A Data Warehouse -Ralph Kimball [father of DWH-ii]
1. A Data warehouse is a relational database management system which is specifically designed for business analysis purpose, rather than for Business Transactional Processing.
2. A data Ware house is designed to support Decision Making Processing [DMP].
3. Decision Support System [DSS]:- since that Data Warehouse is designed to support Decision Making Processing. Hence it is call as DSS.
Read only Database
The data warehouse is designed to make query the data required for analysis but not for business Transactional processing. Hence it is known as Read only Database.
Historical Database
Since a data warehouse contains a historical data which is required for analysis. Hence it is known Historical Database.
A Data Warehouse-W.H. Inmon[father of DWH-I]
A data warehouse is a 1. Time variant, 2.Non-volatile, 3.Subject Oriented, 4. Integrated database.
Characteristics features of Data Warehouse
1. Time Variant: - A data warehouse is time variant database which support the business users in analysis the business and companies the business with different time periods. This is known as time series analysis.
Data warehouse
2. Non-volatile:- A data warehouse is non-volatile database, once the data enter into the data warehouse it does not reflex to the change which takes place at operational source system. Hence the data is static in Data warehouse.
3. Integrated Database: - A data warehouse is an integrated database which store data in integrated format collected from multiple operational source systems.
4. Subject oriented:-A Data warehouse is subject oriented database which support the business needs of middle management users. A subject represents department specification business needs.
Example: HR, Sales, Accounts, Marketing, Loans, Insurance.
Differences between Operational Database and Data warehouse.
Operational Database Data warehouse
1. It is designed to support 1. It is designed to support Decision Making Decision making Processing. processing
2. Designed for Business Operation. 2. Decision for Business analysis.
3. Volatile data 3. Non-volatile data.
4. Current data 4. Historical data.
5. Less history [3-6 months] 5. More history [15-20 years].
6. Detail data 6. Summary data.
7. Normalization 7. De-normalization.
8. Designed for clerical access 8. Managerial access.
9. E-R Modeling 9. Dimensional Modeling.
10. More joins 10.Few joins.
Data Marts and Types of Data Marts
A data Marts is subset of an Enterprise Data warehouse. A data mart is subject
oriented database which supports business needs of middle management
or department specific users. A data mart is also known as High Performance
Query Structures (HPQS).
oriented database which supports business needs of middle management
or department specific users. A data mart is also known as High Performance
Query Structures (HPQS).
Data warehouse store enterprises specific historical data.
There are 2 types’ data marts.
1. Dependent data marts.
2. Independent data marts.
Difference between data mart and enterprise data warehouse
Data Mart Data warehouse
1. It is a department specific database it is an enterprises specific database.
2. Sigel subject integration of multiple subjects.
3. Middle management access top management access.
Top down data warehousing approach [W.H.INMON]
According to the w.h.inmon first we need to design enterprises specific database
known as data warehouse, from the enterprise data warehouse derive the
subject oriented database known as data marts.
known as data warehouse, from the enterprise data warehouse derive the
subject oriented database known as data marts.
Dependent data mart
In a top down approach a data mart development depends on an enterprise data
warehouse. Hence such data mart known as dependent data marts.
warehouse. Hence such data mart known as dependent data marts.
Bottom up approach [Ralf Kimball]
According to the Kimball first designed the department specific databases known as
data marts integrated the data marts into an enterprise data warehouse.
data marts integrated the data marts into an enterprise data warehouse.
Independent data marts
In the bottom up approach data mart development is independent of an enterprise
data warehouse such data marts known as independent data marts.
data warehouse such data marts known as independent data marts.
Operational Data Store[ODS 60-90days]
An ODS is integrated view of operational data sources or operational databases
designed to support business transactional processing.
designed to support business transactional processing.
Differences between ODS and DWH
1. Integrated database integrated database.
2. Subject oriented subject oriented.
3. Detail data summary data.
4. Current data[60-90 days] historical data[15-20 years]
5. Volatile data non volatile data.
6. Normalization de normalization.
This article had provided me with efficient data warehouse solutions due to which I was capable of managing a large amount of data comfortably. Though the maintenance and transformation of the big data are complicated, after taking help from your blog, the process has become more comfortable and I had managed the critical data by following some easy steps.
ReplyDeleteData warehouse is a system that collects data from multiple sources, processes it, and then makes it available for sharing across applications and business units. Data warehousing has become a popular solution for enabling data-driven applications.
ReplyDelete