Post on 23-Apr-2018
© 2012 IBM Corporation
1
IBM NetezzaTaking advantage of the new wealth of information t o make more intelligent decisions at the time of impa ct
Carlo Marchesi – TechSales / System Engineer IBM Net ezza
© 2012 IBM Corporation
Information Management
2
Würden Sie Google nutzenWenn es Sie 3 Tage und 7 Personen kostet,um ein Suchergebnis zu bekommen?
© 2012 IBM Corporation
Information Management
Nahezu 70% aller Data Warehouse Anwendungen leiden unter Leistungseinschränkungen der unterschiedlichsten Art.
3
- Gartner 2010 Magic Quadrant
Monate bis zur ersten Nutzung
Nur für Experten durchschaubar
andauerndes TuningTage für eine einzige Abfrage
© 2012 IBM Corporation
Information Management
� Zu komplexe Infrastruktur
� Zu komplizierter Einsatz
� Zu viel Tuning notwendig
� Zu ineffiziente Analysen
� Zu viel Personal für die Wartung
� Zu kostspielig im Betrieb
Traditionelle Data Warehouse Anwendungen
Sie basieren auf Datenbanken die für die Transaktionsverarbeitung optimiert wurden – NICHT um die Anforderungen von fortschrittlichen Analysen auf großen Datenbeständen abzubilden
4
sind einfach zu komplex
Zu zeitraubend für schnelle Antworten
© 2012 IBM Corporation
Information Management
Agenda
� What is Netezza– Customer Cases
� Architecture– Hardware– Scalability
� High Availability
� Integration– ETL– BI Analytics– Backup & Restore
� Migration
� IBM Warehousing Offerings
5
© 2012 IBM Corporation
Information Management
WHAT IS NETEZZAIBM Netezza
6
© 2012 IBM Corporation
Information Management
7
Netezza’s Market-Leading Evolution
Source: If applicable, describe source origin
World’s First
Data Warehouse
Appliance
World’s First
100 TB Data
Warehouse Appliance
World’s First
Petabyte Data
Warehouse Appliance
Impa
ct
World’s First
Analytic Data
Warehouse Appliance
NPS®
8000 Series
TwinFin™with i-Class™
Advanced Analytics(300X Performance)
NPS®
10000 Series (50X Performance)
TwinFin™
(150X Performance)
2003 2006 2009 2010
© 2012 IBM Corporation
Information Management
8
� Appliance simplicity
� Integrated database, server & storage
� 10-100x faster than traditional systems
� Peta-scale data capacity
� Advanced analytics
� Standard interfaces
� Trust with compliance & auditing
© 2012 IBM Corporation
Information Management
99
“…when something took 24 hours I could onlydo so much with it, but when something takes10 seconds, I may be able to completelyrethink the business process…”
- SVP Application Development, Nielsen
15,000 users running 800,000+ queries per day (50X faster than before)
Eine echte Appliance ermöglicht - Speed
Source: http://www.youtube.com/watch?v=yOwnX14nLrE&feature=player_embedded
© 2012 IBM Corporation
Information Management
DAYS
WEEKS
MONTHS
Eine echte Appliance ermöglicht - Simplicity
10
“Allowing the business users access to the Netezza box was what sold it.”
Steve Taff,
Executive Dir. of IT Services
Up an running 6 months before having any training
© 2012 IBM Corporation
Information Management
1111
“NYSE … has replaced an Oracle IO relational database
with a data warehousing appliance from Netezza,
allowing it to conduct rapid searches of 650 terabytes
of data.”
ComputerWeekly.com
Source: http://www.computerweekly.com/Articles/2008/04/14/230265/NYSE-improves-data-management-with-datawarehousing.htm
Eine echte Appliance ermöglicht - Scalability
1 PB on Netezza (including DR)
7 years of historical data
100-200% annual data growth
© 2012 IBM Corporation
Information Management
Eine echte Appliance ermöglicht - Smarts
12
“Using results derived from Netezza system, Catalina Marketing is able to give shoppers (more relevant) coupons at point of sale for items they would like want to buy in future visits.”
Editorial Director,
DM Review
30% redemption increase
© 2012 IBM Corporation
Information Management
ARCHITECTUREIBM Netezza
13
© 2012 IBM Corporation
Information Management
The Netezza TwinFin™ Appliance
14
High-performance database
engine streaming joins,
aggregations, sorts, etc.
SQL Compiler
Query Plan
Optimize
Admin
Processor &
streaming DB logic
Slice of User Data
Swap and Mirror partitions
High speed data streaming
SMP Hosts
S-Blades™
(with FPGA-based
Database Accelerator)
Disk Enclosures
© 2012 IBM Corporation
Information Management
S-Blade™ Components
Intel Quad-Core
Dual-Core FPGADRAM
IBM BladeCenter Server Netezza DB Accelerator
SAS ExpanderModule
SAS ExpanderModule
© 2012 IBM Corporation
Information Management
The Netezza AMPP™ Architecture
Advanced AnalyticsAdvanced Analytics
LoaderLoader
ETLETL
BIBI
Applications
FPGA
Memory
CPU
FPGA
Memory
CPU
FPGA
Memory
CPU
HostsHost
Disk Enclosures S-Blades™
NetworkFabric
Netezza Appliance
© 2012 IBM Corporation
Information Management
Asymmetric Massively Parallel Processing™
Massively Parallel Intelligent Storage
1
2
3
920
�
�
�
NetzwerkSMP Host
Front End
Netezza TwinFin Appliance
High-Speed Loader/Unloader
ODBC 3.XJDBC Type 4
OLE-DBSQL/92
Execution Engine
SQL Compiler
Query Plan
Optimize
Admin
Quell-Systeme
Client
High Performance
Loader
3rd PartyApps
DBA CLI
ETL Server
SOLARIS
LINUX
HP-UX
AIX
WINDOWS
TRU64
High-PerformanceDatabase EngineStreaming joins,
aggregations, sorts
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
© 2012 IBM Corporation
Information Management
High-PerformanceDatabase EngineStreaming joins,
aggregations, sorts
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
Execution Engine
Asymmetric Massively Parallel Processing™
Massively Parallel Intelligent Storage
1
2
3
920
�
�
�
NetzwerkSMP Host
Front End
Netezza TwinFin Appliance
High-Speed Loader/Unloader
SQL Compiler
Query Plan
Optimize
Admin
SQL
1 2 3
1 2 3
1 2 3
1 2 3
Snippets
1 2 31 2 3
SQL
Quell-Systeme
Client
High Performance
Loader
3rd PartyApps
DBA CLI
ETL Server
SOLARIS
LINUX
HP-UX
AIX
WINDOWS
TRU64
© 2012 IBM Corporation
Information Management
Our Secret Sauce
FPGA Core CPU Core
Uncompress Project Restrict,Visibility
Complex ∑Joins, Aggs, etc.
select DISTRICT,PRODUCTGRP,sum(NRX)
from MTHLY_RX_TERR_DATAwhere MONTH = '20091201'and MARKET = 509123and SPECIALTY = 'GASTRO'
Slice of tableMTHLY_RX_TERR_DATA
(compressed)
where MONTH = '20091201'and MARKET = 509123and SPECIALTY = 'GASTRO'
sum(NRX)
select DISTRICT,PRODUCTGRP,sum(NRX)
© 2012 IBM Corporation
Information Management
High-PerformanceDatabase EngineStreaming joins,
aggregations, sorts, etc.
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
S-Blade
Processor &
streaming DB logic
Asymmetric Massively Parallel Processing™
Massively Parallel Intelligent Storage
1
2
3
920
�
�
�
NetzwerkSMP Host
Front End
Netezza TwinFin Appliance
High-Speed Loader/Unloader
SQL Compiler
Query Plan
Optimize
Admin
1 2 3
1 2 3
1 2 3
1 2 3
Consolidate
Execution Engine
ODBC 3.XJDBC Type 4
OLE-DBSQL/92
Quell-Systeme
Client
High Performance
Loader
3rd PartyApps
DBA CLI
ETL Server
SOLARIS
LINUX
HP-UX
AIX
WINDOWS
TRU64
© 2012 IBM Corporation
Information Management
TwinFin™ 24 Specification
• 16 (8*2) Disk Enclosures• 192 (96*2) 1TB SAS Drives • (8 hot spares)• RAID 1 Mirroring
• 24 Netezza S-Blades:• 192 Core’s ( Intel Quad-Core 2.5 GHz)• 192 FPGA’s ( 125 MHz )• 384 GB DDR2 RAM (1+TB compressed)• Linux 64-bit Kernel
• 2 Hosts (Active-Passive):• 24 Cores (Quad-Core Intel 2.6 GHz)• 96 GB Memory• 4x146 GB SAS Drives• Red Hat Linux 5 64-bit• 10G Internal Network
• User Data Capacity: 250 TB• Data Scan Speed: 290 TB/hr• Load Speed (per system): 2.0 TB/hr
• Power/Rack: 7,400 Watts• Cooling/Rack: 25,500 BTU/Hour
© 2012 IBM Corporation
Information Management
HIGH AVAILABILITY
22
© 2012 IBM Corporation
Information Management
Disk Mirroring and Failover
� All user data and temp space mirrored
� Disk failures transparent to queries and transactions
� Failed drives automatically regenerated
� Bad sectors automatically rewritten or relocated
Primary
Mirror
Temp
© 2012 IBM Corporation
Information Management
S-Blade™ Failover
� Drives automatically reassigned to active S-Blades within a chassis
� Read-only queries that have not returned data yet automatically restarted.
� Transactions and loads interrupted
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
S-Blades
© 2012 IBM Corporation
Information Management
INTEGRATION
25
© 2012 IBM Corporation
Information Management
Loading and Quering Netezza
26
SQ
L
OD
BC
J
DB
C
OLE
-DB
Datenintegration
� Ab Initio� Business Objects/SAP� Composite Software� Expressor Software� Oracle GoldenGate� Informatica� IBM Information Server
(Datastage)� Oracle Data Integrator
(Sunopsis)� WisdomForce� Talend Open Studio� …
Reporting & Analyse
� Actuate� Business Objects/SAP� Cognos (IBM)� Information Builders� Kalido� KXEN� MicroStrategy� Oracle OBIEE� QlikTech� Quest Software� SAS� SPSS (IBM)� Unica (IBM)� …
laden lesen
SQ
L
OD
BC
J
DB
C
OLE
-DB
© 2012 IBM Corporation
Information Management
IBM Netezza Analytics for…
Client
Developer
Custom Analytics
R, Hadoop, Java, C, C++, Python, Fortran
Analyst
Model Building & Scoring
IBM SPSS, Revolution Analytics, Fuzzy Logix, SAS,
R
Business Manager
BI & Visualization
IBM Cognos, SAS, Microstrategy, Business
Objects, ESRI, MS Excel …
IBM Netezza appliance
Predictive Analytics
Geospatial Analytics
Advanced Statistics
Data MiningData Preparation
© 2012 IBM Corporation
Information Management
Advanced Analytics – the Traditional Way
28
Fraud Detection
Demand Forecasting
SAS
R, S+
Analytics
Grid
Data
Warehouse
C/C++, Java, Python,
Fortran, …
Data
SQL
SQL
ETL
SQL
ETL
ETL
© 2012 IBM Corporation
Information Management
Advanced Analytics with IBM Netezza
29
Fraud Detection
Demand Forecasting
Analytic
ToolsAnalytics
Grid
Data
Warehouse
Data
SQL
SQL
ETL
SQL
ETL
ETL
C/C++, Java, Python,
Fortran, …
© 2012 IBM Corporation
Information Management
Advanced Analytics with IBM Netezza
30
Fraud Detection
Demand Forecasting
Analytic
Tools
© 2012 IBM Corporation
Information Management
IBM Netezza AnalyticsTechnical Overview
31
Software Development Kit
Adaptors for R, Hadoop
Java, C, C++, Python, Fortran
Adaptors for R, Hadoop
Java, C, C++, Python, Fortran
Eclipse Client
Plug-in
Eclipse Client
Plug-in
User-Defined
Extensions
User-Defined
Extensions
IBM SPSS
Modeler
IBM SPSS
Modeler
Analytic Packages
Revolution
Analytics
Revolution
AnalyticsSASSAS
IBM Netezza AMPP™ Platform
Client Connectors / Partner Engines
(e.g., Fuzzy Logix)
IBM Netezza Analytics
RR SpatialSpatialHadoopHadoop MatrixMatrix
Language Support & AdaptorsLanguage Support & Adaptors
Data Prep
Predictive Analytics
Data Mining
In-Database Analytics
Client
IBM Netezza
data warehouse appliance
© 2012 IBM Corporation
Information Management
IBM In-Database Analytics
32
Analytic Task Function Types
Data Exploration and Discovery
Data ProfilingDescriptive StatisticsPrincipal Components AnalysisBayesian Network
Data Transformation StandardizationNormalizationBinningMissing Data Treatment
Model Building Classification: Decision Trees, Naïve BayesEstimation: Linear Regression, Regression TreesClustering: k-Means, HierarchicalAssociation: a Priori, FP-Growth
Model Diagnostics Mean/Relative Error StatisticsHypothesis TestingClassification/Misclassification Statistics
Model Scoring Linked to Model Building functions
© 2012 IBM Corporation
Information Management
MONITORING
33
© 2012 IBM Corporation
Information Management
nzPortal: Monitoring and Capacity Planning
� Visual representation of Query History database
� Search for and compare runtimes of identical queries
© 2012 IBM Corporation
Information Management
BACKUP & RESTORE
35
© 2012 IBM Corporation
Information Management
Backup and Restore Features
� Enterprise-class integration and certification with 3rd-party BAR tools – Simplify deployment with leading backup and restore tools– X/Open Backup Services API (XBSA) support – Certification with Veritas NetBackup™ from Symantec– Certification with IBM Tivoli Storage Manager
� Incremental backup and restore– Significantly shorten backup times compared to full backups– Available in NZBACKUP utility– Full or partial restore options
Sun Mon Tue Wed Thu Fri Sat
Ful
l Diff Diff
Cum
Diff Diff Diff
© 2012 IBM Corporation
Information Management
MIGRATION
37
© 2012 IBM Corporation
Information Management
Oracle to Netezza Conversion Tool
� The mission– Tool to automatically convert Oracle SQL (and PL/SQL) to Netezza SQL (and NZPLSQL)– To minimize manual effort during code conversion from Oracle to Netezza.
Assessment Estimation Conversion Testing
• Size and complexity of Data
• Complexity of Database Objects / Data types
• Complexity of PL/SQL codebase (if any)
• Assessment data fed into an Estimation tool
• Development of a tool like MEET in plan
• Tool to minimize manual conversion of Oracle Code
• The following Oracle entities can be converted to Netezza equivalents….
• Correctness of Syntax / Semantics
•Data type compatibility
• Are Netezza best practices followed ?
Data Definition statements
Data Manipulation statements
Oracle PL/SQL & Stored Procedures
Oracle Data Types and Formats
© 2012 IBM Corporation
Information Management
Conversion Capability - Today� Oracle Datatypes
– Character Datatypes (CHAR,VARCHAR2,LONG,NVARCHAR2,NCHAR, RAW,LONG RAW)– Numeric Datatypes (NUMBER,NUMERIC,FLOAT,DEC,INT,SMALLINT,REAL,DOUBLE,)– Temporal Datatypes (DATE,TIMESTAMP, TIMESTAMP with timezones, INTERVAL DAY/YEAR)– Large Objects (BFILE,BLOB,CLOB,NCLOB)
� Basic Oracle DDLs– Create Statements (TABLES,VIEWS, SEQUENCES,MATERIALIZED VIEWS,USERS,GROUPS)– Alter Statements– Rename Statements– Truncate
� Basic Oracle DMLs– INSERTs– UPDATEs– DELETEs– SELECTs (Including Oracle’s OUTER JOIN (+) syntax)– Correlated Subqueries in UPDATE statements
� PL/SQL & Stored Procedures– Stored Proc declarations with and without parameters– Nested Stored Procedures (Will be Unnested for Netezza)– PL Constructs (loops)– Oracle Exceptions– Output statements (DBMS_OUTPUT)
� Anything ANSI does not require conversion!
© 2012 IBM Corporation
Information Management
What cannot be converted (non-exhaustive)
� Oracle Datatypes– RowID Datatypes (ROWID,UROWID) X
� Oracle DDLs– INDEX related statements X– TABLESPACE related statements X– All other Oracle db objects unsupported in Netezza X
� Basic Oracle DMLs– CURSOR Operations ( Oracle + format) X
� PL/SQL & Stored Procedures– RECORD Variable declarations and operations X– Oracle EXCEPTIONS X– Output statements (DBMS_OUTPUT) X
– Key: X - Not Required by Netezza X – Manual conversion will be advised
© 2012 IBM Corporation
Information Management
National Stock Exchange : Netezza PoC In Progress
� The PoC is underway with the following Generic succe ss criteria
Task Current Time NSE ’’’’s Expectation Results on Netezza
Data Load Rate 40 GB per Hour 400 GB per Hour (10x) 1.1 TB per Hour
Data Compression 65% Compression > 65% > 75%
ETL processing 70 mins 7 mins (10x) 2m30Secs
BO queries 70 mins 3.5 mins (20x) 1m30Secs
Intraday Processing 6Hrs 30Mins 39 mins (10x) In Progress
EOD Processing 6Hrs 30Mins 39 mins (10x) In Progress
� NSE’’’’s current environment is Oracle on a 16 Core HP ser ver
� Netezza configuration being used for the PoC – TwinFin 12
� Competitiors – Exadata, Greenplum
© 2012 IBM Corporation
Information Management
IBM DATA WAREHOUSE OFFERINGS
42
© 2012 IBM Corporation
Information Management
Appliance family for data life-cycle management
43
Netezza-100 Smart Analytics System 5710
Netezza-1000 Netezza High Capacity
Dev & Test System Complete BI Solution Data WarehouseHigh Performance Analytics
Queryable ArchivingBack-up / DR
1 TB to 10 TB 1.5 TB to 13 TB 1 TB to 1.5 PB 100 TB to 10 PB
© 2012 IBM Corporation
Information Management
IBM DB2 Analytics Accelerator
10Gb
OSA-Express3
10 GbE
Primary
Backup
zEnterprise IBM DB2 Analytics Acelerator
BladeCenter
Private Service Network
Admin ClientData Studio Foundation
DB2 Analytics Accelerator
Admin Plug-in
AnalyticQueries
OnlineTransactions
(Data Warehouse)enabled for IBM DB2 Analytics Accelerator
z/OS: Recognized leader in Security, Availability, &
Recoverability for OLTP
Netezza: Recognized leader in cost-effective high speed deep
analytics
DB2 for z/OS
© 2012 IBM Corporation
Information Management
45
Price/Performance
Leader
Price/Performance
Leader
Price/Performance
Leader
Data mart to
enterprise
warehouses
Data mart to
enterprise
warehouses
Data mart to
enterprise
warehouses
Low TCO,
fast time to value
Low TCO,
fast time to value
Low TCO,
fast time to value
Deliver analytics
to the masses
Deliver analytics
to the masses
Deliver analytics
to the masses
Speed
• Hardware-based data streaming
Scalability
• True MPP offers enterprise scale-out
Simple
• Black-box appliance with no tuning or storage administration
Smart
• Built-in advanced analytics pushed deep into database
Netezza Differentiators & Business Impact
© 2012 IBM Corporation
Information Management
USE CASES
46
© 2012 IBM Corporation
Information Management
Netezza Deal Scenarios
1 EDW Replacement
2
High Value, Focused Data Marts3
Retail Analytics “Out of the Box”
Install Next-gen EDW
4
5 SAP BW Co-exist / Replacement
© 2012 IBM Corporation
Information Management
Netezza Deal Scenarios
1. Teradata, EDW ReplacementSolution • Netezza Twin Fin
Target Accounts
• Tired of paying TD/ORCL maintenance and cost for expansion.• Not using a lot of TD/ORCL proprietary applications on the platform.• Trying to change business (add new analytic capabilities).• Sponsor exists in account who will be a change agent.
Target Audience
• Typically, CIO will sponsor this or SVP of BI / DW at a larger account.
Value Prop/ Differentiated Positioning
• Establish EDW that’s faster, simpler, cheaper (esp TCO) than Teradata, Oracle, et al.
• Unique architectural approach (esp DB DW) enables CIO to achieve “operating leverage” over life of technology.
• Better serve needs of business immediately and over long run –architectural features enable biz growth and change cheaply, easily, quickly (vs. ORCL, etc.).
© 2012 IBM Corporation
Information Management
Yum! Brands - Teradata EDW Replacement
A $11B global quick serve leader with 37,000 restaurants in 110 countries. Brands include KFC, Pizza Hut, Taco Bell, Long John Silver s, and A&W.
� Growing Business w/ expanding demand for BI and analytics
� BI user satisfaction was declining
� Cost/Performance
� Appliances were maturing
� Good Time for a Change
49
� Goal: Better, Faster, Cheaper BI
� Lower TCO on legacy playform, by replacing Teradata
� Separate platforms for high performance analytics and BI from operational reporting.
� 3x POC w/ Teradata, DataAllegro and Netezza
� Netezza was 2-4x faster than Teradata
� Netezza deployed in 3 mo for forst brand, <1 yr for all brands
� BI Platform deployed to 1,400 users
� Queries run 2-11x faster than previously with Teradata
� 60% reduction in TCO over Teradata
� Shift in focus from admin task to BI and analytics
Challenge Solution Result
© 2012 IBM Corporation
Information Management
Oracle Exadata Results In IBM Netezza Client Advantage
Costs
– Acquisition cost can exceed $7M per rack• Hardware $1M• Software is more than $6M!
– High maintenance and software subscription
– Continuing high admin costs
High cost of ownership
– Low, transparent initial cost– Simple install requires no
additional professional services– Standard maintenance includes
hardware /software support & software upgrades
– Netezza Migrator for easy migration of Oracle applications
– Fast deployment & time-to-value
– Easily understood, predictable costs
– Minimal “extra” services so easier to budget for Netezza
Smart– Very limited push-down analytics– RAC bottleneck for
analytic performance
Poor analytic performance
– Push down of many diverse analytics (SAS, R, Gnu, etc.) with Netezza i-Class
– Fast time to insight– Advanced analytics on big
data easily accessible
Simplicity
– Complexity of Oracle Real Application Clusters (RAC)
– Constant tuning for performance– Complex maintenance and patch
process
Complex administration
– True Appliance with HW/SW created to provide high performance for data warehousing
– No tuning
– More time spent delivering business value rather than tuning for acceptable performance
Scalability
– No proof points of scaling– Backup/restore bottlenecks– Parallelism for joins limited to
RAC tier
Scalability Bottlenecks
– Proven scalability at peta-scale– No scalability bottlenecks
– No restrictions on business and data growth
Speed
– Tuned for OLTP (e.g. FlashCache)– RAC unfit for data warehouse
workloads
Poor data warehouse
performance
– Appliance tuned for data warehousing and advanced analytics
– Highest data warehouse performance
– Operational Simplicity
Architecture
– Two layer:• Clustered SMP DB Layer (RAC)• Shared disk MPP Storage Layer
Compromised performance
– True MPP with FPGA acceleration of processing in each MPP node
– Best architecture for data warehouse and advanced analytics due to minimization of contention/bottlenecks
Details: IBM Netezza is better value than Oracle Exadata
© 2012 IBM Corporation
Information Management
Netezza Deal Scenarios
2. Install Next-gen EDWSolution • Netezza Twin Fin
Target Accounts
• Less mature retailer on outdated and/or disorganized technologies.• Looking to rationalize data platforms and achieve better performance
as efficiently as possible.• Sponsor exists in account who will be a change agent.
Target Audience
• Typically, CIO will sponsor this.
Value Prop/ Differentiated Positioning
• Establish EDW that’s faster, simpler, cheaper (esp TCO) than alternative vendors Teradata, Oracle, et al.
• Unique architectural approach (esp DB DW) enables CIO to achieve “operating leverage” over life of technology.
• Better serve needs of business immediately and over long run –architectural features enable biz growth and change cheaply, easily, quickly (vs. ORCL, etc.).
© 2012 IBM Corporation
Information Management
Carter’s – Next-gen EDWLeading children’s apparel retailer, with 6 brands, more than 375 Carter’s and OshKosh retail stores and distribution through national department stores, chain and specialty stores, and discount retailers. In 2009 Carter’s revenue was $1.68B
� No single view of the biz– 2005 acquired OshKosh– Multiple brands, product
hierarchies, distribution channels, planning and allocation systems
� Could not support detailed analysis: SKU Store Trasaction
� Missing SLAs
� Poor Performance: 12 hrs for an analytic report
� Oracle system admin was absorbing too many resources
52
• Netezza: Most straightforward, minimal administration, lowest TCO
• Exadata: Essentially a bigger, faster version of existing platform; same administration requirements and Oracle limitations; platform partner concerns HP, SUN?
• Teradata: In transition phase from big DW player to appliance vendor
• Consistently meeting SLAs
• Faster, better analytics: fresh sales trends data, market basket analysis (NEW), imminent stock-out (NEW)
• 1 platform, 1 data model, 1 product hierarchy for all brands and channels
• Reduced admin. – reports cut from 120 to 15
• 12 hour analytical data reformat reduced to 90 minutes on Netezza
Challenge Solution Result
© 2012 IBM Corporation
Information Management
Netezza Deal Scenarios
3. High Value, Focused Data Marts Solution • Netezza Twin Fin
TargetAccounts
• Has new strategic initiative involving big data, big analytics (a la customer hub, marketing analytics).
• Struggling to make merchandising data warehouse functional (e.g., Oracle Retail RDW port).
• Typically larger, Tier 1 retailers (e.g., Best Buy, Target, Tesco).• Project can be implemented alongside existing EDW to enhance/extend.
• Looking for speed-to-value for business with minimal disruption to existing IT foundation.
Target Audience
• Position with Chief Marketing Officer, Chief Merchant, and/or CIO.
Value Prop/ Differentiated Positioning
• Establish high value data hub for business in weeks.• Enable apps and/or analysis quickly to support key business initiatives (e.g., multi-channel customer experience).
• Realize value of investment in merchandising OLTP (Oracle Retail RDW).
• Can easily, cheaply evolve to play bigger role in EDW if desired.
© 2012 IBM Corporation
Information Management
Netezza Deal Scenarios
4. Retail Analytics ““““out of the box ””””Solution • Netezza Retail Analytic Appliance
Target Accounts
• Under invested in technology and looking to scale/ evolve technology to support maturing business with minimal risk and business disruption.
• Has no DW/BI or is not happy with existing functionality.• Haven’t capitalized on BI investment (MS, Cognos) and/or data in
merchandising OLTP apps.• Has too many reports, and/or has widespread use of Excel offline.• Avoid established Teradata accounts, where unlikely to replace just for
this solution.
Target Audience
• Position with Chief Merchant / GMM, VP of Planning, Head of Stores, CFO and/or CIO.
Value Prop/ Differentiated Positioning
• Only complete, proven solution – HW, BI & content (data model, ETL, metrics, reports, dashboards, playbooks) out-of-the-box deployed at leading retailers.
• Fast, Efficient, Low Risk - Live in 4-8 weeks with less risk and cost than build or alternative offerings.
Page 54
© 2012 IBM Corporation
Information Management
Summary Key Messages
55
• Netezza is far less expensive than comparable competitive systems • Netezza is the ONLY technology that provides “operating leverage” for IT / CFO.
Financial
• Netezza is designed as a data warehouse appliance.• Our performance is driven by an elegant design, not by brute force.• Designed for simplicity drives speed to value for application development.
Technical
• Simplicity means that you can better service the business by taking off the ‘guard rails.’• Let the business analyze at the ‘speed of thought’ in the context of a business process.
Business people don’t do well when they are constrained by IT. They are creative by nature.
Business
• Everybody talks about having an ‘appliance,’ but the only thing they have in common with Netezza is that they are in a single cabinet.
• Customers buy because of price / performance, while they stick with us because of simplicity.
Competitive
© 2012 IBM Corporation
Information Management
Netezza und SAP
� Einfache Integration von SAP und Nicht-SAP Daten
� Schnittstellen zur bestehenden BI Lösung Cognos
� Ergänzung oder kompletter Ersatz von SAP BW– Mit SAP BW: Basierend auf SAP OpenHub Interface– Ohne SAP BW: Integration basierend auf SAP ECC
� Einfache Skalierbarkeit ohne zusätzliche Komplexität
© 2012 IBM Corporation
Information Management
Netezza und SAP mit NewFrontiers
� NewFrontiers liefert vordefiniert:– ETL-Prozess– Datenmodell– InfoProbes (Reports)
� Rasche und einfache Integration– POC in 10 Tagen
� Referenzen:
– Heineken
– Unilever
– Generali
– Roche
– ING
– ...
DataMart(s) ODS Staging
InfoProbes • reporting, analysis • alerts, dashboards • analytics
Non SAP
ETL/ELT process
© 2012 IBM Corporation
Information Management
58
Test Drive
TwinFin Your Data. Your Site. Our Appliance.
Information Management