The following is intended to outline our general product ... · HITACHI USP V TERADATA 2550 NETEZZA...

The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions.The development, release, and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.

<Insert Picture Here>

Exadata V2 - Sun Oracle Database Machine

Database Machine Success

• Exadata V1 is succeeding in all geographies and industries against every competitor

“Every query was faster on Exadata compared to our current systems. The smallest performance improvement was 10x and

the biggest one was an incredible 72x.”

Simeon Dimitrov, Enterprise Resources Manager

“Call Data Record queries that used to run for over 30 minutesnow complete in under 1 minute. That's extreme performance.”

Grant Salmon, CEO, LGR Telecommunications

“A query that used to take 24 hours now runs in less than 30 minutes. The Oracle Database Machine beats competing solutions on bandwidth, load rate, disk capacity, and transparency.”

Christian Maar, CIO

Exadata V1 Data Warehousing Customers

Exadata Database Machine

• Version 1• World’s Fastest Machine for Data Warehousing

• Extreme Performance for Sequential I/O

• 10x Faster than other Oracle D/W Systems

• Version 2• World’s Fastest Machine for OLTP

• Extreme Performance for Random I/O

• 5x Version 1 Data Warehousing Performance

• Dramatic new Exadata Software Capabilities

The Architecture of the Future

Data Warehousing

and OLTP

Massively Parallel Grid

Sun Oracle Database Machine Hardware

• Complete, Pre-configured, Tested for Extreme Performance• Database Servers

• Exadata Storage Servers

• InfiniBand Switches

• Ethernet Switch

• Pre-cabled

• Keyboard, Video, Mouse (KVM) hardware

• Power Distribution Units (PDUs)

• Ready to Deploy• Plug in power

• Connect to Network

• Ready to Run Database

9

Exadata is Smart Storage

• Database Server

• Compute and memory intensive data processing executes in database servers

• Fully-parallelized joins and aggregations

Bandwidth Intensive Searches

Compute Intensive Processing

• Database Servers and Exadata storage work in conjunction to execute SQL

• Exadata cell is smart storage, not a complete database node

• Exadata Storage Server

• IO-bandwidth intensive searches executes in storage servers

• Exadata Smart Scans and Exadata Storage Indexes filter out data that is not relevant to a query

© 2009 Oracle Corporation – Proprietary and Confidential – 10 –

Sun Exadata Storage Server Hardware

24 GB DRAM

12 x 3.5” Disk Drives

2 Quad-Core Intel®

Xeon® Processors

Pre-installed Software:

• Oracle Exadata Storage Server Software

• Oracle Enterprise Linux

• Drivers

Disk Controller HBA with 512M battery backed cache

InfiniBand QDR (40Gb/s) dual port card

ILOM

Dual-redundant, hot-swappable power supplies

4 x 96GB Sun Flash PCIe Cards

Sun Oracle Database Machine

Full Rack Half Rack Quarter Rack Basic System

Scale Performance and Capacity

• Scalable• Scales to 8 rack database machine

by just adding wires

• More with external InfiniBand switches

• Scales to hundreds of storage servers

• Multi-petabyte databases

• Redundant and Fault Tolerant• Failure of any component

is tolerated

• Data is mirrored across storage servers

© 2009 Oracle Corporation – Proprietary and Confidential -13-13

Sun Oracle Database Machine

Hardware Improvements

• Same architecture as previous Database Machine• Same number and type of Servers, CPUs, Disks

Plus Flash Storage!

NewLatest Technologies

Faster

Better

1.8x Faster CPUs

1.33x More SAS Disk Capacity

2x More SATA Disk Capacity

1.5x Faster Disk Throughput

2x Faster Networking

2.25x More Memory

2x Faster Memory

2x More Ethernet Connectivity

Xeon 5500 Nehalem

600 GB SAS Disks

2 TB SATA Disks

6 Gb SAS Links

40 Gb InfiniBand

72 GB per DB Node

DDR3 DRAM

4 Ethernet links per DB Node

New

© 2009 Oracle Corporation – Proprietary and Confidential -14-

Sun Oracle Database Machine Full Rack

Pre-Configured for Extreme Performance

• 8 Sun Fire™ X4170 Oracle Database servers

• 14 Exadata Storage Servers (All SAS or all SATA)

• 3 Sun Datacenter InfiniBand Switch 36

• 36-port Managed QDR (40Gb/s) switch

• 1 “Admin” Cisco Ethernet switch


• Redundant Power Distributions Units (PDUs)

• Single Point of Support from Oracle

• 3 year, 24 x 7, 4 Hr On-site response

Add more racks for additional scalability


Sun Oracle Database Machine:

System Architecture

High BandwidthHigh BandwidthHigh BandwidthHigh BandwidthInterconnectInterconnectInterconnectInterconnect

Massively Parallel Massively Parallel Massively Parallel Massively Parallel StorageStorageStorageStorage

Clustered Clustered Clustered Clustered Database ServersDatabase ServersDatabase ServersDatabase Servers

…Compute Intensive Processing

Sun Oracle Database ServerSun Oracle Database ServerSun Oracle Database ServerSun Oracle Database Server




Sun Oracle Database ServerSun Oracle Database ServerSun Oracle Database ServerSun Oracle Database Server …

Data Intensive

Processing

Sun Oracle Exadata Storage ServerSun Oracle Exadata Storage ServerSun Oracle Exadata Storage ServerSun Oracle Exadata Storage Server

Data Intensive

Processing


Data Intensive

Processing


Data Intensive

Processing



Sun Oracle Database Machine:

5.3 TB SunFire Flash Cache

High BandwidthHigh BandwidthHigh BandwidthHigh BandwidthInterconnectInterconnectInterconnectInterconnect

Massively Parallel Massively Parallel Massively Parallel Massively Parallel StorageStorageStorageStorage

Clustered Clustered Clustered Clustered Database ServersDatabase ServersDatabase ServersDatabase Servers

…Compute Intensive Processing





Sun Oracle Database ServerSun Oracle Database ServerSun Oracle Database ServerSun Oracle Database Server …

Data Intensive

Processing


SunFire

Data Intensive

Processing


SunFire

Data Intensive

Processing


SunFire

Data Intensive

Processing


SunFire





• 36-port Managed QDR (40Gb/s) switch



• Redundant PDUs



Can Upgrade to a Full Rack

Sun Oracle Database Machine Half Rack






• 36-port Managed QDR (40Gb/s) InfiniBand switch



• Redundant PDUs



Can Upgrade to an Half Rack

Sun Oracle Database Machine Quarter Rack



Exadata Product Capacity

28 TB14 TB6 TB2 TBSAS

SATA

SATA

SAS 100 TB50 TB21 TB7.2 TB

100 TB50 TB21 TB7 TB

User Data2(assuming no

compression)

5.3 TB2.6 TB1.1 TB384 GBRaw Flash1

336 TB168 TB72 TB24 TBRaw Disk1

Full RackHalf RackQuarter RackSingle Server

1 – Raw capacity calculated using 1 GB = 1000 x 1000 x 1000 bytes and 1 TB = 1000 x 1000 x 1000 x 1000 bytes.

2 - User Data: Actual space for end-user data, computed after single mirroring (ASM normal redundancy) and after allowing space for database structures such as temp, logs, undo, and indexes. Actual user data capacity varies by application. User Data capacity calculated using 1 TB = 1024 * 1024 * 1024 * 1024 bytes.

Hardware Warranty & Support

• Included with ALL Exadata Products

• 3 year parts and labor are covered

• On-site support and repair is provided

• 24x7 coverage with a 4 hour response (Priority 1)

• Disk Retention Service (DMR): Failed drive, under warranty, can be replaced without returning the drive to Sun/Oracle

• Hardware support beyond the first 3 years is available, via separate contract, from Sun

• Additional higher levels of support can be purchased from Sun

Drastically Simplified Deployments

• Database Machine eliminates the complexity of deploying database systems• Months of configuration, troubleshooting, tuning

• Database Machine is ready on day one• Pre-built, tested, standard, supportable configuration

• Runs existing applications unchanged

• Extreme performance out of the boxMonths to

Days

22

Performance

Why is Oracle Faster?

� DB Processing in Storage

� Better Compression (10x)

� Smart Flash Cache

� Faster Interconnect (40Gb/sec)

� More Disks

� Faster Disks (15K RPM)

© 2009 Oracle Corporation

21

1011.4

7.5

0

10

20

30

40

50

60

Query Throughput Query Throughput with Flash

50

HITACHIUSP V

TERADATA2550

NETEZZATwinFin 12

SUN ORACLEDatabase Machine

23

Exadata Flash Extreme Performance

• Sun Oracle Database Machine has 5.3 TB of flash storage• OLTP - Over 1 Millions IOPS

• DW – 50 GB/s throughput

• Smart Flash Cache caches hot data• Not just simple LRU

• Knows when to avoid caching to avoid flushing cache

• Allows optimization by application tableOracle is the First Flash Optimized Database

New

Semiconductor Cache Hierarchy

Massive throughput and IOsthrough innovative Cache Hierarchy

• Database DRAM Cache• 400GB raw capacity

• Up to 4TB compressed user data

• 100 GB/sec

• Exadata Smart Flash Cache• 5TB raw capacity


• 50 GB/sec raw scan

• 1 million IO/sec

• Exadata disks• 100TB or 336TB raw


• 21 GB/sec scan

• 50K IO/sec

25

Sun Oracle Database MachineProduct Family – SAS Disk Drives

1M – 8M1,000,000500,000225,00075,000I/O per Second (IOPS)

N/A

3.6 GB/sec

1.5 GB/sec

2 TB

7.2 TB

1

1

Basic

100 - 400 GB/sec50 GB/sec25 GB/sec11 GB/secI/O Throughput (flash)

42 - 168 GB/sec21 GB/sec10.5 GB/sec4.5 GB/secI/O Throughput (disks)

2-8111Racks

56 – 224 TB28 TB14 TB6 TBUser Data (uncompressed)

200 – 800TB100 TB50 TB21 TBTotal Disk Capacity

28-1121473Exadata Storage Servers

16-64842Database Servers

2-8 Full RacksFull RackHalf RackQuarter Rack

© 2008 Oracle Corporation – 26 –

Scale-Out Architecture

• The database, ASM, and Exadata Cells each play a role in Oracle’s scale-out storage architecture

• Responsibilities are placed in the optimal location

• DB – Business Data Protection

• ASM – Reliable Storage Pool

• Exadata Cell – Database Intelligent Storage

• Seamless integration simplifies management

ASMReliable Storage

Pool

BusinessData Protection

Database Intelligent Storage

Exadata



ASMReliable Storage

Pool



Exadata


Scale-Out Business Data Protection

• Business Data Protection is implemented by the Database• Only the database can protect and recover business level objects

• Data Guard provides disaster protection and corruption protection • Automatically maintained standby copy of database

• Standby database is available for reads, reports, backups – not idle

• Flashback provides point-in-time recovery• Snapshot-like capabilities to rewind database to before error

• At row, transaction, table, or complete database level

• Recovery Manager (RMAN) provide backup to disk and tape• Archiving and corruption protection

• Data is validated as it is backed up

Storage Grid

• These work with Exadata with no changes

• Users and database administrators continue to use familiar tools


Exadata Storage Management & Administration

• Enterprise Manager • Manage & administer Database and ASM

• Exadata Storage Plug-in• Enterprise Manager Grid Control Plug-in to monitor & manage Exadata Storage Cells

• Comprehensive CLI• Local Exadata Storage cell management

• Distributed shell utility to execute CLI across multiple cells

• Lights-out Management (ILOM)• Remote management and administration of hardware



ASMReliable Storage

Pool



Exadata


Exadata Cell Exadata Cell

Exadata Storage Layout Example Grid Disks

• Cell Disks are logically partitioned into Grid Disks

• Grid Disk is the entity allocated to ASM as an ASM disk

• Minimum of one Grid Disk per Cell Disk

• Can be used to allocate “hot”, “warm” and “cold” regions of a Cell Disk or to separate databases sharing Exadata Cells

Grid

Disk


Exadata Storage Layout Example ASM Disk Groups and Mirroring

• Two ASM disk groups defined

• One for the active, or “hot” portion, of the database and a second for the “cold” or inactive portion

• ASM striping evenly distributes I/O across the disk group

• ASM mirroring is used protect against disk failures

• Optional for one or both disk groups

Exadata Cell Exadata Cell

Hot Hot Hot Hot Hot Hot

Cold Cold Cold Cold Cold Cold

Hot ASM

Disk Group Cold ASM

Disk Group


ASMReliable Storage

Pool



Exadata


Database Intelligence:

Smart Scan

Data Reduction

and Host Offloading


Traditional Scan Processing

• Smart Scan Example:

• Telco wants to identify customers that spend more than $200 on a single phone call

• With traditional storage, all database intelligence resides in the database hosts

• Most data returned from storage is discarded by database

• Discarded data consumes valuable resources, and impacts the performance of other workloads

��

IOs Executed:

1 terabyte of data

returned to hosts

��

DB Host reduces

terabyte of data to 1000

customer names that

are returned to client

��

Rows Returned

��

SELECT

customer_id

FROM calls

where amount >

200;

��

Table

Extents

Identified

��

I/Os Issued


Exadata Smart Scan Processing

• Only the relevant columns • customer_id

and required rows

• where amount>200

are are returned to database

• CPU consumed by predicate evaluation is offloaded

• Moving scan processing off the database frees CPU cycles and eliminates lots of unproductive messaging• Returns the needle, not the entire hay stack

��

2MB of data

returned to server

��

Rows Returned

��

Smart Scan

Constructed And

Sent To Cells

��

Smart Scan

identifies rows and

columns within

terabyte table that

match request

��

Consolidated

Result Set

Built From

All Cells

��

SELECT

customer_id

FROM calls

where amount >

200;


More Smart Scans

• Join filtering

• Join filtering is performed within Exadata cells

• Predicates on dimension tables can be used to filter fact tables

• Example - “Select total sales of all Italian wines”

• Item numbers of Italian wines are read from Item table

• Then translated into a Bloom filter

• Bloom filter is applied by the cells during the scan of Sales table to identify sales of Italian wines

• Index Scans

• When only the columns in an index are selected, the smart scan can be performed on the index instead of the full table


Smart Scans for Backup

• To optimize disk performance, the database issues large I/Os to read multiple blocks that need backup

• Even when some unneeded blocks are between the blocks needing backup, the database will issue a single efficient large I/O

• The database then discards the unneeded blocks

• Cells offload the overhead of filtering unneeded blocks

Large I/Os for multiple nearby blocks

Small I/O for isolated block


End-to-End Data Validation

• Exadata performs extensive validation of database blocks

• Checksums, block locations, magic numbers, head and tail checks, alignment, etc.

• End-to-End validation from database memory to cell

• Stops corrupted data from being written to disk

• Database blocks are validated when they are read from disk

• If block became corrupted on disk, the mirror block is read from another cell


Smart Scan Transparency

• Smart Scans correctly handle complex cases including

• Uncommitted data and locked rows

• Chained rows

• Compressed tables

• National Language Processing

• Date arithmetic

• Regular expression searches

• Partitioned tables

• Smart scans are transparent to the application

• No application or SQL changes required

• Returned data is fully consistent and transactional

• If a cell dies during a smart scan, the uncompleted portions of the smart scan are transparently routed to another cell

New

Exadata Hybrid Columnar Compression

• Data is grouped by columnand then compressed

• Query Mode for data warehousing• Optimized for speed

• 10X compression typical • Scans improve proportionally

• Archival Mode for infrequently accessed data• Optimized to reduce space

• 15X compression is typical

• Up to 50X for some data

10 10 10 1116

19 19 19 20 21

29

43

0

5

10

15

20

25

30

35

40

45

50

Size Reduction Factor by Table

OLTP Compression (avg=3.3)

Query Compression (avg=14.6)

Archive Compression (avg=22.6)

Real-World Compression RatiosOracle Production E-Business Suite Tables

• Columnar compression ratios

• Query = 14.6X

• Archive = 22.6X

• Vary by application and table

52

Exadata Storage IndexTransparent I/O Elimination with No Overhead

• Exadata Storage Indexes maintain summary information about table data in memory• Store MIN and MAX values of columns

• Typically one index entry for every MB of disk

• Eliminates disk I/Os if MIN and MAX can never match “where” clause of a query

• Completely automatic and

transparent

A B C D

1

3

5

5

8

3

Min B = 1

Max B =5

Table Index

Min B = 3

Max B =8

Select * from Table where B<2 - Only first set of rows can match

New

Best Machine for

Data Warehousing

Best Data Warehouse Machine

• Massively parallel high volume hardware to quickly process vast amounts of data• Exadata runs data intensive processing

directly in storage

• Most complete analytic capabilities • OLAP, Statistics, Spatial, Data Mining, Real-time

transactional ETL, Efficient point queries

• Powerful warehouse specific optimizations• Flexible Partitioning, Bitmap Indexing, Join indexing, Materialized Views, Result

Cache

• Dramatic new warehousing capabilitiesData Mining

OLAP

ETLETL

New

Benefits Multiply

1 TBwith compression

10 TB of user dataRequires 10 TB of IO

100 GBwith partition pruning

20 GB with Storage Indexes

5 GB Smart Scan on Memory or Flash

Subsecond

On Database Machine

Data is 10x Smaller, Scans are 2000x faster

In-Memory Parallel Queries

• One Sun Oracle Database Machine rack• 400GB of DRAM usable for caching

• Exadata Hybrid Columnar Compression enables 4TB data in DRAM

• Database release 11.2 introduces parallel query processing on DRAM cached data• Harnesses DRAM capacity of entire database cluster for

queries

• Technology for world record benchmark

315,842

1,018,321

1,166,976

ParAccel Exasol Oracle & HP

Exadata

QphH: 1 TB TPC-H

Faster than specialized in-memory warehouse databases DRAM has 100x more bandwidth than Disk

New

Source: Transaction Processing Council, as of 9/14/2009: Oracle on HP Bladesystem c-Class 128P RAC, 1,166,976 QphH@1000GB, $5.42/QphH@1000GB, available 12/1/09. Exasol on PRIMERGY RX300 S4, 1,018,321 QphH@1000GB, $1.18/QphH@1000GB, available 08/01/08.ParAccel on SunFire X4100 315,842 QphH@1000GB, $4.57 /QphH@1000GB, available 10/29/07.

DBFS - Scalable Shared File System

• Database Machine comes with DBFS shared Linux file system• Shared storage for ETL staging, scripts, reports and other application files

• Files stored as SecureFile LOBs in database tables stored in Exadata • Protected like any DB data – mirroring, DataGuard, Flashback, etc.

• 5 to 7 GB/sec file system I/O throughput

ETL Files in DBFS

Load into database using External Tables

New

ETL

More File Throughput than High-End NAS Filer

Best Machine for OLTP

Best OLTP Machine

• Only Oracle runs real-world business applications “on the Grid”

• Unique fault-tolerant scale-out OLTP database• RAC, Data Guard, Online Operations

• Unique fault-tolerant scale-out storage suitable for OLTP• ASM, Exadata

• Dramatic New OLTP

Capabilities

The Disk Random I/O Bottleneck

• Disk drives hold vast amounts of data• But are limited to about 300 I/Os per second

• Flash technology holds much less data• But can run tens of thousands of I/Os

per second

• Ideal Solution• Keep most data on disk for low cost

• Transparently move hot data to flash

• Use flash cards instead of flash disks to avoid disk controller limitations

• Flash cards in Exadata storage

• High bandwidth, low latency interconnect

300 I/O per Sec

Tens of Thousands of I/O’s per Second

Exadata Smart Flash CacheExtreme Performance

• Database Machine achieves:

• 20x more random I/Os• Over 1 million per second

• 2x faster sequential query I/O• 50 GB/sec

• 10x better I/O response time• Sub-millisecond

• Greatly Reduced Cost• 10x fewer disks needed for I/O

• Lower Power5X More I/Os than

1000 Disk Enterprise

Storage Array

WAN

Complete, Open, Integrated AvailabilityMaximum Availability Architecture

• Protection from• Server Failures

• Storage Failures

• Network Failures

• Site Failures

• Real-time remote standby open for queries

• Human error correction • Database, table, row, transaction level

• Online indexing and table redefinition

• Online patching and upgrades

Real

Application

Clusters

Real

Application

Clusters

ASMASM

Fast

Recovery Area

Fast

Recovery Area

Active

Data Guard

Active

Data Guard

Secure Backup

Secure Backup

Complete, Open, Integrated Security

Data Masking

Advanced Security

SecureBackup

Encryption and Masking

Database Vault

Label Security

Access Control

Configuration Management

Audit Vault Total

Recall

Monitoring

Best Machine forConsolidating Databases

Consolidating Databases

• Biggest driver of ongoing cost is• Multitudes of special-purpose systems

• Consolidate onto Database Machine• High performance for all applications

• Low cost platform for all applications

• Predictable response times in a shared environment

• Handles all data management needs

• Complete, Open, Integrated

ERP

CRM

Warehouse

Data Mart

HR

CRMERP

Warehouse Data Mart

HR


Exadata Storage GridI/O Resource Management

• With traditional storage,creating a managing shared storage is hampered by the inability to balance the work between users on the same database or on multiple databases sharing the storage subsystem• Hardware isolation is the approach to ensure separation

• Exadata I/O resource management ensures user defined SLAs are met• Coordination and prioritization between different groups/classes of work within a database and between databases

Storage

Switch/Network

.

Database

A

.

Database

B

Database

C

. . .


Exadata I/O Resource Management

• Exadata implements Smart I/O Resource management at three levels:

• Transaction Level

• Ensures higher priority transactions get better throughput

• Database Level

• Ensures each database gets fair share of disk bandwidth

• Usage Category Level

• Ensures interactive users are not slowed by reporting or batch

• Each Exadata cell coordinates its I/Os based on a grid-wide plan


Exadata I/O Resource ManagementMixed Workload Environments

• Ensure different transactions and users within a database are allocated the correct relative amount of I/O

• The same Database Plan that is used to direct CPU resources is used to direct I/O

• Host side RM has been in Oracle since 9iR1

• Example Database Plan:

• Interactive Txns: 70% of I/O bandwidth

• Reporting Txns: 30% of I/O bandwidth

• also any bandwidth not taken by Interactive

Interactive 70%

Reporting 30%


Exadata I/O Resource ManagementMulti-Database Environment

• Ensure different databases are allocated the correct relative amount of I/O resources

• Example Inter-database Plan:• Database A: 33% of I/O bandwidth

• Database B: 67% of I/O bandwidth

• Unused bandwidth can go to the other

• Or• Database B has lower priority than A

• E.g testing or development DB

• Within each database prioritize based on database plan

Database B

(RAC)

Database A

(Single-Instance)

33% 67%

Conclusion


Exadata V2 Summary

• Fastest Data Warehousing & OLTP Performance

• Best Data Warehousing & OLTP Cost/Performance

• Fault Tolerant & Scalable On-Demand

Sun Oracle Database MachineExtreme Performance for all Data Management

• Best for Data Warehousing• Parallel query on memory or Flash

• Compressed 4TB of data in DRAM, 50 TB in flash• 10x compressed tables with storage offload• Overall up to 5X faster than 11.1 for Warehousing

• Best for OLTP• Only database that scales real-world applications on grid• Smart flash cache provides 1 million IOs per second

• Compressed 1.2 TB of data in DRAM, 15 TB in Flash • Up to 50x compression for archival data• Secure, fault tolerant

• Best for Database Consolidation• Only database machine that runs and scales all workloads

• Predictable response times in multi-database, multi-application, multi-user environments

Q & A

If you have more questions later, feel free to ask

DB Machine Technology Comparison

----5 TBFlash

7.5 GB/sec10 GB/sec21 GB/secDisk IO Bandwidth

----50 GB/secFlash IO Bandwidth

12.6 TB72 TB140 TBUser data

(compressed)

128 GB108 GB576 GBMemory

144 x 300GB disks96 x 1TB disks168 x 600GB disksDisks

32 DB Cores8 DB Cores 64 DB CoresDatabase cores

0 Storage Cores96 Storage Cores112 Storage CoresStorage cores

32 Cores112 Cores176 CoresTotal cores

8.8 TB32 TB28 TBUser data

(uncompressed)

Footprint 1 rack1 rack1 rack

Teradata

2550

Netezza

TwinFin 12

Sun Oracle

Database

Machine

�

��

��

��

The following is intended to outline our general product ... · HITACHI USP V TERADATA 2550 NETEZZA...

Documents

Transcript of The following is intended to outline our general product ... · HITACHI USP V TERADATA 2550 NETEZZA...