Be Right back in 5 minutes!

Thank you for joining

IBM PureData System

for Analytics
(Formerly known as, IBM Netezza

- Ravi

What is Netezza?
Datawarehouse Appliance
H/W & S/W pre-bundled, pre-configured

Little configuration needed after deployment

Solves the traditional datawarehouse complexities!

Traditional DW Complexities!

Netezza Appliance: Datawarehousing simplified!

Netezza Architecture (major) Principles

Processing close to the data source
Balanced massively parallel architecture
Appliance Simplicity
Flexible configurations and extreme scalability

FPGA in Action!

SELECT DISTRICT, PRODUCTGRP, SUM (NRX)

FROM MTHLY_PROD_DATA
WHERE PDATE=20140401
AND MARKET = 2014
And SPECIALITY = GASTRO

SELECT
DISTRICT,
PRODUCTGRP,
NRX

SUM (NRX)

Slice of table
MTHLY_PROD_DATA
(Compressed)

S-Blade View
S-Blade: Where extreme performance happens

AMPP Architecture

What happens when you submit a query?

What happens when you submit a query?

Host compiles the query & divides into snippets
Optimizer creates a query execution plan by making intelligent decisions like join order/
redistribution/broadcast
Each snippet has two elements: Compiled code & FPGA parameters
Object Cache: Improves query performance. You can avoid code compilation
Scheduler: Maintains maximum utilization and throughput

S-Blades execute these snippets in parallel. Sends the results back to host
Host accumulates the results and results will be returned to Client

Netezza TwinFin Specifications

Netezza TwinFin Specifications

Various Datawarehouse appliances in the market!

IBM (Netezza)
HP (Vertica)
EMC (Greenplum)
SAP (HANA-High Performance Analytics Appliance)

Oracle (Exadata)
Teradata (Teradata, Asterdata)
Microsoft (DATAllegro)

http://ybigdata.blogspot.com/2013/01/vertica-vs-aster-data-vs-greenplum-vs.html

Netezza Delivers
Speed: 10-100x faster than traditional custom systems

Simplicity & Ease: Minimal tuning & administration and greater resilience
Fast time to value: 5 TB/Hour load speed
Smart: Complex algorithms in minutes. A rich library of integrated analytics

Data-In/Out of Netezza Appliance

Course Contents (Development)

About Netezza Performance Server (NPS)

NPS AMPP Architecture & Various Netezza appliance models
Installing Netezza GUI client Interfaces
ODBC/JDBC/OLEDB Client connectivity
Installing Netezza Emulator for day-to-day practice
Netezza Command Line Interface (CLI)
Netezza SQL (NZSQL) language
NZSQL DDL/DML/DCL/TCL statements
NZSQL command line options & Internal slash commands
Netezza Environment & System Variables
Managing User Access to the Netezza Database
Working with Databases & Tables (both from NZSQL & GUI interfaces)
Netezza Data Types, Schemas, Users, Groups, Privileges
Functions, Operators, Constraints, Sequences, & Synonyms
Data Distribution (Hash, Random) & Best practices
Transactions, Generate Statistics, Zone Maps, Materialized Views
Groom Table, CBT, SQL Identifiers
Data Loading/Unloading using External Tables, NZLOAD (CLI & GUI tools)
Netezza system tables, views, user tables, Materialized views
Netezza Joins, Sub queries, stored procedures, Aggregates, UDFs
Netezza SQL Extensions & analytic functions
Techniques to improve Netezza query performance
Working with IBM Netezza support to resolve issues

Course Contents (DBA)

About Netezza Performance Server (NPS)

NPS AMPP Architecture & Various Netezza appliance models
Netezza High Availability Architecture (Clustering, Mirroring, failover)
Installing Netezza system and client softwares
Installing Netezza Emulator for day-to-day practice
NzAdmin: GUI Admin Tool (Installation & Setup)
Netezza Command Line Interface (CLI)
Manage NPS with CLI commands
Manage User access to Netezza Databases
Monitoring Netezza and Linux logs
Netezza Events (Setup & Monitoring)
Databases & Tables
Data Distribution (Hash, Random), Cluster Base Tables, Table Skew
Generate Statistics, Zone Maps, Materialized Views, Groom Table
Backup & Restore (Host Level, Database Level, Table Level)
Netezza Appliance/Database migration (For Example: 6.x to 7.x migration)
Data Loading/Unloading using External Tables, NZLOAD, NZ_MIGRATE
Data Loading/Unloading using GUI Tools
Optimizer and query plans
Query history collection & Reporting
Workload management
Netezza Replication/DR Architecture
Techniques to improve Netezza performance
Frequent DBA activities such as SPU replacements, Table Skew monitoring, etc
Working with IBM Netezza Support to resolve issues

IBM Netezza Features

http://www-01.ibm.com/software/data/puredata/analytics/features.html