| Date post: | 11-Oct-2015 |
| Category: |
Documents |
| Upload: | rahul-kale |
| View: | 15 times |
| Download: | 0 times |
of 169
DATA WAREHOUSING ANDDATA MINING
S. SudarshanKrithi Ramamritham
IIT Bombay
[email protected]@cse.iitb.ernet.in
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
2Course Overviewz The course:
what and how
z 0. Introductionz I. Data Warehousingz II. Decision Support
and OLAPz III. Data Miningz IV. Looking Ahead
z Demos and Labs
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
30. Introduction
z Data Warehousing, OLAP and data mining: what and why (now)?
z Relation to OLTPz A case study
z demos, labs
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
4Which are our lowest/highest margin
customers ?Who are my customers
and what products are they buying?
Which customers are most likely to go to the competition ?
What impact will new products/services
have on revenue and margins?
What product prom--otions have the biggest
impact on revenue?
What is the most effective distribution
channel?
A producer wants to know.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
5Data, Data everywhereyet ... z I cant find the data I need
y data is scattered over the network
y many versions, subtle differences
z I cant get the data I needy need an expert to get the data
z I cant understand the data I foundy available data poorly documented
z I cant use the data I foundy results are unexpectedy data needs to be transformed
from one form to other
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
6What is a Data Warehouse?
A single, complete and consistent store of data obtained from a variety of different sources made available to end users in a what they can understand and use in a business context.
[Barry Devlin]
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
7What are the users saying...
z Data should be integrated across the enterprise
z Summary data has a real value to the organization
z Historical data holds the key to understanding data over time
z What-if capabilities are required
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
8What is Data Warehousing?
A process of transforming data into information and making it available to users in a timely enough manner to make a difference
[Forrester Research, April 1996]Data
Information
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
9Evolution
z 60s: Batch reportsy hard to find and analyze informationy inflexible and expensive, reprogram every new
request
z 70s: Terminal-based DSS and EIS (executive information systems)y still inflexible, not integrated with desktop tools
z 80s: Desktop data access and analysis toolsy query tools, spreadsheets, GUIsy easier to use, but only access operational databases
z 90s: Data warehousing with integrated OLAP engines and tools
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
10
Warehouses are Very Large Databases
35%
30%
25%
20%
15%
10%
5%
0%5GB
5-9GB10-19GB 50-99GB 250-499GB
20-49GB 100-249GB 500GB-1TB
InitialProjected 2Q96
Source: META Group, Inc.
Res
pond
ents
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
11
Very Large Data Bases
z Terabytes -- 10^12 bytes:
z Petabytes -- 10^15 bytes:
z Exabytes -- 10^18 bytes:
z Zettabytes -- 10^21 bytes:
z Zottabytes -- 10^24 bytes:
Walmart -- 24 Terabytes
Geographic Information Systems
National Medical Records
Weather images
Intelligence Agency Videos
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
12
Data Warehousing -- It is a process
z Technique for assembling and managing data from various sources for the purpose of answering business questions. Thus making decisions that were not previous possible
z A decision support database maintained separately from the organizations operational database
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
13
Data Warehouse
z A data warehouse is a y subject-orientedy integratedy time-varyingy non-volatile
collection of data that is used primarily in organizational decision making.
-- Bill Inmon, Building the Data Warehouse 1996
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
14
Explorers, Farmers and Tourists
Explorers: Seek out the unknown and previously unsuspected rewards hiding in the detailed data
Farmers: Harvest informationfrom known access paths
Tourists: Browse information harvested by farmers
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
15
Data Warehouse Architecture
Data Warehouse Engine
Optimized Loader
ExtractionCleansing
AnalyzeQuery
Metadata Repository
RelationalDatabases
LegacyData
Purchased Data
ERPSystems
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
16
Data Warehouse for Decision Support & OLAP
z Putting Information technology to help the knowledge worker make faster and better decisionsy Which of my customers are most likely to go
to the competition?
y What product promotions have the biggest impact on revenue?
y How did the share price of software companies correlate with profits over last 10 years?
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
17
Decision Support
z Used to manage and control businessz Data is historical or point-in-timez Optimized for inquiry rather than updatez Use of the system is loosely defined and
can be ad-hoc
z Used by managers and end-users to understand the business and make judgements
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
18
Data Mining works with Warehouse Data
z Data Warehousing provides the Enterprise with a memory
z Data Mining provides the Enterprise with intelligence
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
19
We want to know ...z Given a database of 100,000 names, which persons are the
least likely to default on their credit cards?
z Which types of transactions are likely to be fraudulent given the demographics and transactional history of a particular customer?
z If I raise the price of my product by Rs. 2, what is the effect on my ROI?
z If I offer only 2,500 airline miles as an incentive to purchase rather than 5,000, how many lost responses will result?
z If I emphasize ease-of-use of the product as opposed to its technical capabilities, what will be the net effect on my revenues?
z Which of my customers are likely to be the most loyal?
Data Mining helps extract such information
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
20
Application Areas
Industry ApplicationFinance Credit Card AnalysisInsurance Claims, Fraud Analysis
Telecommunication Call record analysisTransport Logistics managementConsumer goods promotion analysisData Service providersValue added dataUtilities Power usage analysis
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
21
Data Mining in Use
z The US Government uses Data Mining to track fraud
z A Supermarket becomes an information broker
z Basketball teams use it to track game strategy
z Cross Sellingz Warranty Claims Routingz Holding on to Good Customersz Weeding out Bad Customers
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
22
What makes data mining possible?
z Advances in the following areas are making data mining deployable:y data warehousing y better and more data (i.e., operational,
behavioral, and demographic) y the emergence of easily deployed data
mining tools and y the advent of new data mining
techniques. -- Gartner Group
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
23
Why Separate Data Warehouse?
z Performancey Op dbs designed & tuned for known txs & workloads.y Complex OLAP queries would degrade perf. for op txs.y Special data organization, access & implementation
methods needed for multidimensional views & queries.
z Functiony Missing data: Decision support requires historical data, which
op dbs do not typically maintain.y Data consolidation: Decision support requires consolidation
(aggregation, summarization) of data from many heterogeneous sources: op dbs, external sources.
y Data quality: Different sources typically use inconsistent data representations, codes, and formats which have to be reconciled.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
24
What are Operational Systems?
z They are OLTP systemsz Run mission critical
applicationsz Need to work with
stringent performance requirements for routine tasks
z Used to run a business!
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
25
RDBMS used for OLTP
z Database Systems have been used traditionally for OLTPy clerical data processing tasksy detailed, up to date datay structured repetitive tasksy read/update a few recordsy isolation, recovery and integrity are
critical
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
26
Operational Systems
z Run the business in real timez Based on up-to-the-second dataz Optimized to handle large
numbers of simple read/write transactions
z Optimized for fast response to predefined transactions
z Used by people who deal with customers, products -- clerks, salespeople etc.
z They are increasingly used by customers
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
27
Examples of Operational Data
Data IndustryUsage Technology Volumes
CustomerFile
All TrackCustomerDetails
Legacy application, flatfiles, main frames
Small-medium
AccountBalance
Finance Controlaccountactivities
Legacy applications,hierarchical databases,mainframe
Large
Point-of-Sale data
Retail Generatebills, managestock
ERP, Client/Server,relational databases
Very Large
CallRecord
Telecomm-unications
Billing Legacy application,hierarchical database,mainframe
Very Large
ProductionRecord
Manufact-uring
ControlProduction
ERP,relational databases,AS/400
Medium
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
So, whats different?
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
29
Application-Orientation vs. Subject-Orientation
Application-Orientation
Operational Database
LoansCredit Card
Trust
Savings
Subject-Orientation
DataWarehouse
Customer
VendorProduct
Activity
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
30
OLTP vs. Data Warehouse
z OLTP systems are tuned for known transactions and workloads while workload is not known a priori in a data warehouse
z Special data organization, access methods and implementation methods are needed to support data warehouse queries (typically multidimensional queries)y e.g., average amount spent on phone calls
between 9AM-5PM in Pune during the month of December
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
31
OLTP vs Data Warehouse
z OLTPy Application
Orientedy Used to run
businessy Detailed datay Current up to datey Isolated Datay Repetitive accessy Clerical User
z Warehouse (DSS)y Subject Orientedy Used to analyze
businessy Summarized and
refinedy Snapshot datay Integrated Datay Ad-hoc accessy Knowledge User
(Manager)
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
32
OLTP vs Data Warehouse
z OLTPy Performance Sensitivey Few Records accessed at
a time (tens)
y Read/Update Access
y No data redundancyy Database Size 100MB
-100 GB
z Data Warehousey Performance relaxedy Large volumes accessed
at a time(millions)y Mostly Read (Batch
Update)y Redundancy presenty Database Size
100 GB - few terabytes
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
33
OLTP vs Data Warehouse
z OLTPy Transaction
throughput is the performance metric
y Thousands of usersy Managed in entirety
z Data Warehousey Query throughput is
the performance metric
y Hundreds of usersy Managed by
subsets
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
34
To summarize ...
z OLTP Systems are used to run a business
z The Data Warehouse helps to optimize the business
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
35
Why Now?
z Data is being producedz ERP provides clean dataz The computing power is availablez The computing power is affordablez The competitive pressures are strongz Commercial products are available
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
36
Myths surrounding OLAP Servers and Data Marts
z Data marts and OLAP servers are departmental solutions supporting a handful of users
z Million dollar massively parallel hardware is needed to deliver fast time for complex queries
z OLAP servers require massive and unwieldy indices
z Complex OLAP queries clog the network with data
z Data warehouses must be at least 100 GB to be effective
Source -- Arbor Software Home Page
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
37
Wal*Mart Case Study
z Founded by Sam Waltonz One the largest Super Market Chains
in the US
z Wal*Mart: 2000+ Retail Stores z SAM's Clubs 100+Wholesalers Stores
x This case study is from Felipe Carinos (NCR Teradata) presentation made at Stanford Database Seminar
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
38
Old Retail Paradigm
z Wal*Marty Inventory
Management y Merchandise Accounts
Payable y Purchasing y Supplier Promotions:
National, Region, Store Level
z Suppliers y Accept Orders y Promote Products y Provide special
Incentives y Monitor and Track
The Incentives y Bill and Collect
Receivables y Estimate Retailer
Demands
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
39
New (Just-In-Time) Retail Paradigm
z No more dealsz Shelf-Pass Through (POS Application)
y One Unit Pricex Suppliers paid once a week on ACTUAL items sold
y Wal*Mart Managerx Daily Inventory Restockx Suppliers (sometimes SameDay) ship to Wal*Mart
z Warehouse-Pass Throughy Stock some Large Items
x Delivery may come from suppliery Distribution Center
x Suppliers merchandise unloaded directly onto Wal*Mart Trucks
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
40
Wal*Mart System
z NCR 5100M 96 Nodes;
z Number of Rows:z Historical Data:z New Daily Volume:
z Number of Users:z Number of Queries:
24 TB Raw Disk; 700 - 1000 Pentium CPUs
> 5 Billions65 weeks (5 Quarters)Current Apps: 75 MillionNew Apps: 100 Million +Thousands60,000 per week
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
41
Course Overview
z 0. Introductionz I. Data Warehousingz II. Decision Support
and OLAPz III. Data Miningz IV. Looking Ahead
z Demos and Labs
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
42
I. Data Warehouses:Architecture, Design & Construction
z DW Architecturez Loading, refreshingz Structuring/Modelingz DWs and Data Martsz Query Processing
z demos, labs
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
43
Data Warehouse Architecture
Data Warehouse Engine
Optimized Loader
ExtractionCleansing
AnalyzeQuery
Metadata Repository
RelationalDatabases
LegacyData
Purchased Data
ERPSystems
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
44
Components of the Warehouse
z Data Extraction and Loadingz The Warehouse z Analyze and Query -- OLAP Toolsz Metadata
z Data Mining tools
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
Loading the Warehouse
Cleaning the data before it is loaded
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
46
Source Data
z Typically host based, legacy applicationsy Customized applications, COBOL,
3GL, 4GLz Point of Contact Devices
y POS, ATM, Call switchesz External Sources
y Nielsens, Acxiom, CMIE, Vendors, Partners
Sequential Legacy Relational ExternalOperational/Source Data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
47
Data Quality - The Reality
z Tempting to think creating a data warehouse is simply extracting operational data and entering into a data warehouse
z Nothing could be farther from the truth
z Warehouse data comes from disparate questionable sources
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
48
Data Quality - The Reality
z Legacy systems no longer documentedz Outside sources with questionable quality
proceduresz Production systems with no built in
integrity checks and no integrationy Operational systems are usually designed to
solve a specific business problem and are rarely developed to a a corporate plan
x And get it done quickly, we do not have time to worry about corporate standards...
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
49
Data Integration Across Sources
Trust Credit cardSavings Loans
Same data different name
Different data Same name
Data found here nowhere else
Different keyssame data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
50
Data Transformation Exampleen
codi
ngun
itfie
ld
appl A - balanceappl B - balappl C - currbalappl D - balcurr
appl A - pipeline - cmappl B - pipeline - inappl C - pipeline - feetappl D - pipeline - yds
appl A - m,fappl B - 1,0appl C - x,yappl D - male, female
Data Warehouse
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
51
Data Integrity Problems
z Same person, different spellingsy Agarwal, Agrawal, Aggarwal etc...
z Multiple ways to denote company namey Persistent Systems, PSPL, Persistent Pvt.
LTD.z Use of different names
y mumbai, bombayz Different account numbers generated by
different applications for the same customerz Required fields left blankz Invalid product codes collected at point of sale
y manual entry leads to mistakesy in case of a problem use 9999999
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
52
Data Transformation Terms
z Extractingz Conditioningz Scrubbingz Mergingz Householding
z Enrichmentz Scoringz Loadingz Validatingz Delta Updating
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
53
Data Transformation Terms
z Extractingy Capture of data from operational source in
as is status
y Sources for data generally in legacy mainframes in VSAM, IMS, IDMS, DB2; more data today in relational databases on Unix
z Conditioningy The conversion of data types from the source
to the target data store (warehouse) -- always a relational database
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
54
Data Transformation Terms
z Householdingy Identifying all members of a household
(living at the same address)y Ensures only one mail is sent to a
householdy Can result in substantial savings: 1 lakh
catalogues at Rs. 50 each costs Rs. 50 lakhs. A 2% savings would save Rs. 1 lakh.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
55
Data Transformation Terms
z Enrichmenty Bring data from external sources to
augment/enrich operational data. Data sources include Dunn and Bradstreet, A. C. Nielsen, CMIE, IMRA etc...
z Scoring y computation of a probability of an
event. e.g..., chance that a customer will defect to AT&T from MCI, chance that a customer is likely to buy a new product
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
56
Loads
z After extracting, scrubbing, cleaning, validating etc. need to load the data into the warehouse
z Issuesy huge volumes of data to be loadedy small time window available when warehouse can be
taken off line (usually nights)y when to build index and summary tablesy allow system administrators to monitor, cancel, resume,
change load ratesy Recover gracefully -- restart after failure from where
you were and without loss of data integrity
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
57
Load Techniques
z Use SQL to append or insert new datay record at a time interfacey will lead to random disk I/Os
z Use batch load utility
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
58
Load Taxonomy
z Incremental versus Full loadsz Online versus Offline loads
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
59
Refresh
z Propagate updates on source data to the warehouse
z Issues:y when to refreshy how to refresh -- refresh techniques
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
60
When to Refresh?
z periodically (e.g., every night, every week) or after significant events
z on every update: not warranted unless warehouse data require current data (up to the minute stock quotes)
z refresh policy set by administrator based on user needs and traffic
z possibly different policies for different sources
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
61
Refresh Techniques
z Full Extract from base tablesy read entire source table: too expensivey maybe the only choice for legacy
systems
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
62
How To Detect Changes
z Create a snapshot log table to record ids of updated rows of source data and timestamp
z Detect changes by:y Defining after row triggers to update
snapshot log when source table changesy Using regular transaction log to detect
changes to source data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
63
Data Extraction and Cleansing
z Extract data from existing operational and legacy data
z Issues:y Sources of data for the warehousey Data quality at the sourcesy Merging different data sourcesy Data Transformationy How to propagate updates (on the sources) to
the warehousey Terabytes of data to be loaded
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
64
Scrubbing Data
z Sophisticated transformation tools.
z Used for cleaning the quality of data
z Clean data is vital for the success of the warehouse
z Exampley Seshadri, Sheshadri,
Sesadri, Seshadri S., Srinivasan Seshadri, etc. are the same person
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
65
Scrubbing Tools
z Apertus -- Enterprise/Integrator z Vality -- IPEz Postal Soft
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
Structuring/Modeling Issues
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
67
Data -- Heart of the Data Warehouse
z Heart of the data warehouse is the data itself!
z Single version of the truthz Corporate memoryz Data is organized in a way that
represents business -- subject orientation
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
68
Data Warehouse Structure
z Subject Orientation -- customer, product, policy, account etc... A subject may be implemented as a set of related tables. E.g., customer may be five tables
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
69
Data Warehouse Structure
y base customer (1985-87)x custid, from date, to date, name, phone, dob
y base customer (1988-90)x custid, from date, to date, name, credit rating,
employer
y customer activity (1986-89) -- monthly summary
y customer activity detail (1987-89)x custid, activity date, amount, clerk id, order no
y customer activity detail (1990-91)x custid, activity date, amount, line item no, order no
Time is Time is part of part of key of key of each tableeach table
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
70
Data Granularity in Warehouse
z Summarized data storedy reduce storage costsy reduce cpu usagey increases performance since smaller
number of records to be processedy design around traditional high level
reporting needsy tradeoff with volume of data to be
stored and detailed usage of data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
71
Granularity in Warehouse
z Can not answer some questions with summarized datay Did Anand call Seshadri last month? Not
possible to answer if total duration of calls by Anand over a month is only maintained and individual call details are not.
z Detailed data too voluminous
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
72
Granularity in Warehouse
z Tradeoff is to have dual level of granularityy Store summary data on disks
x 95% of DSS processing done against this data
y Store detail on tapesx 5% of DSS processing against this data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
73
Vertical Partitioning
Frequentlyaccessed Rarely
accessed
Smaller tableand so less I/O
Acct.No Name BalanceDate Opened
InterestRate Address
Acct.No Balance
Acct.No Name Date Opened
InterestRate Address
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
74
Derived Data
z Introduction of derived (calculated data) may often help
z Have seen this in the context of dual levels of granularity
z Can keep auxiliary views and indexes to speed up query processing
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
75
Schema Design
z Database organizationy must look like businessy must be recognizable by business usery approachable by business usery Must be simple
z Schema Typesy Star Schemay Fact Constellation Schemay Snowflake schema
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
76
Dimension Tables
z Dimension tablesy Define business in terms already
familiar to usersy Wide rows with lots of descriptive texty Small tables (about a million rows) y Joined to fact table by a foreign keyy heavily indexedy typical dimensions
x time periods, geographic region (markets, cities), products, customers, salesperson, etc.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
77
Fact Table
z Central tabley mostly raw numeric itemsy narrow rows, a few columns at mosty large number of rows (millions to a
billion)y Access via dimensions
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
78
Star Schema
z A single fact table and for each dimension one dimension table
z Does not capture hierarchies directly
T ime
prod
cust
city
fact
date, custno, prodno, cityname, ...
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
79
Snowflake schema
z Represent dimensional hierarchy directly by normalizing tables.
z Easy to maintain and saves storageT ime
prod
cust
city
fact
date, custno, prodno, cityname, ...
region
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
80
Fact Constellation
z Fact Constellationy Multiple fact tables that share many
dimension tablesy Booking and Checkout may share many
dimension tables in the hotel industry
Hotels
Travel Agents
Promotion
Room TypeCustomer
Booking
Checkout
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
81
De-normalization
z Normalization in a data warehouse may lead to lots of small tables
z Can lead to excessive I/Os since many tables have to be accessed
z De-normalization is the answer especially since updates are rare
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
82
Creating Arrays
z Many times each occurrence of a sequence of data is in a different physical location
z Beneficial to collect all occurrences together and store as an array in a single row
z Makes sense only if there are a stable number of occurrences which are accessed together
z In a data warehouse, such situations arise naturally due to time based orientationy can create an array by month
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
83
Selective Redundancy
z Description of an item can be stored redundantly with order table -- most often item description is also accessed with order table
z Updates have to be careful
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
84
Partitioning
z Breaking data into several physical units that can be handled separately
z Not a question of whether to do it in data warehouses but how to do it
z Granularity and partitioning are key to effective implementation of a warehouse
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
85
Why Partition?
z Flexibility in managing dataz Smaller physical units allow
y easy restructuringy free indexingy sequential scans if neededy easy reorganizationy easy recoveryy easy monitoring
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
86
Criterion for Partitioning
z Typically partitioned by y datey line of businessy geographyy organizational unity any combination of above
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
87
Where to Partition?
z Application level or DBMS levelz Makes sense to partition at
application levely Allows different definition for each year
x Important since warehouse spans many years and as business evolves definition changes
y Allows data to be moved between processing complexes easily
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
Data Warehouse vs. Data Marts
What comes first
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
89
From the Data Warehouse to Data Marts
DepartmentallyStructured
IndividuallyStructured
Data WarehouseOrganizationallyStructured
Less
More
HistoryNormalizedDetailed
Data
Information
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
90
Data Warehouse and Data MartsOLAPData MartLightly summarizedDepartmentally structured
Organizationally structuredAtomicDetailed Data Warehouse Data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
91
Characteristics of the Departmental Data Mart
z OLAPz Smallz Flexiblez Customized by
Departmentz Source is
departmentally structured data warehouse
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
92
Techniques for Creating Departmental Data Mart
z OLAPz Subsetz Summarizedz Supersetz Indexedz Arrayed
Sales Mktg.Finance
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
93
Data Mart Centric
Data Marts
Data Sources
Data Warehouse
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
94
Problems with Data Mart Centric Solution
If you end up creating multiple warehouses, integrating them is a problem
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
95
True Warehouse
Data Marts
Data Sources
Data Warehouse
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
96
Query Processing
z Indexing
z Pre computed views/aggregates
z SQL extensions
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
97
Indexing Techniques
z Exploiting indexes to reduce scanning of data is of crucial importance
z Bitmap Indexesz Join Indexesz Other Issues
y Text indexingy Parallelizing and sequencing of index
builds and incremental updates
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
98
Indexing Techniques
z Bitmap index:y A collection of bitmaps -- one for each
distinct value of the columny Each bitmap has N bits where N is the
number of rows in the tabley A bit corresponding to a value v for a
row r is set if and only if r has the value for the indexed attribute
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
99
BitMap Indexes
z An alternative representation of RID-listz Specially advantageous for low-cardinality
domainsz Represent each row of a table by a bit
and the table as a bit vectorz There is a distinct bit vector Bv for each
value v for the domainz Example: the attribute sex has values M
and F. A table of 100 million people needs 2 lists of 100 million bits
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
100Customer Query : select * from customer where
gender = F and vote = Y
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
Bitmap Index
M
F
F
F
F
M
Y
Y
Y
N
N
N
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
101
Bit Map Index
Cust Region RatingC1 N HC2 S MC3 W LC4 W HC5 S LC6 W LC7 N H
Base TableBase TableRow ID N S E W
1 1 0 0 02 0 1 0 03 0 0 0 14 0 0 0 15 0 1 0 06 0 0 0 17 1 0 0 0
Row ID H M L1 1 0 02 0 1 03 0 0 04 0 0 05 0 1 06 0 0 07 1 0 0
Rating IndexRating IndexRegion IndexRegion Index
Customers whereCustomers where Region = WRegion = W Rating = MRating = MAndAnd
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
102
BitMap Indexes
z Comparison, join and aggregation operations are reduced to bit arithmetic with dramatic improvement in processing time
z Significant reduction in space and I/O (30:1)z Adapted for higher cardinality domains as well.z Compression (e.g., run-length encoding)
exploitedz Products that support bitmaps: Model 204,
TargetIndex (Redbrick), IQ (Sybase), Oracle 7.3
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
103
Join Indexes
z Pre-computed joinsz A join index between a fact table and a
dimension table correlates a dimension tuple with the fact tuples that have the same value on the common dimensional attributey e.g., a join index on city dimension of calls
fact tabley correlates for each city the calls (in the calls
table) from that city
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
104
Join Indexes
z Join indexes can also span multiple dimension tablesy e.g., a join index on city and time
dimension of calls fact table
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
105
Star Join Processing
z Use join indexes to join dimension and fact table
CallsC+T
C+T+L
C+T+L+P
Time
Loca-tion
Plan
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
106
Optimized Star Join Processing
Time
Loca-tion
Plan
Calls
Virtual Cross Productof T, L and P
Apply Selections
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
107
Bitmapped Join Processing
AND
Time
Loca-tion
Plan
Calls
Calls
Calls
Bitmaps101
001
110
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
108
Intelligent Scan
z Piggyback multiple scans of a relation (Redbrick)y piggybacking also done if second scan
starts a little while after the first scan
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
109
Parallel Query Processing
z Three forms of parallelismy Independenty Pipelinedy Partitioned and partition and replicate
z Deterrents to parallelismy startup y communication
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
110
Parallel Query Processing
z Partitioned Datay Parallel scansy Yields I/O parallelism
z Parallel algorithms for relational operatorsy Joins, Aggregates, Sort
z Parallel Utilitiesy Load, Archive, Update, Parse, Checkpoint,
Recovery z Parallel Query Optimization
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
111
Pre-computed Aggregates
z Keep aggregated data for efficiency (pre-computed queries)
z Questionsy Which aggregates to compute?y How to update aggregates?y How to use pre-computed aggregates
in queries?
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
112
Pre-computed Aggregates
z Aggregated table can be maintained by they warehouse servery middle tier y client applications
z Pre-computed aggregates -- special case of materialized views -- same questions and issues remain
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
113
SQL Extensions
z Extended family of aggregate functionsy rank (top 10 customers)y percentile (top 30% of customers)y median, modey Object Relational Systems allow
addition of new aggregate functions
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
114
SQL Extensions
z Reporting featuresy running total, cumulative totals
z Cube operatory group by on all subsets of a set of
attributes (month,city)y redundant scan and sorting of data can
be avoided
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
115
Red Brick has Extended set of Aggregates
z Select month, dollars, cume(dollars) as run_dollars, weight, cume(weight) as run_weightsfrom sales, market, product, period twhere year = 1993and product like Columbian%and city like San Fr%order by t.perkey
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
116
RISQL (Red Brick Systems) Extensions
z Aggregatesy CUMEy MOVINGAVGy MOVINGSUMy RANKy TERTILEy RATIOTOREPORT
z Calculating Row Subtotalsy BREAK BY
z Sophisticated Date Time Supporty DATEDIFF
z Using SubQueries in calculations
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
117
Using SubQueries in Calculationsselect product, dollars as jun97_sales, (select sum(s1.dollars)from market mi, product pi, period, ti, sales siwhere pi.product = product.productand ti.year = period.yearand mi.city = market.city) as total97_sales,100 * dollars/(select sum(s1.dollars)from market mi, product pi, period, ti, sales siwhere pi.product = product.productand ti.year = period.yearand mi.city = market.city) as percent_of_yrfrom market, product, period, saleswhere year = 1997and month = June and city like Ahmed%order by product;
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
118
Course Overviewz The course:
what and how
z 0. Introductionz I. Data Warehousingz II. Decision Support
and OLAPz III. Data Miningz IV. Looking Ahead
z Demos and Labs
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
II. On-Line Analytical Processing (OLAP)
Making Decision Support Possible
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
120
Limitations of SQL
A Freshman in Business needs a Ph.D. in SQL
-- Ralph Kimball
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
121
Typical OLAP Queries
z Write a multi-table join to compare sales for each product line YTD this year vs. last year.
z Repeat the above process to find the top 5 product contributors to margin.
z Repeat the above process to find the sales of a product line to new vs. existing customers.
z Repeat the above process to find the customers that have had negative sales growth.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
122
* Reference: http://www.arborsoft.com/essbase/wht_ppr/coddTOC.html* Reference: http://www.arborsoft.com/essbase/wht_ppr/coddTOC.html
What Is OLAP?
z Online Analytical Processing - coined by EF Codd in 1994 paper contracted by Arbor Software*
z Generally synonymous with earlier terms such as Decisions Support, Business Intelligence, Executive Information System
z OLAP = Multidimensional Databasez MOLAP: Multidimensional OLAP (Arbor Essbase,
Oracle Express)z ROLAP: Relational OLAP (Informix MetaCube,
Microstrategy DSS Agent)
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
123
The OLAP Market z Rapid growth in the enterprise market
y 1995: $700 Milliony 1997: $2.1 Billion
z Significant consolidation activity among major DBMS vendorsy 10/94: Sybase acquires ExpressWayy 7/95: Oracle acquires Express y 11/95: Informix acquires Metacubey 1/97: Arbor partners up with IBMy 10/96: Microsoft acquires Panorama
z Result: OLAP shifted from small vertical niche to mainstream DBMS category
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
124
Strengths of OLAP
z It is a powerful visualization paradigm
z It provides fast, interactive response times
z It is good for analyzing time series
z It can be useful to find some clusters and outliers
z Many vendors offer OLAP tools
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
125
Nigel Pendse, Richard Creath - The OLAP ReportNigel Pendse, Richard Creath - The OLAP Report
OLAP Is FASMI
z Fastz Analysisz Sharedz Multidimensionalz Information
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
126MonthMonth
1 1 22 3 3 4 4 776 6 5 5
Prod
uct
Prod
uct
Toothpaste Toothpaste
JuiceJuiceColaColaMilk Milk
CreamCream
Soap Soap
Regio
n
Regio
n
WWS S
N N
Dimensions: Dimensions: Product, Region, TimeProduct, Region, TimeHierarchical summarization pathsHierarchical summarization paths
Product Product Region Region TimeTimeIndustry Country YearIndustry Country Year
Category Region Quarter Category Region Quarter
Product City Month WeekProduct City Month Week
Office DayOffice Day
Multi-dimensional Data
z HeyI sold $100M worth of goods
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
127
Data Cube Lattice
z Cube latticey ABC
AB AC BC A B C none
z Can materialize some groupbys, compute others on demand
z Question: which groupbys to materialze?z Question: what indices to createz Question: how to organize data (chunks, etc)
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
128
Visualizing Neighbors is simpler
1 2 3 4 5 6 7 8AprMayJunJulAugSepOctNovDecJanFebMar
Month Store SalesApr 1Apr 2Apr 3Apr 4Apr 5Apr 6Apr 7Apr 8May 1May 2May 3May 4May 5May 6May 7May 8Jun 1Jun 2
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
129
A Visual Operation: Pivot (Rotate)
1010
4747
30301212
JuiceJuice
ColaCola
Milk Milk
CreamCream
NYNYLALA
SFSF
3/1 3/2 3/3 3/43/1 3/2 3/3 3/4DateDate
Month
Month
Regi
onRe
gion
ProductProduct
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
130
Slicing and Dicing
Product
Sales Channel
Regio
ns
Retail Direct Special
Household
Telecomm
Video
Audio IndiaFar East
Europe
The Telecomm Slice
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
131
Roll-up and Drill Down
z Sales Channelz Regionz Countryz State z Location Addressz Sales
Representative
Roll
Up
Higher Level ofAggregation
Low-levelDetails
Drill-D
own
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
132
Nature of OLAP Analysisz Aggregation -- (total sales,
percent-to-total)z Comparison -- Budget vs.
Expensesz Ranking -- Top 10, quartile
analysisz Access to detailed and
aggregate dataz Complex criteria
specificationz Visualization
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
133
Organizationally Structured Data
z Different Departments look at the same detailed data in different ways. Without the detailed, organizationally structured data as a foundation, there is no reconcilability of data
marketing
manufacturing
sales
finance
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
134
Multidimensional Spreadsheetsz Analysts need
spreadsheets that supporty pivot tables (cross-tabs)y drill-down and roll-upy slice and dicey sorty selectionsy derived attributes
z Popular in retail domain
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
135
OLAP - Data Cube
z Idea: analysts need to group data in many different waysy eg. Sales(region, product, prodtype, prodstyle,
date, saleamount)y saleamount is a measure attribute, rest are
dimension attributesy groupby every subset of the other attributes
x materialize (precompute and store) groupbys to give online response
y Also: hierarchies on attributes: date -> weekday, date -> month -> quarter -> year
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
136
SQL Extensionsz Front-end tools require
y Extended Family of Aggregate Functionsx rank, median, mode
y Reporting Featuresx running totals, cumulative totals
y Results of multiple group byx total sales by month and total sales by
product
y Data Cube
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
137
Relational OLAP: 3 Tier DSSData Warehouse ROLAP Engine Decision Support Client
Database Layer Application Logic Layer Presentation Layer
Store atomic data in industry standard RDBMS.
Generate SQL execution plans in the ROLAP engine to obtain OLAP functionality.
Obtain multi-dimensional reports from the DSS Client.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
138
MD-OLAP: 2 Tier DSSMDDB Engine MDDB Engine Decision Support Client
Database Layer Application Logic Layer Presentation Layer
Store atomic data in a proprietary data structure (MDDB), pre-calculate as many outcomes as possible, obtain OLAP functionality via proprietary algorithms running against this data.
Obtain multi-dimensional reports from the DSS Client.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
139
16 81 256 10244096
16384
65536
0
10000
20000
30000
40000
50000
60000
70000
2 3 4 5 6 7 8
Data Explosion SyndromeData Explosion Syndrome
Number of DimensionsNumber of Dimensions
Num
ber o
f Agg
rega
tions
Num
ber o
f Agg
rega
tions
(4 levels in each dimension)(4 levels in each dimension)
Typical OLAP Problems Data Explosion
Microsoft TechEd98
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
140
Metadata Repository
z Administrative metadatay source databases and their contentsy gateway descriptionsy warehouse schema, view & derived data definitionsy dimensions, hierarchiesy pre-defined queries and reportsy data mart locations and contentsy data partitionsy data extraction, cleansing, transformation rules,
defaultsy data refresh and purging rulesy user profiles, user groupsy security: user authorization, access control
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
141
Metdata Repository .. 2
z Business datay business terms and definitionsy ownership of datay charging policies
z operational metadatay data lineage: history of migrated data and
sequence of transformations appliedy currency of data: active, archived, purgedy monitoring information: warehouse usage
statistics, error reports, audit trails.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
Recipe for a Successful Warehouse
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
143
For a Successful Warehouse
z From day one establish that warehousing is a joint user/builder project
z Establish that maintaining data quality will be an ONGOING joint user/builder responsibility
z Train the users one step at a timez Consider doing a high level corporate data
model in no more than three weeks
From Larry Greenfield, http://pwp.starnetinc.com/larryg/index.html
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
144
For a Successful Warehouse
z Look closely at the data extracting, cleaning, and loading tools
z Implement a user accessible automated directory to information stored in the warehouse
z Determine a plan to test the integrity of the data in the warehouse
z From the start get warehouse users in the habit of 'testing' complex queries
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
145
For a Successful Warehouse
z Coordinate system roll-out with network administration personnel
z When in a bind, ask others who have done the same thing for advice
z Be on the lookout for small, but strategic, projects
z Market and sell your data warehousing systems
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
146
Data Warehouse Pitfalls
z You are going to spend much time extracting, cleaning, and loading data
z Despite best efforts at project management, data warehousing project scope will increase
z You are going to find problems with systems feeding the data warehouse
z You will find the need to store data not being captured by any existing system
z You will need to validate data not being validated by transaction processing systems
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
147
Data Warehouse Pitfalls
z Some transaction processing systems feeding the warehousing system will not contain detail
z Many warehouse end users will be trained and never or seldom apply their training
z After end users receive query and report tools, requests for IS written reports may increase
z Your warehouse users will develop conflicting business rules
z Large scale data warehousing can become an exercise in data homogenizing
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
148
Data Warehouse Pitfalls
z 'Overhead' can eat up great amounts of disk space
z The time it takes to load the warehouse will expand to the amount of the time in the available window... and then some
z Assigning security cannot be done with a transaction processing system mindset
z You are building a HIGH maintenance systemz You will fail if you concentrate on resource
optimization to the neglect of project, data, and customer management issues and an understanding of what adds value to the customer
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
149
DW and OLAP Research Issuesz Data cleaning
y focus on data inconsistencies, not schema differencesy data mining techniques
z Physical Designy design of summary tables, partitions, indexesy tradeoffs in use of different indexes
z Query processingy selecting appropriate summary tablesy dynamic optimization with feedbacky acid test for query optimization: cost estimation, use of
transformations, search strategiesy partitioning query processing between OLAP server and
backend server.
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
150
DW and OLAP Research Issues .. 2
z Warehouse Managementy detecting runaway queriesy resource managementy incremental refresh techniquesy computing summary tables during loady failure recovery during load and refreshy process management: scheduling queries,
load and refreshy Query processing, cachingy use of workflow technology for process
management
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
Products, References, Useful Links
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
152
Reporting Toolsz Andyne Computing -- GQL z Brio -- BrioQuery z Business Objects -- Business Objects z Cognos -- Impromptu z Information Builders Inc. -- Focus for Windows z Oracle -- Discoverer2000 z Platinum Technology -- SQL*Assist, ProReports z PowerSoft -- InfoMaker z SAS Institute -- SAS/Assist z Software AG -- Esperant z Sterling Software -- VISION:Data
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
153
OLAP and Executive Information Systemsz Andyne Computing -- Pablo z Arbor Software -- Essbase z Cognos -- PowerPlay z Comshare -- Commander
OLAP
z Holistic Systems -- Holos z Information Advantage --
AXSYS, WebOLAP
z Informix -- Metacubez Microstrategies --DSS/Agent
z Microsoft -- Platoz Oracle -- Express z Pilot -- LightShip z Planning Sciences --
Gentium
z Platinum Technology -- ProdeaBeacon, Forest & Trees
z SAS Institute -- SAS/EIS, OLAP++
z Speedware -- Media
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
154
Other Warehouse Related Products
z Data extract, clean, transform, refreshy CA-Ingres replicatory Carleton Passporty Prism Warehouse Managery SAS Accessy Sybase Replication Servery Platinum Inforefiner, Infopump
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
155
Extraction and Transformation Tools
z Carleton Corporation -- Passport
z Evolutionary Technologies Inc. -- Extract
z Informatica -- OpenBridge
z Information Builders Inc. -- EDA Copy Manager
z Platinum Technology -- InfoRefiner
z Prism Solutions -- Prism Warehouse Manager
z Red Brick Systems -- DecisionScape Formation
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
156
Scrubbing Tools
z Apertus -- Enterprise/Integrator z Vality -- IPEz Postal Soft
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
157
Warehouse Products
z Computer Associates -- CA-Ingres z Hewlett-Packard -- Allbase/SQL z Informix -- Informix, Informix XPSz Microsoft -- SQL Server z Oracle -- Oracle7, Oracle Parallel Serverz Red Brick -- Red Brick Warehouse z SAS Institute -- SAS z Software AG -- ADABAS z Sybase -- SQL Server, IQ, MPP
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
158
Warehouse Server Products
z Oracle 8z Informix
y Online Dynamic Servery XPS --Extended Parallel Servery Universal Server for object relational
applicationsz Sybase
y Adaptive Server 11.5y Sybase MPPy Sybase IQ
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
159
Warehouse Server Products
z Red Brick Warehousez Tandem Nonstopz IBM
y DB2 MVSy Universal Servery DB2 400
z Teradata
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
160
Other Warehouse Related Products
z Connectivity to Sourcesy Apertusy Information Builders EDA/SQLy Platimum Infohuby SAS Connecty IBM Data Joinery Oracle Open Connecty Informix Express Gateway
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
161
Other Warehouse Related Products
z Query/Reporting Environmentsy Brio/Queryy Cognos Impromptuy Informix Viewpointy CA Visual Expressy Business Objectsy Platinum Forest and Trees
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
162
4GL's, GUI Builders, and PC Databases
z Information Builders -- Focus z Lotus -- Approach z Microsoft -- Access, Visual Basic z MITI -- SQR/Workbench z PowerSoft -- PowerBuilder z SAS Institute -- SAS/AF
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
163
Data Mining Products
z DataMind -- neurOagent z Information Discovery -- IDIS z SAS Institute -- SAS/Neuronets
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
164
Data Warehouse
z W.H. Inmon, Building the Data Warehouse, Second Edition, John Wiley and Sons, 1996
z W.H. Inmon, J. D. Welch, Katherine L. Glassey, Managing the Data Warehouse, John Wiley and Sons, 1997
z Barry Devlin, Data Warehouse from Architecture to Implementation, Addison Wesley Longman, Inc 1997
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
165
Data Warehouse
z W.H. Inmon, John A. Zachman, Jonathan G. Geiger, Data Stores Data Warehousing and the Zachman Framework, McGraw Hill Series on Data Warehousing and Data Management, 1997
z Ralph Kimball, The Data Warehouse Toolkit, John Wiley and Sons, 1996
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
166
OLAP and DSS
z Erik Thomsen, OLAP Solutions, John Wiley and Sons 1997
z Microsoft TechEd Transparencies from Microsoft TechEd 98
z Essbase Product Literaturez Oracle Express Product Literaturez Microsoft Plato Web Sitez Microstrategy Web Site
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
167
Data Mining
z Michael J.A. Berry and Gordon Linoff, Data Mining Techniques, John Wiley and Sons 1997
z Peter Adriaans and Dolf Zantinge, Data Mining, Addison Wesley Longman Ltd. 1996
z KDD Conferences
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
168
Other Tutorials
z Donovan Schneider, Data Warehousing Tutorial, Tutorial at International Conference for Management of Data (SIGMOD 1996) and International Conference on Very Large Data Bases 97
z Umeshwar Dayal and Surajit Chaudhuri, Data Warehousing Tutorial at International Conference on Very Large Data Bases 1996
z Anand Deshpande and S. Seshadri, Tutorial on Datawarehousing and Data Mining, CSI-97
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
169
Useful URLs
z Ralph Kimballs home pagey http://www.rkimball.com
z Larry Greenfields Data Warehouse Information Centery http://pwp.starnetinc.com/larryg/
z Data Warehousing Institutey http://www.dw-institute.com/
z OLAP Councily http://www.olapcouncil.com/
www.jntuworld.com
www.jntuworld.com
www.jwjobs.net
