oracle’s zero data loss · • scale-out storage servers for persistence and deduplication •...

Oracle’s Zero Data Loss Recovery Appliance: A New Era In Data Protection

Tim Chien Principal Product Manager Oracle Corporation Steve Wertheimer Senior Director of Development Oracle Corporation September 29, 2014

Copyright © 2014, Oracle and/or its affiliates. All rights reserved. |


Safe Harbor Statement 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.


Program Agenda

Today’s Data Protection Goals & Challenges

Recovery Appliance: A Fundamentally Different Approach

Recovery Appliance Configuration Overview

Customer & Analyst Experiences

Summary/Q&A

1

2

3

4

5


Key Database Protection Goals

• Business Goals – Never lose critical business data – No impact on business applications

• I.T. Goals – Ensure database level recoverability – Centralized service to protect all databases

5

Current backup solutions fail to achieve these goals


Backup Technology Evolution

6

Industry focus on Optimization of Storage Capacity, not on Application Availability

1980’s 1990’s Early 2000’s 2010’s

Tape Libraries

Network Attached Storage

Virtual Tape Libraries (VTLs)

Deduplication

Late 2000’s

Purpose-built Backup Appliances (PBBAs)


Backup Appliances Are Not Designed for Database Treat Databases as Just Files to Periodically Copy

7

Daily Backup Window Large performance impact on production

Data Loss Exposure Lose all data since last backup

Many Systems to Manage

Scale by deploying more backup appliances

Poor Database Recoverability

Many files are copied but protection state of database is unknown


Data Protection: Still A Complex Task

• Significant challenges for data protection?

• #1: Backing up and managing increasingly large data volumes

8

Ref. Oracle Database And Data Protection Survey (200 IT professionals), January 2014 http://www.dbta.com/DBTA-Downloads/ResearchReports/Oracle-Database-and-Data-Protection-Survey-Results-4331.pdf

Backups are too slow

Backups need constant management

Complex Recovery

Production slow down

Coordination of multiple backups


Zero Data Loss Recovery Appliance

9

Need a Fundamentally Different Approach to Protect Business Critical Database Data…

Copyright © 2014, Oracle and/or its affiliates. All rights reserved. | 10

Recovery Appliance Unique Benefits for Business and I.T.

Minimal Impact Backups Production databases only send changes. All backup and tape processing offloaded

Eliminate Data Loss Real-time redo transport provides instant protection of ongoing transactions

Cloud-Scale Protection

Easily protect all databases in the data center using massively scalable service

Database Level Recoverability

End-to-end reliability, visibility, and control of databases - not disjoint files


Zero Data Loss Recovery Appliance Overview

11

Delta Push • DBs access and send only changes

• Minimal impact on production • Real-time redo transport instantly

protects ongoing transactions

Protected Databases

Protects all DBs in Data Center • Petabytes of data • Oracle 10.2-12c, any platform • No expensive DB backup agents

Delta Store • Stores validated, compressed DB changes on disk • Fast restores to any point-in-time using deltas • Built on Exadata scaling and resilience • Enterprise Manager end-to-end control

Recovery Appliance

Replicates to Remote Recovery Appliance

Offloads Tape Backup


Fundamentally Different Approach Eliminate Data Loss


Data Guard Like Protection for all Databases

13

Generic Bit-Copy Backup Appliance • Backup taken once a day • Data loss exposure: all transactions since last backup

Zero Data Loss Recovery Appliance • Uses Data Guard continuous real-time redo transport

from in-memory buffers to immediately protect ongoing transactions

Protected Databases

Data loss is inherently bad. Even worse..it creates massive consistency issues across databases


Tape Library

One Way

Bi- Directional

Hub & Spoke

Remote Data Center Local Data Center

Data Loss Protection from Site Disasters

• Replication to Remote Appliance protects data from disasters or site failures

• Automated restore from Local Appliance or directly from Remote Appliance

14


Data Loss Protection from Users or Errors Autonomous Tape Archival

• Optional low-cost Tape Archival provides unalterable protection from: – Intentional attacks by hackers, employees, and viruses – Accidental data deletion using tape as long-term archive – Malfunction of primary data protection system

• Autonomous Tape Archival is built-in: just connect tape drives – Offloads tape backup activity from production servers – No expensive media manager backup agents on production systems – Tape drives run all day without slowing production systems – Oracle Secure Backup included, or use your own tape software

15

Supports Most Major Tape Libraries

16Gb FC


• Data validated on receive • Data periodically revalidated • Data validated on restore

• Data validated on receive, restore, and periodically on-disk

• Data validated when copied to and restored from tape

Tape Archive

Remote Replica

Production Appliance

Data Loss Protection from Corruptions Recovery Appliance Understands and Validates Database Formats End-to-end

16


Fundamentally Different Approach Minimal Impact Backup


Delta-Only Architecture No More Full Backups: Database Optimized Incremental-Forever

Delta Push

Delta Push Source Deduplication • Fast Incremental Backup - Never reads duplicate blocks - Never sends duplicate blocks • Eliminates Undo Blocks for committed transactions • Eliminates Unused Blocks

Delta Store Backup Management • Stores only change data • Compresses at block-level • Ships only Deltas to Replica

Protected Database Compressed Delta Store

Dramatic Database I/O & Network Savings

Change Data

No more full backups, only changes

Disaster Recovery

18


Space-Efficient “Virtual” Full Backups

• After one-time full backup, incrementals used to create virtual full database backups on a daily basis • Pointer-based representation of physical

full backup as of incremental backup time • Virtual backups typically 10x space efficient • Enables long backup history to be kept with

the smallest possible space consumption • “Time Machine” for database

Delta Store Protected Databases

Day N Incr

Day 1 Virtual Full

Day N Virtual Full

Day 1 Incr

Day 0 Full

19

No More Full Backups: Incrementals Forever Architecture


Fast Restore to Any Point-in-Time No Load On Production Servers to Merge Old Backups

• Directly restore any virtual full backup • All blocks referenced from virtual full are

efficiently retrieved • Eliminates production server overhead of

traditional restore and merge of incrementals

• Supported by the scalability and performance of the underlying Exadata hardware architecture

20

Delta Store Protected Database

RESTORE DATABASE TO DAY ‘N’

Day 0 Full

Day 1 Incr

Day N Incr

Day ‘N’ Full Backup


Minimal Impact Backups: Boost Production Server Performance Offload Backup Processing, Eliminate Expensive Backup Agents, Reduce Network Load

Database Server Today

Performance degrades with backups

Disk / Tape / Dedupe Backup Agents

Backup Operations: Merge, Compress, Validate, Delete,

Full/Tape Backups

Performance improves with backup offload

Database Server with Recovery Appliance

Disk / Tape / Dedupe Backup Agents

Backup Operations: Merge, Compress, Validate, Delete,

Full/Tape Backups

21

Delta Push


Fundamentally Different Approach Database-Level Recoverability


3 Fragmented backup management with no ownership

Today: Islands of Fragmented Backup Processes Managing Backups & Data Growth: #1 Enterprise IT Concern

1 Set of Databases

Well Defined Oracle Blocks

RMAN Backups

Stream of bits

2

Media Server

23


Recovery Appliance: End-to-End Data Protection Visibility Policy-based Management: Application-oriented Data Protection

1 Application-oriented

Protection Policy

Well Defined Oracle Blocks

RMAN Backup Set

End-to-end Oracle Block Validation

2

Recovery Appliance

24

3 EM Cloud Control: Integrated

Management


Unified End-to-End Control

• Recovery Appliance Admin centrally monitors and manages all database protection activity across all tiers

• Database Admin monitors the protection status of their database from disk, to tape, to replica – Offloaded replicas and tape backups

appear in Recovery Catalog

25

Tape

Remote Appliance

Enterprise Manager

Tape


Managing & Monitoring Enterprise Database Protection

with Zero Data Loss Recovery Appliance

Recovery Appliance Home Page

Protected Databases Activity Summary

Protected Databases Recovery Status Summary

Protected Databases Status Filters Select ‘Not Backed Up’

Notify DBA that this database needs a full backup ASAP!

At-a-Glance Backup Volume & Performance Charts

Filter Performance by Incoming Backup Rate

Filter Performance by Replication Rate

Drill Down Into Protected Databases Status Details

Protected Databases Overview Page

< 1 sec

Drill down into STORE36 status details

< 1 sec

Days GB

Current Recovery Window of 4.8 Days > Goal of 2 Days - OK

< 1 sec

< 1 sec

Days GB

Unprotected Data Window of Sub-Secs < Threshold of 24 Hrs - OK

< 1 sec

< 1 sec

Days GB

< 1 sec

< 1 sec

Recovery Window Space Needed of 350 GB < Reserved Space of 1 TB - OK

Days GB

Current Deduplication Ratio: 17:1 Last Copy To Tape Time: Sep 24, 2:42 AM PDT

Links to Data Protection Reports

< 1 sec

Days GB

< 1 sec

Protected Database Status Report

Protected Database Status Report Generated and Emailed on Daily/Weekly Basis

Similar data as in EM Protected Database Details

Total Data on Tape: 1.7 TB

Incoming Backup Volume for Sep 25: 4.5 TB Outgoing Backup Volume to Tape: 6.1 TB

< 1 sec

Protected Database Backup Report

RMAN Backup Job Start Time RMAN Backup Job Duration

RMAN Backup Size RMAN Backup Output Rate


Fundamentally Different Approach Cloud-Scale Protection


Policy-Based Cloud-Scale Database Protection

Recovery Appliance Protection Policies • Standardized

recovery window, tape retention, replication policies

Gold Policy – Customer Critical Disk: 35 days Tape: 90 days

Tape

Silver Policy – Internal Critical Disk: 10 days Tape: 45 days

Bronze Policy - Test/Dev Disk: 3 days Tape: 30 days

Replica

Replica Recovery Appliance also Policy-Based

55


Dynamic Policy-Based Space Allocation

• Space dynamically reallocated between databases to meet recovery window goals – E.g. Recover to any time in the

last 35 days

• Avoids storage islands and over-allocation typical of host or LUN oriented provisioning

56

Recovery Window Goals

Sales DB

Web DB

Sales Web


Easy to Provision Databases for Recovery Appliance Protection

• Recovery Appliance Administrator – Chooses new database in Enterprise Manager list – Assigns protection policy – Sets new database credentials

• Database Administrator – Selects Recovery Appliance target in Enterprise Manager – Enables real-time redo transport

57

New Database Fully Configured

in Recovery Appliance


Oracle Database 10.2, 11g, 12c

• Scale-out compute servers for data processing • Scale-out storage servers for persistence and deduplication • Scalable InfiniBand internal fabric

Connectivity scales with

capacity

Fibre Channel connectivity scales with

capacity

Modern Cloud-Scale Database Protection No Bottlenecks, No Single Point of Failure

10 GigE or InfiniBand

Single System Scales to Protect an Entire Data Center

58

Protected Databases


Configuration Overview


• Base Rack

– 2x Compute Servers with high speed connectivity * • Eight 10 Gb Ethernet ports per rack • Twelve 40 Gb InfiniBand ports per rack • Optional Four 16Gb Fibre Channel ports for tape connectivity

– 3x Storage Servers * • Each storage server has 12 high capacity disks

• Customers choose number of additional storage servers at purchase or later to increase capacity

• Full Rack: 2x Compute Servers, 14x Storage Servers

Start with Base Rack Configuration and Scale

Extra Storage

Fully Redundant

Recovery Appliance Base Rack

* Ref. Datasheet for latest configuration info


Recovery Appliance Sizing

• Recovery Appliance delta space needed is typically 1-2X size of protected databases

• Accurate sizing requires knowing: – Size and growth rate of database

• Minus temp, free space, committed undo

– Percent of database that changes each day – Redo generation rate – Desired recovery window in days – Compressibility of database

• Depends heavily on compression already used in DB

61

224 TB Usable Delta Space / Rack

Typically keeps 2 weeks history for databases with

total size 110TB to 220TB


• Add InfiniBand connected racks to scale

• Can expand with new generation hardware

• Scale to 18 Racks – Up to 216 TB/hour Delta Ingest and Restore

Recovery Appliance Scaling

Up to 120 TB/hour Virtual Backup Rate Up to 12 TB/hour Sustained Delta Ingest

Up to 12 TB/hour Restore Rate

Performance Per Full Rack


Edition-based Redefinition, Online Redefinition, Data Guard, GoldenGate – Minimal downtime maintenance, upgrades, migrations

Active Data Guard – Data Protection, DR – Query Offload

GoldenGate – Active-active replication – Heterogeneous

Active Replica

RMAN, Oracle Secure Backup, Zero Data Loss Recovery Appliance – Backup to disk, tape or cloud

Enterprise Manager Cloud Control – Site Guard, Coordinated Site Failover Application Continuity – Application HA Global Data Services – Service Failover / Load Balancing

RAC – Scalability – Server HA

ASM – Local storage

protection

Production Site

Flashback – Human error

correction

Recovery Appliance & Oracle MAA

63


ZS3 Backup Appliance • Fast low-cost storage for backup • RAID-Z and compression for high storage efficiency • Built-in snapshots and cloning for dev/test

Zero Data Loss Recovery Appliance • Real-time redo transport, store online history for weeks or months • Offload disk and tape backup activity from production servers • Scalable database protection as a service for the enterprise

Active Data Guard • Fast failover for for one database, highest application availability (RTO ~=0) • Real-time redo transport (RPO = 0) • Optimize performance and availability by offloading queries

Fast & Low-Cost Backup Store

Enterprise Data Protection

Dedicated System for Failover + Data Protection

Recovery Appliance & Oracle High Availability Technologies

64


Customer & Analyst Experiences

“ Recovery Appliance is a terrific addition to the Oracle Engineered Systems Portfolio. With its support for real-time redo transport, the new system is designed to bring Data Guard-like robust data protection not just for our mission critical applications - but for all the Oracle databases in our Data Center. ” Mach Terry Senior Vice President and Head of Information Technology Division Acleda Bank Plc

“ Virtual full backups using Zero Data Loss Recovery Appliance allow overall database recovery times to be reduced – this would be a critical benefit in CERN's environment where the monthly backup traffic is constantly growing and approaching the Petabyte range. ” Szymon Skorupinski DBA & Backup Service Manager CERN IT

“ Enterprises need to protect vital data for their critical business applications in real time without the downtime or data loss often experienced when using traditional approaches. The Zero Data Loss Recovery Appliance meets this demand with a simple, yet powerful solution that easily scales to protect databases enterprise-wide and meet ever-stringent recovery point objectives. ” Laura Dubois Program Vice President - Storage Practice IDC


Summary/Q&A


A New Era in Protection: Zero Data Loss Recovery Appliance Ground-Breaking Data Protection Solution Integrated with the Oracle Database

70

Minimal Impact Backups Production databases only send changes. All backup and tape processing offloaded

Eliminate Data Loss Real-time redo transport provides instant protection of ongoing transactions

Cloud-Scale Protection

Easily protect all databases in the data center using massively scalable service

Database Level Recoverability

End-to-end reliability, visibility, and control of databases - not disjoint files


Resources

• Recovery Appliance Best Practices Session – Tuesday September 30, Moscone South, Room 305, 3:45-4:30 PM

• Live Demonstrations – outside Keynote Hall, Moscone North • Recovery Appliance Web Page:

– http://www.oracle.com/recoveryappliance

• Oracle High Availability – http://www.oracle.com/goto/ha

• Oracle Maximum Availability Architecture – http://www.oracle.com/goto/maa

http://www.oracle.com/recoveryappliance�

http://www.oracle.com/goto/ha�

http://www.oracle.com/goto/maa�


Key HA Sessions and Demos by Oracle Development Monday, 29 September, Moscone South

11:45 MAA with Oracle Multitenant – Seeing is Believing, 104 1:30 Oracle Database 12c HA for Consolidation and Cloud, 306 2:45 Zero Data Loss Recovery Appliance, New Era in Data Protection, 307 4:00 Oracle GoldenGate 12c for Oracle Database 12c, 305 5:15 Maximizing Oracle RAC Uptime, 103 Tuesday, 30 September, Moscone South 10:45 Active Data Guard and GoldenGate HA Best Practices, 308 12:00 Zero-Downtime Mantra for Applications with Oracle RAC, 309 3:45 Zero Data Loss Recovery Appliance Best Practices, 305 5:00 Oracle WebLogic Server 12c: Oracle Database Integration, 304 5:00 Geodistributed Oracle GoldenGate and Active Data Guard:

Global Data Services, 307

Wednesday, 1 October, Moscone South 10:15 Resource Manager Best Practices 11:30 RMAN Best Practices in Oracle Database 12c, 104 12:45 Active Data Guard: Best Practices and Deep Dive, 104 2:00 Expert High-Availability Best Practices for Oracle Exadata, 102 4:45 GoldenGate Performance and Tuning for Oracle, NORTH 130

Thursday, 2 October, Moscone South 9:30 Best Practices for Zero Downtime, 103 12:00 Data Protection,Recovery and HA for Private Cloud, 103 Demos – Moscone South

Oracle Maximum Availability Architecture, SLD-140 Oracle Active Data Guard, SLD-145 Global Data Services, SLD-144

Continuous Availability, SLD-125 RMAN, Database Backup Cloud Service, Flashback, SLD-141 Oracle Secure Backup, SLD-142 Oracle Real Application Clusters, SLD-128

oracle.com/goto/availability https://blogs.oracle.com/MAA @OracleMAA

oracle’s zero data loss · • scale-out storage servers for persistence and deduplication •...

Documents