Data Warehousing Concepts in SSIS
Data Warehousing Concepts
DATA: data is composed of observable and recordable facts that are often found in operational for transactions systems.
OLTP: OLTP is nothing but observation of online transaction processing. The system is an applicable application that modifies data the instance it receives and has a large number of concurrent users.
OLAP: OLAP is abbreviation of online analytical processing this system is an application that collects manager processes and presents multidimensional data for analysis and management purpose.
DATA MINING: data mining is the process of analyzing data from different perspectives and summarizing it in to useful information.
BI: BI is the leveraging of BW to help make business decisions and recommendations. Information and data rules engines are leveraged here to help make these decisions along with statistical analysis tools and data mining tools.
DWH DEFINITIONS: Bill Inmon, considered to be the father of the data ware housing provides the following definition
“A data ware house is a Subject oriented, Integrated, Non volatile, And time variant.
Collection of data in support of management’s decision seam Kelly, another leading data ware housing Practitioner defines the data ware house in the following way.
The data in the data ware house is,
In the data ware house, data is not stored by operational applications, but by business subjects.
Operational applications data ware house subject
Data ware house Subjects
Data inconsistencies are removed; data from diversified operational applications is integrated
Here are some of the items that would need standardization:
- Naming conventions
- Data attributing
Time variant data:
The time- variant nature of the data in a data ware house
- Allows for analysis of the past
- Relates information to the presents
- Enables forecasts for the future
Non volatile data:
Usually the data in the data ware house is not updated for deleted,
Data granularity in a data ware house refers to the level of detail data. The lower level is detail, the finer the data granularity.
- Depending on the requirements multiple levels of details may be present. Many data ware house have at least dual levels of granularity.
Three data levels in a banking data ware house
|Daily detail||monthly summery||quarterly summery|
|Amount||Number of transactions||Number of transactions|
|Deposit/withdrawal||With drawlsDepositsBeginning balance
|With drawlsDepositsBeginning balance
Data warehouse approaches
(Bottom up approach)W.H 2 NMON
(Top down approach)Here data marts designed first, later from data marts for ware houseEnterprise DWH concentrated first, next data marts
Top- down approach:
The advantages of this approach are:
- A truly corporate efforts, an enterprise view of data
- Inherently architected – not a union of the separate data marts
- Single, central storage of data about the content
- Centralized rules and control
- May sea quick results if implemented with iterations
The disadvantages are:
- Takes longer to build even with an iterative method
- High exposure risk to failure
- Needs high level of cross – fundamental skills
- High outlay without proof of concept.
Bottom – up approach:
The advantages of these approaches are:
- Faster and easier implementation of manageable pieces
- Favorable returns on investment and proof of concept
- Less risk of failure
- Inherently, incremental: can schedule important data marts first
- Allows project team to learn and grow
The disadvantages are:
- Each data mart has its own narrow view of data
- Permits redundant data in every data mart
- Perpetuates inconsistent and irreconcilable data
- Proliferates, unmanageable interfaces.
DWH Life Cycle
Difference between data warehouse and OLTP
|Designed for transaction processing
Store current data only
Store detail data
Normalized (more tables )
Low level managers access
Read, update, insert, delete
Access frequency high and in sec
|Designed for decision support
Store historical data
Store summarized data
De normalized (less tables )
High level manager access
Low, sec to minutes
Types of OLAP
ROLAP: (Relational OLAP):
Applied on relational sources both data and aggregate information store under relational sources.
Ex: BO, congo’s crystal reports, micro strategy
MOLAP: (multi dimensional OLAP):
Here, analysis will be done on multi dimensional applicatioHere, data and aggregate information store under multi dimensional sources.
Eg: congo’s SSAS, hyperion, Micro-strategy etc…………
HOLAP: (hybrid OLAP)
Here data store in relational sources and aggregated values store under multi dimensional sources “cubes”
Eg: cognos, BO, Micro strategic……………
a) Code based.
Development cost, testing, maintenance
Eg: sas base, sas access, tera data utilities, sql, plsql.
b) GUI based etl- informatica, data stage, abinitio etc …..
a) relational sources: ORACLE, SQL SERVER, TD, DB2, INFORMIX, SY BASE, RED BRICK
b) ERP sources: SAP R/3, DEOPLESOFT, J.D. EDWARDS, BANN, RAMCO MARSHALL.
c) Main frames files: COBOL FILES, IMS FILES, JCL FILES, DB2, JCL
d) File sources: FLAT FILES (TEXT FILES), XML FILES
e) Other sources: WEB LOGIC FILES, TIBCO M2 SERVICES, EXCEL SHEET, and PDF ETC……………….
Enroll for Live Instructor Led MSBI Training