What is data warehousing?

Constructing and using a warehouse: collecting, cleaning, integrating data and providing query, OLAP and mining tools

What is the first step in creating a warehouse?

Requirement analysis of goals, subjects, users and key queries

What is decided in the design-the-model step?

Architecture (top-down/bottom-up), dimensional schema, granularity and metadata

Why does an organisation need a warehouse?

To get a consistent view and analyse history without disturbing operational systems

To remove its operational systems

Year not given · 3+3+4 marks

Define data warehousing and creation process

Short answer

Data warehousing integrates heterogeneous data into a subject-oriented historical repository for decision support. Organizations need it for a consistent view, historical analysis, improved quality, and competitive advantage. Creation involves requirement analysis, source identification, dimensional modeling, ETL development, database building, access layer setup, testing, deployment, and maintenance.

How do you define data warehousing? Why do any organization need a data warehouse? Explain the process of creating a data warehouse.

Data warehousing

Data warehousing is the process of constructing and using a data warehouse: collecting data from multiple heterogeneous sources, cleaning and integrating it, storing it in a subject-oriented historical repository, and providing tools for query, OLAP and mining to support decision making.

Why an organisation needs a data warehouse

To get one consistent view of data scattered across many systems.
To analyse historical trends without disturbing operational systems.
To improve data quality and reporting speed.
To support planning, forecasting and data mining for competitive advantage.

Process of creating a data warehouse

Requirement analysis: identify business goals, subject areas, users and key queries.
Identify data sources: operational databases and external data to be included.
Design the model: choose architecture (top-down/bottom-up), design dimensional schema (facts, dimensions, granularity) and metadata.
ETL design and development: extract data from the sources, clean it (errors, duplicates, missing values), transform it (formats, codes, derived values) and load it.
Build the warehouse database: create tables, indexes, aggregates and data marts; choose hardware and DBMS.
Build access layer: OLAP server, report, query and mining tools.
Test and deploy: check data accuracy and performance, train users, release.
Maintain: periodic refresh, backup, security, tuning and growth.

 Plan -> Design -> Build ETL -> Load -> Test -> Deploy
   ^                                              |
   +------------- Maintain / refresh <------------+

✓ Answer checked by Claude Opus 5.5 (high effort)

Data warehouse, data mart and need · Data Mining, chapter 1
National data warehouse process architecture · Data Mining, chapter 1
Typical process flow in warehouse · Data Mining, chapter 1
Data warehouse schemas · Data Mining, chapter 1
Data warehouse architecture and components · Data Mining, chapter 1
Short note: Noise in data warehouse · Data Mining, chapter 2
General process of creating a PCB · Instrumentation II, chapter 8
Data warehouse and data mining · Information System, chapter 4 · asked 2×
Need for organization and its principles · Organization and Management, chapter 1 · asked 3×
Short note: data mining and warehousing · Information System, chapter 4

All Data Mining chapter 1 questions →

Define data warehousing and creation process

Data warehousing

Why an organisation needs a data warehouse

Process of creating a data warehouse

Related questions