What Extreme Performance Means for Your Business
What Extreme Performance Means for Your Business
2
Data Warehousing at an Inflection Point
“2011 will be the year when data warehousing reaches what could well be its most-significant inflection point since its
inception.”
Source: The State of Data Warehousing in 2011
3
Q: Why can‟t I get
access to the
data I need for
decision making ?
Trusted
Information
•Accurate
•Consistent
•Quality
Q: Why do I have
so many duplicate
copies of data?
Much of it is
inaccurate!
Q: Why are my
applications
referring to last
week‟s numbers?
Accessible
Information
•Available
•Secure
•Reliable
Up-to-Date
Information
•Fast access
•Multiple sources
•Actionable
Information Management Challenges
4
ETL
OLAP Data Mining
OLAP Data Mining
ETL
Data Marts
Data Marts
Today’s Reality: No Single Source of Truth Expensive Data Warehouse Architecture
5
Getting to Your Data Becomes a Circus Act
6 6
Difficult to Manage
Limited Scalability
Inaccessible Information
Untrustworthy Data
Fragmented data Silos
Fragmented data Silos • Higher Costs
• Increased Risk
• Poor Decision Making
• Poor Decisions
• Higher Costs
• Increased Risk
What’s the Impact on Your Business?
7
Oracle Database 11g
Oracle Exadata Database Machine
Data
Marts
Data Mining
Online
Analytics ETL
Consolidate onto a Single Platform Faster Performance, Single Source of Truth
8
• Data Warehousing
• OLTP
• Database Consolidation
Exadata is Oracle’s strategic database platform for ALL Oracle Database
workloads
Oracle Exadata Database Machine One Architecture for …
9
Exadata Changes the Equation A Major Leap in Performance, Capacity and Value
Perform database queries 10x faster
Consume 10x less storage
Consolidate platforms, databases, power, cooling, administration
Eliminate systems integration trial-and-error
Run existing applications unchanged
Apply existing personnel, skills, Oracle licenses
“Exadata is the fastest growing product in Oracle’s history”
Larry Ellison, Oracle CEO
10
Are You Ready?
11
Exadata Hardware Architecture Pre-integrated, Optimized Oracle Database Platform
• Database Grid
– Intel-based database servers
– Oracle Linux or Solaris 11
– 10 Gig Ethernet (to data center)
• Storage Grid
– Intel-based storage servers
– Up to 336 terabytes raw disk
– 5.3 terabytes Flash storage
• InfiniBand Network
– Internal high-bandwidth connectivity
A complete system – compute, storage and network; pre-configured,
balanced and optimized.
12
The Exadata Difference Dedicated to running the Oracle Database
Exadata DB Machine Custom Configuration
Storage scans, filters data Storage just ships blocks
Storage offloads DB* DB-unaware storage
Flash caches relevant data No DB-aware flash management
40 Gb/sec network 8 – 10 Gb/sec network
Pre-built for DB workload Assembled by customer
Redundancy built-in Build-your-own HA
Compression built-in Compression optional
Workload mgmt built-in Workload mgmt optional
* Backups, compression, decryption, data mining
Exadata is not a general-purpose system, It’s a Database Machine
168 CPU
cores in storage
13
Traditional Query Problem
Select sum(sales) where salesdate= „22-Jan-2010‟…
Sum
Return entire Sales table
What Were Yesterday’s
Sales?
• Data is pushed to database server for processing
• I/O rates are limited by speed and number of disk drives
• Network bandwidth is strained, limiting performance and concurrency
Discard most of
sales table
14
Exadata Smart Scan Improve Query Performance by 10x or More
Select sum(sales) where salesdate= „22-Jan-2010‟…
Sum
Return Sales for Jan 22 2010
What Were Yesterday’s
Sales?
• Off-load data intensive processing to Exadata Storage Server
• Exadata Storage Server only returns relevant rows and columns
• Wide Infiniband connections eliminate network bottlenecks
15
Exadata Smart Flash Cache Extreme Performance for OLTP Applications
• Automatically caches frequently-accessed „hot‟ data in flash storage
• Assigns the rest to less expensive disk drives
• Know when to avoid trying to cache data that will never be reused
• Process data at 50Gb/sec and up to 1million I/Os per second
Infrequently Used Data
Frequently Used Data
16
Exadata Hybrid Columnar Compression
Warehousing and Archiving
Warehouse/Query Compression
• 10x average storage savings
• 10x Scan I/O reduction
Archive Compression
• 15x average storage savings
– Up to 50x on some data
• Some access overhead
• For cold or historical data
Optimized for Speed Optimized for Space
Smaller Warehouse
Faster Performance
Reclaim 93% of Disks
Keep Data Online
Can mix compression types by partition for ILM
17
Exadata Hybrid Columnar Compression Reduce Disk Space Requirements
3x
10x 15x
1.4x
2.5 x
Uncompressed Data
Data Warehouse Appliances
OLTP Data DW Data Archive Data
Oracle
18
Exadata Delivers Extreme Data Capacity Greater Than 10x Data Compression is Common
Exadata 10x compression
Teradata 2650 1.4x compression
Netezza TwinFin 2-4x compression
EMC VMAX 3x (Oracle) compression
Competitors require multiple racks of storage
to equal 1 rack of Exadata storage*
* Using largest disks and best
compression for each vendor
19
Built-in Analytics Secure, Scalable Platform for Advanced Analytics
• Complex and predictive analytics embedded into Oracle Database 11g
• Reduce cost of additional hardware, management resources
• Improve performance by eliminating data movement and duplication
Oracle Data Mining
Uncover and predict
Oracle OLAP
Analyze and summarize
20
Complete Family Of Database Machines From Quarter-rack to Multi-rack Configurations
Oracle Exadata X2-2 2 to 8 (12 core) database nodes
Oracle Exadata X2-8 2 (64 core) database nodes
Quarter Half
Full, Multi-Rack
Full-Rack
Multi-Rack Exadata models and configurations
span the full range of enterprise database deployments
21
22
23
Exadata Delivers Extreme Consolidation Create a “Database-as-a-Service” Platform
• Large Memory
– Many databases can be consolidated
• Extreme Performance
– OLTP, DW, data mining, batch, reporting, loading,
backups, files in the database
– Encryption, compression
• Workload Management
– Manage SLAs via Quality of Service (QoS)
– CPU and I/O resource management
– Instance caging
Shrink data center costs, increase system utilization and promote
application integration
24
Exadata Delivers Extreme Value 1/6 the Price Versus IBM‟s Latest
$18,860,000
2 Exadata X2-8
$3,000,000
• More CPU cores
• More I/O performance
• Same storage capacity
– (excludes compression)
• HA configuration
IBM P795 + 4 DS8700s w/ Flash
6:1 advantage*
* Hardware list price comparisons
25
Standardized and Simple to Deploy Eliminate Systems Integration Trial-and-Error
• No Unique Configurations
– Same as Oracle engineering
– Delivered tested and ready-to-run
– Highly optimized, top-to-bottom
– Highly supportable
• Existing Apps, Tools, Practices
– Oracle Database 11g
– Full 30 years of Oracle DB capabilities
– No Exadata certification required
– Skills, knowledge base, people, partners
Deploy in Days, Not Months
26
Exadata in the Marketplace
Customers report astounding performance with NO changes to their
applications
27
Value Proposition
• Improve query performance by 10x
– Better insight into customer requirements
– Expand revenue opportunities
• Consolidate OLTP and analytic
workloads
– Lower admin and maintenance costs
– Reduce points of failure
• Lower risk
– Streamline deployment
– One support contact
28
Ed Forman, CTO, Targetbase
“We are an analytical driven company and we need an environment that is going to keep up with our analysts‟ train-of-thought. We have seen analytical processes that used to run in 20 minutes now run in 38 seconds. It is a fantastic performance improvement, but it is also a business changer for our clients.”
DRAFT
Metin Yilmaz
Manager, DBA
Turkcell IIetisim Hizmetleri A.S.
“With Exadata, analysts from the finance and marketing departments get the results quickly and can be much more productive. This will provide us competitive advantages in a highly competitive telecommunication market.”
DRAFT
Kolin Ohi
Business Intelligence Architect
Hotwire.com
“Oracle Exadata allows us to support increasingly complex analytic work on rapidly increasing data volumes. By removing system performance constraints, it encourages more innovative data use and gives us a significant advantage in the highly competitive discount travel industry.”
DRAFT
31
For More Information
http://search.oracle.com
or
www.oracle.com/exadata
oracle exadata
32