicates the skill sets of the business people. It should state
whether a business person has beginning, advanced, or expert skills in
computer usage as well as knowledge about the business functions pertaining
to the BI application.
4.
Issues log
Document any issues that came up during prototyping (whether they were
resolved or not) in an issues log, indicating status or final resolution, impact on
the real BI application, action items to pursue, and to whom the action items
were assigned.
5.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
221/631
[ Team LiB ]
Roles Involved in These Activities
Application lead developer
The application lead developer should review the application requirements
document with the business representative, and together they should prepare a
short and informal prototype charter. The application lead developer should
also review the existing or mock-up report and query layouts, which will form
the basis for the prototype design. He or she also has to make plans for
scheduling the prototype demonstrations.
Business representative
The primary responsibility of a business representative is to use the prototype
to learn about the feasibility and the look and feel of the BI application. The
business representative should assist the application lead developer in creating
the prototype charter. The business representative also needs to review the
overall BI project requirements and revise them if necessary. This can be
accomplished only if the business representative participates in the design and
the review of the prototype and is actively involved in the demonstrations.
Database administrator
The database administrator is responsible for designing and loading the
prototype database with sample source data or sample test data. He or she
should also review all database access calls (Structured Query Language [SQL]
statements).
Stakeholders
The stakeholders do not directly participate in the BI project activities, but they
have a vested interest in the project. They should take part in the prototype
demonstrations and provide input regarding any additional common access
requirements for reports, queries, and ad hoc usage.
Subject matter expert
The subject matter expert has to analyze and discuss the access and analysis
requirements for reports and queries with the prototyping team. He or she
should work with the business representative and project manager to determine
the purpose, goals, and primary use of the prototype.
Web master
The Web master has to review the existing tools slated for use with Web access
to the BI application. The Web master also has to evaluate the usability of the
prototype design with regard to Web access and has to determine the necessary
interfaces.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
222/631
[ Team LiB ]
Risks of Not Performing Step 6
The main purpose of prototyping is to make sure that the design of the database, the
design of the access and analysis application, and the BI technologies selected will
be able to meet the business requirements when the BI application is implemented
as intended. By building a successful prototype, you can also validate the accuracy of
your cost and time estimates for the full-scale final BI application. The risk of not
performing this step is that you may build a BI solution that will cost much more and
take much longer than you expected—and that you will not realize it until it is too
late.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
223/631
[ Team LiB ]
Bibliography and Additional Reading
Bajaj, Chandrajit. Trends in Software: Data Visualization Techniques. New York: John
Wiley & Sons, 1999.
Beck, Kent. Extreme Programming Explained: Embrace Change. Boston, MA:
Addison-Wesley, 2000.
Cockburn, Alistair. Agile Software Development. Boston, MA: Addison-Wesley, 2002.
Devlin, Barry. Data Warehouse: From Architecture to Implementation. Reading, MA:
Addison-Wesley, 1997.
Humphrey, Watts S. Winning with Software: An Executive Strategy. Boston, MA:
Addison-Wesley, 2002.
Kimball, Ralph, Laura Reeves, Margy Ross, and Warren Thornthwaite. The Data
Warehouse Lifecycle Toolkit: Expert Methods for Designing, Developing, and
Deploying Data Warehouses. New York: John Wiley & Sons, 1998.
Shneiderman, Ben. Designing the User Interface: Strategies for Effective Human-
Computer Interaction. Boston, MA: Addison-Wesley, 1998.
Turkle, Sherry. Life on the Screen: Identity in the Age of the Interface. New York:
Simon & Schuster, 1997.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
224/631
[ Team LiB ]
Chapter Seven. Step 7: Meta Data
Repository Analysis
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
225/631
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
226/631
Chapter Overview
This chapter covers the following topics:
Things to consider when analyzing whether to license (buy) or build a meta
data repository
Why it is important to deliver meta data with every BI project
The differences between the two categories of meta data: business meta data
and technical meta data
How a meta data repository can help business people find and use their
business data after the data has been standardized for the BI decision-support
environment
The four groupings of meta data components: ownership, descriptive
characteristics, rules and policies, and physical characteristics
How to prioritize meta data for implementation purposes
Five common difficulties encountered with meta data repository initiatives:
technical, staffing, budget, usability, and political challenges
The entity-relationship (E-R) meta model used to document the meta data
requirements
A definition and examples of meta-meta data
Brief descriptions of the activities involved in meta data repository analysis, the
deliverables resulting from those activities, and the roles involved
The risks of not performing Step 7
Things to Consider
Meta Data Repository Usage
Who will be using the meta data in the meta data repository?
What standards do we have in place for its use? What standards do
we need to develop?
Do we already have a meta data repository? If not, will we license
(buy) one or build one?
Will this meta data repository support only the BI decision-support
environment, or will it be used for all systems throughout the
organization?
How will we know if we are using the meta data repository
effectively?
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
227/631
Meta Model Requirements
Do we need to create a meta model for the meta data repository,
or do we already have one?
If we have one, what meta data objects do we need to add to it?
Which meta data repository products support our meta model?
Are these meta data repository products extendable?
Meta Data Repository Security
What type of security will we need for the meta data repository?
Who will be authorized to enter and maintain the meta data?
Will everyone be authorized to access any meta data at any time?
Meta Data Capture
What types of business meta data do we need to capture?
Will we be using a computer-aided software engineering (CASE)
tool to capture the business meta data?
What type of technical meta data do we need to capture?
Will we be capturing technical meta data in the
extract/transform/load (ETL), online analytical processing (OLAP), and
other access and analysis tools?
How will we extract the meta data from these tools and migrate it
to the meta data repository? Who is responsible for migrating it?
Who will connect the business meta data from the CASE tool to the
technical meta data from the ETL, OLAP, and other tools?
Meta Data Delivery
How will meta data be displayed? How will it be accessed? Will we
have a Web interface to the meta data repository?
Will we need to produce meta data reports? What types of reports?
How will these reports be distributed?
Will there be a help function (online tutorial)? Will the help function
be context sensitive?
Staffing
Do we already have a meta data administrator? If not, do we have
a data administrator with technical skills who can perform the
functions of a meta data administrator?
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
228/631
Will we have to hire more meta data administrators?
How will meta data responsibilities be divided among data
administrators and meta data administrators?
A meta data repository is a database. But unlike ordinary databases, a meta data
repository is not designed to store business data for a business application. Instead,
it is designed to store contextual information about the business data. Examples of
contextual information about business data include the following:
Meaning and content of the business data
Policies that govern the business data
Technical attributes of the business data
Specifications that transform the business data
Programs that manipulate the business data
Contextual information about business data inherently exists in every organization,
whether it is documented or not. When contextual information is documented, it is
known as meta data. When the information is not documented, it is usually not
known to everyone in the organization. As a result, business people often invent their
own business rules and create their own redundant data along with redundant
processes, not realizing (or sometimes not caring) that what they need may already
exist. Forty years of such practices have now brought most organizations to the brink
of drowning in data and dehydrating from lack of information.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
229/631
[ Team LiB ]
The Importance of Meta Data
Meta data describes an organization in terms of its business activities and the
business objects on which the business activities are performed. Consider, for
example, a sale of a product to a customer by an employee. The sale is a business
activity and the product, customer, and employee are the business objects on which
the sale activity is performed. Business activities and business objects, whether
manual or automated, behave according to a set of relationships and rules, which
are defined by the business. These activities and objects, and the relationships and
rules that govern them, provide the context in which the business people use the
business data every day.
Meta data is so important for the BI decision-support environment because it helps
metamorphose business data into information. The difference between data and
information is that information is raw data within a business context. Meta data
provides that business context; that is, meta data ensures the correct interpretation
(based on activities, objects, relationships, and rules) of what the business data
actually means.
For example, what is profit? Is it the amount of money remaining after a product has
been sold and everybody who was involved in that product has been paid? Or is it a
more complicated calculation, such as "total annual revenue minus sum of average
base cost per product minus actual staff overhead minus accumulated annual
production bonuses minus wholesale discounts minus coupons divided by twelve?"
Does every business person have the same understanding of profit? Is there one and
only one calculation for profit? If there are different interpretations of profit, are all
interpretations legitimate? If there are multiple legitimate versions for profit
calculations, then multiple data elements must be created, each with its own unique
name, definition, content, rules, relationships, and so on. All of this contextual
information about profit is meta data.
Since meta data provides the business context in which business data is used, meta
data can be viewed as a semantic (interpretive) layer of the BI decision-support
environment. This semantic layer helps the business people navigate through the BI
target databases, where the business data resides. It also helps the technicians
manage the BI target databases as well as the BI applications.
Some important characteristics of meta data and meta data repositories are listed
below.
A meta data repository is populated with meta data from many different tools,
such as CASE tools, ETL tools, OLAP tools, and data mining tools.
Meta data documents the transformation and cleansing of source data and
provides an audit trail of the periodic data loads.
Meta data helps track BI security requirements, data quality measures, and
growth metrics (for data volume, hardware, and so on).
Meta data provides an inventory of all the source data that populates the BI
applications.
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
230/631