NetApp Oracle Solutions - website.bgoug.online · WAFL®, Snapshot™, and FlexClone™: The Foundation NetApp’s Unique DNA WAFL® – Write Anywhere File Layout – Core block

NetApp Oracle

Lars Ejskjaer

Technical Partner Advisor Fujitsu Alliance [email protected]

NetApp Oracle Solutions

Agenda

� NetApp’s Unified Architecture Model

� NetApp’s Unique DNA

� NetApp’s value to the DBA

– Performance, Utilization and Virtualization

– Backup and Recovery

© 2009 NetApp. All rights reserved. 2

– Data Protection

– Disaster Recovery and Business Continuance

– Application DevTest

– ASM

– SnapManager® for Oracle®

� Q&A

The NetApp Unified Architecture Model

Competitors NetApp

Incompatible silos Compatible family

Platforms

Software &Processes


Experts &IntegrationServicesLots of experts and

integration servicesReduced training & service

requirements

Processes

Incompatible software;different processes

Unified software;Same processes

WAFL®, Snapshot™, and FlexClone™:The Foundation

NetApp’s Unique DNA

� WAFL® – Write Anywhere File Layout

– Core block and file storage services

– Resiliency features (RAID-DP™)� Snapshot

– Near-instantaneous, point-in-time “copy” of file system (vol) or LUN

Active Filesys or LUN

Inodes,

block maps

Snapshot™ FlexClone®


– Near-instantaneous, point-in-time “copy” of file system (vol) or LUN

– Read-only� SnapRestore®

– Near-instantaneous “rollback” of vol or LUN to prior Snapshot copy

� FlexClone– Near-instantaneous, WRITABLE

“copy” of vol or LUN– Same space-sharing

characteristics

A B C D

Disk blocks

D’ Y

Summary

Snapshot = A + B + C + D

Active = A + B + C + D’

FlexClone = Snapshot + Y

Incremental storage = 22 blocks

NetApp’s Value to the DBA

� Performance, Utilization and Virtualization

� Backup and Recovery

� Data Protection

� Disaster Recovery and Business Continuance



� Application DevTest

� ASM

� SnapManager® for Oracle®

Database Performance Tuning and Capacity Planning Challenges

� Storage can be a performance bottleneck

� Duplicate database copies are needed to maintain availability and SLAs

� Performance of each volume is limited to the number of spindles within


within� “Hot” volumes can’t be “helped”

by disks in other volumes� Capacity planning, forecasting

and management are imprecise tasks

� Availability, scalability and capacity must be maintained or increased while driving down costs…

Introducing FlexVol™

� Dynamically grow or shrink nondisruptively

� Dynamic reclamation of unused space

Volumes: Not Tied to Physical Storage


� Physical allocation invisible to users and applications

� Container level soft allocation

- Flexible provisioning- Higher utilization

Disk Group Disk Group Disk Group

Virtualized Pool

Aggregate

LUNs

Application-levelsoft allocation

1 TB

800 GB

Utilization Benefits with FlexVol: Thin Provisioning

FlexVol volumes:

� Separates space visible to users from physical disks

� Increases control of Container-levelsoft allocation

300200150


� Increases control of space allocation– Flexible provisioning– Higher utilization– Higher granularity

� ASM is aware of dynamic volume resizing operations

� Oracle® S.A.M.E.Physical Storage: 1 TB

FlexVol® Volumes:

2TB1 TB

300GB

200GB

200GB

50GB

150GB

100GB

Improved Utilization

waste waste needs more

Legacy

< 40% Used

App 1 App 2 App 3

App 1 App 2 App 3


One shared expansion pool – no wasted space

Disk View

NetAppApp 1 App 2 App 3

VirtualizedApp View

(60-80% used)

Any/Every Application…

Improved Performance

Regular volume� Volume performance limited� “Hot” volumes isolated and

vulnerable

Manage volume specific QoS with FlexShare™ and/or MultiStore…


FlexVol®� Shares spindle performance� Typically >2x improvement

– 10-disk vol. vs. shared 40-disk aggregate

� Automatic load shifting

� Tuning is no longer necessary

Performance Improvement

FlexShare™ and/or MultiStore…

Introducing FlexShare:

Volume-Specific QoS

� Provides workload prioritization

� Critical workloads get optimal

Standard Feature in Data ONTAP®Standard Feature in Data ONTAP®


� Critical workloads get optimal

response even when system is

fully loaded

� More effective storage consolidation

FlexShare Key Features

� Relative priority of different volumes

� Application (user) Vs. Data ONTAP® background (system) prioritization

0 10 20 30 40 50 60 70 80

Low PriorityVolumes

Medium PriorityVolumes

High PriorityVolumes

Latency (msec)

Without FlexShare™

Low PriorityVolumes

Medium PriorityVolumes

High PriorityVolumes

With FlexShare


prioritization

� Per-volume cache policies

� Dynamic configuration updates

0 10 20 30 40 50 60 70 80

Low PriorityVolumes

Latency (msec)

SystemClient

Time

Syste

m L

oa

d (

IOP

S)

Prioritize Client over System

Prioritize System over Client

FlexShare High Benefit Use Cases

� Consolidated Environments

– Production, Test, and Development on the same system

� Backup/Disaster Recovery Throttling

� Prioritize log I/O

� Match NetApp cache management with Oracle® cache management

� Etc.


� Automatic Link Aggregation

– Users can specify multiple Ethernet connections between the

Oracle® Database and the NFS server

– DNFS will distribute the load over all active links in a

balanced fashion (maximum: 16 NICs)

� Improved I/O Concurrency

Performancehttp://media.netapp.com/documents/tr-3622.pdf


� Improved I/O Concurrency

– Supports fully concurrent, asynchronous I/O on all platforms

� Summary

– Any Oracle application can easily take advantage of the

performance available on multiple parallel wires

– No other NFS client can do this…

High performance and high throughput, regardless of operating environment and without the need for expert tuning

Virtualization

� What about VMWare?

– Single Instance only

– Limited supportability

– See Metalink note: 249212.1

� Announced at OOW 2007

� New server virtualization software and support

– Oracle VM server – open source


249212.1

� OOW 2007 Demo and Whitepaper

– Application DevTest

media.netapp.com/documents/tr-3646.pdfmedia.netapp.com/documents/wp-7037.pdf

source

– Oracle VM Manager –private source

– See Metalink note: 464754.1

� Oracle on Demand Case

Study with NetApp

Other VM Platforms: not supported (per Metalink 249212.1)

Oracle On Demand – Virtualization Deployment Profile

Original State Consolidation Virtualization

6 servers 2 servers4 servers

2 x 8 DB

Prod Instance

2 x 8 MT

Prod Instance

2DC x 16 MT

Prod Instance

2DC x 16 DB

Prod Instance

2QC x 32

MT and DB

Prod Instance

18 months

Austin Data Center

�Compute grid: 25,000+ servers

�Storage grid: 11,500+ terabytes of NetApp storage

Legend

MT = Middle Tier

DB = Database


Original State Consolidation Virtualization

Space 12 rack units 4 rack units 2 rack units

Power @ Idle 1518 watts 960 watts 480 watts

Power @ 100% Load 2010 watts 1180 watts 660 watts

Perf Prod 1 2 ~2

Perf Dev & Test 1 1 ~1

2 x 8 DB

Dev Instance

2 x 8 DB

Test Instance

2 x 8 MT

Dev Instance

2 x 8 MT

Test Instance

2DC x 16 MT

Dev & Test

Instances

2DC x 16 DB

Dev & Test

Instances

2QC x 32

MT and DB

Dev & Test

Instances

Prod InstanceDB = Database

DC = Dual Core

QC = Quad Core

http://www.oracle.com/ondemand/collateral/network-appliance-brief.pdf

http://media.netapp.com/documents/tr-3712.pdf




� Data Protection





� ASM


Database Backup and Recovery

Challenges

� RMAN backups consume significant physical resources

� Cold backups lead to lower SLAs

� Separate backups on each platform


platform

� Time-to-recover from tape becomes prohibitive

� DBA’s time spent on non-value-add backup/restore tasks

� Backups performed with less frequency

Backup and Recovery with Snapshot and SnapRestore

� Significant time savings

� Stay online

� Reduce system and storage overhead

� Consolidated backups

� Backup more often

300GB Database


� Backup more often

Time in Hours

Time toBackup

Time toRecover

To Tape (60GB/Hr Best Case)

From Tape

Redo Logs

0 1 2 3 4 5 6 7 8

Snapshot™

Redo Logs

SnapRestore®A B C D

Active Filesys or

LUNSnapshot

D’ Y

FlexClone®

Primary Backup

Secondary BackupSnapVault®

Other NetApp and Oracle Solutions for Backup and Recovery� NetApp NearStore® Virtual Tape Library

– Certified with Oracle® Secure Backup

– Simultaneously backup Oracle Databases

(RMAN) and other file systems (NDMP)

� Archived redo logs, Flash Recovery Area (FRA), RMAN backups, and database exports

2

1


RMAN backups, and database exports

– Place on SATA drives in the same system with

FC drives

FASSystem

FCDisks

SATADisks

Highest Performing Data Files

Lower Performing Data Files

Use NetApp Deduplication for backups

Unified Storage

Architecture

NearStoreVTL600/VTL700

3




� Data Protection





� ASM


Protect Against Data Corruption with SnapValidator® for Oracle®

Database

Block Write

Prevention is better than recovery…

“… the larger outage was caused by a corruption of data files. The corrupted files began to replicate and quickly spread across the system, …”

April 11, 2002

eBay suffers three outages in three days

Oracle® Database

Oracle’s

Hardware


Volume Manager

I/O Driver

Operating System

Host Bus Adapter

SAN / NAS Cloud

Data passes through many layers before reaching the storage subsystem.

Corruption can occur at any point in the data path. http://www.oracle.com/technology/deploy/availability/pdf/corruption_wp.pdf

Hardware and/or firmware failures, misconfiguration, etc.

Hardware

Assisted

Resillient

Data

Initiative

SnapValidator: Prevents Oracle Block Corruption

� Validates writes to volumes that contain Oracle® data via Oracle’s checksum verification

� Supported with Oracle Database 9i and 10g (NFS, iSCSI, FCP)


� NetApp is the only HARD partner to support this initiative across its entire product line

� NetApp is the only HARD partner to support file systems (NFS)




� Data Protection





� ASM


Disaster Recovery Planning

Challenges

� 4x actual storage requirement

� Expensive transport infrastructure

� Cumbersome to install and manage

Secondary Data CenterPrimary Data Center

Database

Server

DedicatedLines andSwitches

Database

Server


manage

� Difficult to keep backup copies synchronized

RAID0 + 1Mirror Copy

RAID0 + 1Mirror Copy

Leveraging DR Resources

� “Ship” Changed Blocks

– SnapMirror

� “Ship” Logs

– Oracle Data Guard

� Logical/Physical

� Snapshot™ Standby

– 3rd Party Solutions

FAS System

FCDrives

SATADrives

Disaster Recovery Site


SLAs: policies and procedures, managed fail over and fail backSnapMirror®: synchronous, asynchronous or semisynchronous Oracle® Data Guard: physical or logical standbyVirtual Environments: virtual machine migration

� Leverage FlexClone®

– Test

– Development

– QA

– Patch Management

– Etc.

Drives

Leveraging Pooled Storage in a Virtualized DR Environment

Primary Site DR Site

VM1 VM2 VM2VM1

Clients

& Web

Ph

ysic

al

Ph

ysic

al

VM3

Ph

ysic

al

NetApp delivers a common storage pool across physical & virtual environments


Ph

ysic

al

Ph

ysic

al

Ph

ysic

al

CIFS

& NFS

Storage Pool Storage Pool

Data ONTAP®

Disaster Recovery Methodology

� Establish and update DR policies and procedures to meet SLAs

� Mirror PROD for initial copy to DR site

– If third-party PROD, use SnapMirror™ for OpenSystems

PROD Remote DR Clones


SnapMirror™ for OpenSystems for SAN-to-SAN

� Clone database replicas as needed

� Rehearse DR procedures to test failover and failback scenarios

Develop ● Test ● Deploy

Database Cloning and the Application Development Process

� Full or partial database copies required for:

– App and DB Development

– Maintenance (OS, DB

upgrade)

– Test and QA

PROD SECONDARY (DR)


– Test and QA

– Training and Demos

– Reporting and DW ETL

� Ability to do this quickly, correctly, and efficiently directly impacts Application Development and Deployment

DEV MAINT TEST/QA RPT/ETL

Traditional Approaches to Cloning: Challenges and Pain Points

� Limited storage resources

– 100% storage capacity overhead per instance, or custom partial extraction scripts

� Long lead-time requirements

– Process heavy (i.e., many “approvals” required)

LOB

Systems

Network


required)

– Storage resource allocation

– Manual or scripted operations subject to human error

� Downtime (offline) or degraded production system performance (online) during copy

� Restoring the baseline requires repeat of this process

Systems

Storage

Database

What if …

� … your DBAs and application developers

could create (and repetitively re-create) a

consistent copy of a database application

environment…….

– nearly instantaneously,


– nearly instantaneously,

– using negligible incremental storage,

– as needed, even for individual developers

– with little or no support of a storage admin?

� How would that impact the efficiency of your

application development team?

Benefits

� Instantaneous copies

� Low resource overhead

� Easy to manage

� Use database clones to test

Cloning Databases with FlexClone

Production Mirrored Copy

Dev 1 Dev NDev 2

Mask Cloned Data for

Compliance

© 2009 NetApp. All rights reserved.

migrations, apply bug fixes, upgrades, and patches

� Easily reallocate storage for a database and its transaction log files

32

Test 1 Test 2 Test N

Dev 1 Dev NDev 2Production DB Clones

A B C D

Active Filesys or

LUNSnapshot™

D’ Y

FlexClone®

more clones = higher productivity

New Database Development Methodology

� Mirror PROD for initial copy (DR)

� Clone database replicas as needed

� Create Snapshot™ copies of replicas for instant SnapRestore® of working

PROD Remote DR Clones


SnapRestore® of working databases

� Exploit multiple mirrors to increase fan-out and support multiple sites

Develop ● Test ● Deploy




� Data Protection





� ASM


Automatic

Storage

Management

NetApp and ASM: The Differentiated Stack

Logical Vol

File System

0010 0010 0010 0010 0010

0010 0010 0010 0010 0010Files

Tablespace

Tables

File System

File Names

Tablespace

Tables

Before ASM ASM


Management

Disks

Logical Vol

Disk Group

Logical Vol

Transparent Virtialization and Dynamic Thin Provisioning FlexVol® and MultiStore®

Space Efficient Integrated Backup, Recovery and Cloning SnapManager® for Oracle

Deduplication NetApp Deduplication

Datacenter Hardened RAID (survive double disk failures/group) RAID-DP™

WORM Storage SnapLock®

Oracle® HARD (end-to-end Oracle checksumming) SnapValidator® for Oracle

Oracle SAME – No need to mirror with RAID-DP! Aggregates

Unified Storage

Architecture

NetApp Adds Value to Oracle ASM

Oracle ASM NetApp Oracle ASM + NetApp

Data Resilience

Protect against Single Disk Failure Yes Yes Yes

Protect against Double Disk failure No Yes Yes

Passive Block corruption detection Yes Yes Yes

Active Block corruption detection Yes Yes Yes

Lost disk write detection No Yes Yes

Performance

Stripe data across ASM Disks Yes No Yes

Balance I/O across ASM Disks Yes No Yes


Balance I/O across ASM Disks Yes No Yes

Stripe data across Physical Disks No Yes Yes

Balance I/O across Physical Disks No Yes Yes

I/O prioritization No Yes Yes

Storage Utilization

Free space management across physical disks

No Yes Yes

Thin provisioning of ASM Disks No Yes Yes

Space efficient Cloning No Yes Yes

Data Protection

Storage Snapshot based Backups No Yes Yes

Storage Snapshot based Restores No Yes Yes




� Data Protection





� ASM


SnapManager for Oracle Overview

Database Server

Oracle Database 10g

Oracle Database 9i

� Provides easy-to-use GUI

� Integrates with the host application

� Automates and simplifies complex manual effort

– Clone – Backup– Restore– Recover

� Provides ability to initiate and manage backups

SnapManager

SnapManager

Oracle Database 11g SnapManager


SnapDrive®

NFS, FCP or iSCSI protocols

backups– Provides disk, LUN and Snapshot™

management infrastructure

ASM and RMAN

NetApp Storage Systems

SnapManager for Oracle Backup

Primary Data Center � SnapManager® backs up in seconds

� Snapshot copies can be verified locally or remotely for integrity then “stored”

App/DBServer

� Optionally catalog with RMAN and/or archive

Backup Steps1. Inventory

database files2. Map database


Benefits:

� Fastest and most storage efficient backups

� More frequent backups due to speed and efficiency

� No performance degradation from online backups

� Instantaneous access to backup data

� Automation eliminates manual errors

Time to backup: seconds

Snapshot

NetAppStorageSystem

2. Map database files to storage volumes

3. Place database in backup mode

4. Backup all files/LUN’s

5. Release database from backup mode

SnapManager for Oracle Restore

Primary Data Center � Near instantaneous restores from Snapshot

� Dramatically shortened recovery with automated log replays

� Automated recovery tasks

– RMAN, SCN, date+time,

database, tablespace,

data file, data block

(RMAN)

App/DBServer


Benefits:

� Extremely fast and accurate data restore and recovery

� Reduce downtime from outages

� Automation saves administrative time

(RMAN)

Time to restore: minutes

Snapshot

Roll transaction logs

NetAppStorageSystem

SnapManager for Oracle Clone

Primary Data Center

Benefits:

� Near-instant clones from Snapshot™

� Automated cloning operation

App/DBServer

Mask Data for Compliance


� Fast clones of Oracle® databases for test and dev

� Automation saves administrative time

� Improved productivity

� Decru LKM - Ideally suited for cloned production databases containing encrypted data.

Time to clone: seconds

Clones

NetAppStorageSystem

Decru Lifetime Key Management ™ (LKM)

Simplify

Simplify

Simplify

Simplifying Database Administrationbase

Operations and Deployment

Performance, Utilization and Virtualization

Backup and Recovery

Data Protection

FlexVol®, MultiStore®FlexShare™ &

OEM-GC Plugin

Snapshot™, SnapVault®, SnapRestore® & VTL

RAID-DP™, Decru®,SnapLock® &

SnapValidator®

© 2009 NetApp. All rights reserved.

42

Simplify

Simplify

Simplify

Simplify

Simplify

Disaster Recovery

ASM, DNFS, & Real App. Testing

SnapManager® for Oracle®

Integration and Optimization

Application DevTest

Arch. Redo, RMAN, Flsh Rec., & OSB

Content Mgmt., ILM, & Compliance

SnapMirror®

SnapValidator®

NetApp & Oracle

NetApp & Oracle

FlexClone® & SnapMirror

NetApp & Oracle

NetApp & Oracle Simplify

Thousands of NetApp customers are running Oracle…


Oracle Austin Data Center

Austin Data Center Statistics

� Compute grid: 25,000+ servers

–Largest Dell/Linux installation on earth

–Scaling at over 100 servers/week

� Storage grid: 11,500+ terabytes of NetApp storage

–Largest NetApp single installation on earth

–Scaling at over 60TBs of storage/month

90% NAS; 5% FC SAN; 5% iSCSI20% of all production DB storage is SATA11,500 terabytes in production globally


–Scaling at over 60TBs of storage/month

� Over 500 mission-critical customers' apps

hosted on the premises

� Primary data center for 70,000+ Oracle employees

� 2 acres of data center raised floor space

� Virtualization is the next phase…Sources: Oracle Magazine Mar/Apr 2005 and Oracle ADC Executives

“…for any of you who have visited our Austin Data Center, it is a

commercial for NetApp. We voted for them with our money.”

Safra Catz, OOW 2006

Oracle Utilization vs. Industry Avg.

80.0%40%

50%

60%

70%

80%

90%Utilization


47.0%

80.0%

00%

10%

20%

30%

40%

Industry Avg. Oracle Avg.

51% improvement in storage utilization with Data ONTAP 7G

Question and Answers


Благодаря

and now

“Simplicity is the ultimate sophistication.”

- Leonardo da Vinci

and now

Todor Todorov

NetApp Oracle Solutions - website.bgoug.online · WAFL®, Snapshot™, and FlexClone™: The Foundation NetApp’s Unique DNA WAFL® – Write Anywhere File Layout – Core block

Documents