Online Analytical Processing

By
 Ouline
 Importance of OLAP

 Key features of OLAP

 OLAP support in SQL

 Why separate OLAP tool?

 Comparison of OLAP with OLTP System

 Multi dimensional Model

 OLAP operators
 OLAP - Introduction
 Useful to answer demand oriented
questions
 Makes Data Synthesis, Analysis and
consolidation quicker, smarter and easier
 Used for efficient retrieval
 Assist in making better business decisions
Codd's Comments on OLAP
Codd: “As enabling as RDBMS have for users,
they were never intended to provide
more powerful functions for data
synthesis, analysis, and consolidation”
(Computer World, 1993).
OLAP makes
Analysis
Data Synthesis
Consolidation - quicker, smarter and easier
 OLAP - Introduction
 Analysis can be made quickly, easily and
smartly by using analytical data.
 OLAP is termed as high capacity, multi user
data manipulation engine specifically
designed to operate on multi dimensional
structures
 For example in market analysis:
 Product, organization, customer-dimensions
 Maturity amount, interest rate etc-measures
On-Line Analytical Processing
OLAP Council definition:
“A Category of software
technology that enables
analysts, managers and
executives to gain insight into
Security
data through fast, consistent,
Customization
interactive access to a wide Interactivity
OLAP

variety of possible views of Visualization

information that has been

transformed from raw data to
reflect the real dimensionality
of the enterprise as
understood by the user.”
On-Line Analytical Processing
 Interactivity – How easy the end user
interacts with the tool?
 Customization – How easy the end user
make changes on the data representation
provided by the tool?
 Visualization – How easy the tool
provide multi-dimensional graphical
representations?
 Security
 OLAP – Nature
 Data contents – Summarized, archived,
derived, historical
 Data organization – by subject
 Data stability – Optimized for complex
queries
 Access frequency – minimum to low
 Access type – Read / aggregate
 Usage – ad hoc / unstructured
 Response time – several seconds to
minutes
Codd's Comments on OLAP
Codd: “As enabling as RDBMS have for users,
they were never intended to provide
more powerful functions for data
synthesis, analysis, and consolidation”
(Computer World, 1993).
OLAP makes
Data synthesis
Analysis
Consolidation quicker, smarter and easier
OLAP – FIVE Key Features
 F – FAST
 A – ANALYSIS
 S – SHARED
 M – Multi Dimensional
 I - Information
 OLAP Support in SQL
 Influenced
by SQL and by
spreadsheets.

 Popularoperators are simple

extensions Group by clause

A common operation is to aggregate

a measure over one or more
dimensions.
 OLAP Queries
 Find total sales.

 Find
total sales for each city, or for
each state.

 Find top five products ranked by

total sales.
Why a Separate OLAP Tool?
 Empowers end-users to perform
their own analysis

 ISbacklog of report requests

reduction

 Easeof use and ease of

understanding

 Views data from all angles

 OLTP
 Transaction processing

 Uses operational databases

 Short-term decisions: fulfill orders,
resolve complaints, provide staffing

 OLAPis an Analytical Processing

system
 Lecture 35
Comparison of OLAP with OLTP System
Comparison of OLTP vs OLAP System
OLTP OLAP

Users Clerk, IT Professional Knowledge worker

Function Day to day operations Decision support

DB design Application oriented Subject oriented

Data Current, up to date- detailed, flat Historical, summarized, multi

relational isolated dimensional integrated,
consolidated
Usage Repetitive ad-hoc

Access Read/write, index/hash on prim key Lots of scans

Unit of work Short, simple transaction Complex query

# records accessed Tens Millions

# users Thousands Hyndreds

DB size 100 MB – GB 100 GB – TB

Metric Transaction throughput Query throughput, response

 Parameters for Systems comparison
 Users and System orientation
 Function
 Design
 Data Contents
 Database Design
 Views and Access Patterns
 Size of the database
 Frequency of Operations
 Performance Metrics
Users and System Orientation
 OLTP
 Is either a clerk or IT professional
 Customer orientation

 OLAP
 Knowledge worker is the main user (Managers,
business executives)
 Market orientation
 Function
 OLTP
 Concentrate on the normal day-to-day operations of
the application.

 OLAP
 Users additional support criteria into account from
the perspective of informational and decision
support requirements.
 Data Contents
 OLTP
 Always maintains the current and up-to-date data
and uses more detailed data for decision making

 OLAP
 Considers the large amounts of historical data and
provides mechanism for summarization and
aggregation
 Stores and manages information at different levels
granularity and will help them to use effectively in
informed decisions.
 View
 OLTP
 Current and latest data within an organization or
department

 OLAP
 Refers to historical data integrated from different
organizations which results in huge amount of data.
Such a huge data is maintained over time.
 Access patterns
 OLTP
 Short and atomic transactions.
 Number of records accessed are tons of thousands
of records

 OLAP
 Uses read-only operations and complex querying.
 Number of records accessed are millions of
thousands of records.
 Database Size
 OLTP
 May take 100 MB to 1GB size

 OLAP
 100 GB to 1 Terabyte of size
 Metric used: query throughput and quick response
time
 Users
 OLTP
 Thousands of users

 OLAP
 Hundreds of users
 Lecture 36
Representation of Data in
Multi Dimensional form and
its associated operations
Why Need Multi dimensional Analysis
 Decision makers are interested to get
complex queries rather than simple
queries.
 Example Query:
How much revenue did the new item X
generate during the last six months,
 Broken down by individual months (m1, m2, …), in
AP State, by an individual stores, (s1, s2, …)
 Broken down by promotions (p1, p2, ….), compared
to estimates, and compared to the previous version
of the product
 Extend comparison further
Why Need Multi dimensional Analysis
 Decision makers are interested to get
Complex queries rather than simple
queries.
 Easy methods to perform complex
analysis along business dimensions
 Can establish using multidimensional
Data
 Analyze data over several dimensions
 Multidimensional Data Model

Timed

Locid

sales
 Collection of numeric measures,

Pid
which depend on a set of
11 1 1 25
dimensions.
 E.g., measure Sales, dimensions 11 2 1 8
Product (key;pid), Location (locid),
11 3 1 15
and Time (timed).
12 1 1 30
12 2 1 20
13 8 10 10
12 3 1 50
Slice
pid

12 30 20 50 13 1 1 8
locid =1
Is shown 25 8 15 13 2 1 10
11
13 3 1 10
1 2 3 locid
11 1 2 35
timed
 OLAP Operators
 Slice
 Dice
 Drilldown
 Roll up
 pivot
 OLAP Operators
 Part of most of OLAP Tools

 Provide flexibility to view data in

different perspectives

 Assist
the user for interactive
analysis
 OLAP Operators Summary
Operator Purpose Description
Slice Focus attention on a Replace a dimension with a
subset of dimensions single member value or
with a summary of its
measure values

Dice Focus attention on a Replace a dimension with a

subset of member subset of members
values
Drill-down Obtain more detail Navigate from a more
about a dimension general level to a more
specific level
Roll-up Summarize about a Navigate from a more
dimension specific level to a more
general level
Pivot Present data in a Rearrange the dimensions
different order in a data cube
On-Line Analytical Processing
 OLAP functionality provides basic
operators for interactivity and better
visualization for performing dynamic
multi-dimensional analysis.

 OLAP ensures fast, flexible access for

summarizing data.

 OLAP supports analytical and navigational

activities.
 Discussion on Four Views
 All
four views are combined into
single structure called Data cube

 Allfour views access the same data

in different manner