Scribd
Upload
43 views

Introduction To SQL Server Analysis Services 2008 Olap: by Kristin Ferrier

This document provides an introduction to SQL Server Analysis Services 2008 OLAP. It discusses how OLAP cubes store data in multidimensional structures optimized for analytics and reporting. It describes how SSAS is Microsoft's OLAP tool, how data is processed and queried using MDX, and key concepts like dimensions, measures, and cube storage modes. Processing types and partitioning techniques are also covered to improve query and processing performance.

Uploaded by

viswakumar
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
43 views

Introduction To SQL Server Analysis Services 2008 Olap: by Kristin Ferrier

This document provides an introduction to SQL Server Analysis Services 2008 OLAP. It discusses how OLAP cubes store data in multidimensional structures optimized for analytics and reporting. It describes how SSAS is Microsoft's OLAP tool, how data is processed and queried using MDX, and key concepts like dimensions, measures, and cube storage modes. Processing types and partitioning techniques are also covered to improve query and processing performance.

Uploaded by

viswakumar
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 23

Introduction to SQL Server Analysis

Services 2008
OLAP
by Kristin Ferrier
Transactional Databases
• Stored in relational database engines
• Data is typically in normalized form
• Optimized for insert, update, and deletion
operations
• Typically used by business applications
OLAP Cubes
• Online Analytic Processing (OLAP)
• Optimized for complex analytics and
reporting
• Data lives in multidimensional structures
• Allow answer questions like
o What were my sales revenue by product, by
store, and sales date?
What is SSAS OLAP?
• Microsoft’s Online Analytic Processing
(OLAP) tool
• Started with MS SQL Server 7.0
• Queried using MDX
Microsoft BI Example
Implementation
Star Schema
SSAS Versions
• SSAS started with version 7.0 when MS
bought Panorama’s OLAP engine
• SSAS 2000 <> SSAS 2005
• SSAS 2005 similar to SSAS 2008
SSAS Development Tools
• Business Intelligence Development Studio
(BIDS)
• BIDSHelper (add-in from Codeplex)
MDX
• Multidimensional Expressions
• A query language for OLAP databases
• Developed by SQL Server engineers
• Now accepted by most major OLAP
vendors, including:
o Microsoft
o Cognos
o Hyperion
o Business Objects
MDX Popular Books
• Fast Track to MDX (for SQL Server 2000)
Second Edition by Whitehorn, Zare, and
Pasumansky
• MDX Solutions: With Microsoft SQL
Server Analysis Services 2005 and
Hyperion Essbase by Spofford, Harinath,
Webb, and Hai Huang
SSAS 2008 Datasources
• SQL Server (7.0 – 2008)
• Oracle 9.0
• IBM DB2 8.1
• Teradata v2R6
• MS Access
• Note: Certain sources are only accepted by certain
levels of SSAS 2008
Data Source Views
• Abstraction layer
• Logical data model of one or more data
sources
• Primarily includes tables, named queries,
relationships, and named calculations.
Dimensions
• Business entities by which you want to
analyze your measures
• In question, what is my sales revenue by
Sales Date, Employee, and Product?
o Sales Date, Employee, and Product are the
dimensions
Measures
• The data being analyzed
• In question, what are my Sales Revenue
and Quantity Sold by Sales Date,
Employee, and Product?
o Sales Revenue and Quantity Sold are both
measures
• Measures of the same granularity exist in
the same measure group
Cube Storage Modes
• Cubes have up to 3 logical storage layers
o Metadata
o Aggregations (optional)
o Data
• There are three storage modes
o MOLAP
o HOLAP
o ROLAP
Cube Storage Modes cont.
• MOLAP
o Metadata, Aggregations, and Data all live on SSAS server
o Considered fastest for query time
• HOLAP (rarely used)
o Metadata and Aggregations live on SSAS Server
o Data lives in relational database
• ROLAP
o Metadata lives on SSAS server
o Data and aggregations live in relational database
o Relational tables are created to hold the aggregations
o Low latency
Demo

Sales Cube
Processing
• Processing is how data is brought in from
relational data sources to the SSAS cube
• Executed using XMLA
• Many places from which to execute
XMLA:
o SSMS
o Visual Studio 2008
o SSIS Package
o SQL Server Agent
o Command Line
Processing Types
• Process Full – Can be applied to any SSAS object.
Complete reprocess. Process Full is needed if
metadata changes.
• Process Update – For dimensions only. Applies
member updates, insertions, and deletions without
invalidating the cubes using the dimension.
• Process Add – Adds new data only. Applies to
dimensions and partitions.
• Process Data – Loads the object’s data only. Does
not build indexes or aggregations.
• Process Index – Builds indexes and aggregations on
data already there. Applies to dimensions, cubes,
measure groups, and partitions.
Partitioning
• Improved Query Performance
• Improved Processing Performance
Partitioning cont.
• Recommended partition size of 10 – 20
million rows
• Example: A year of sales contains 15
million records.
• If partition by year, queries by time will
only scan the partition(s) with data for the
time range of the query.
• Processing can be done only against the
partitions receiving new data.
My Favorite SSAS Blogs
• Chris Webb -
http://cwebbbi.spaces.live.com
• Mosha Pasumansky -
http://sqlblog.com/blogs/mosha
Contact Me
• [email protected]
• http://kferrier.blogspot.com/

You might also like