Oracle OpenWorld 2006 in Review - GCOUG Open World 2006 Part 1.pdf · Oracle OpenWorld 2006 in...
Transcript of Oracle OpenWorld 2006 in Review - GCOUG Open World 2006 Part 1.pdf · Oracle OpenWorld 2006 in...
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 remain at the sole discretion of Oracle.
Agenda
• OOW Highlights
• Database New Features
• Data Warehousing New Features
• Oracle Business Intelligence
• Oracle Enterprise Linux
• Service Oriented Architecture
Oracle OpenWorld 2006
• 41,000 attendees
• 1,621 sessions
• 420 exhibiting partners
• 200 press and analysts
• 550 product demos
• SOLD OUT hands-on labs• Special Programs: X-Treme
Weekend, User Groups,
Industry Tracks, JD Edwards,
Develop, Leaders Circle, CFO Summit
Oracle OpenWorld 2006Attendees
Regions:
• 86% North America
• 8% EMEA
• 5% APAC/Japan
• 2% LAD
Products:
• 49% Database Customers
• 31% Middleware Customers
• 30% E-Business Suite Cust.
• 20% PeopleSoft Customers
• 11% JD Edwards Customers
• 10% Siebel Customers
Industries:
• High Technology
• Financial Services
• Professional Services
• Public Sector
• Education
• Healthcare
• Communications
• Systems Integrators
• Manufacturing
• ISVs
• Retail
Oracle OpenWorld 2006We Painted The Town Red!
• Billboards
• Pole Banners
• Bus Shelters
• Kiosks
• Street Decals
• Newsstands
• Taxi Tops
• Airports
• Java Jackets
• Customer Badges
Oracle OpenWorld 2006Fun Facts
• 10,000 hours to set up network, voice,data, and wireless infrastructure
• 124 miles of fiber and copper network cable
• 1100 network switches, routers and firewalls
• 5125 computers and 300 VoIP telephones
• 412 Connection Center email stations
• 17,600 Ethernet ports and 150 Wireless Access Points
• 35,000 Collaboration Suite accounts for attendees
• 187 Shuttle Buses, 13 routes
• 4,351 signs – 82,000 square feet – 1.89 acres.
• 65,000 hotel room nights
Keynote Presentations
• Hardware Partners• AMD, Dell, H-P, Cisco, Sun, Network Appliance
• Service Provider / Integrator• Cap Gemini
• Oracle Executives• Charles Phillips, Chuck Rozwat, Thomas Kurian, John Wookey,
Larry Ellison
• Replays Available • http://www.oracle.com/openworld
Database Market
Great Year for Oracle Database
14.3%
15.0%
48.6%
22.0%
Oracle IBM Microsoft Other
Data Warehouse Market
Database for SAP
Database for Linux
9%
67%
3% 17%
4%
Oracle IBM Informix M/soft SAPDB
23%40%
16% 10%
11%
Oracle IBM M/soft Teradata Others0%
10%20%30%40%50%60%70%80%90%
100%
2003 2004 2005
Mark
ets
hare
(%)
Oracle
IBM
Others
Sources: Gartner May 2006, Worldwide RDBMS Total Software Revenue IDC May 2006. Worldwide RDBMS 2005 Vendor Shares, AMR March, 2005
Oracle’s 10g Grid Strategy
• Enable Grid Computing• Using Low Cost Servers and Storage
• With Intelligent Software• Database Servers and Fusion Middleware
• And Automated Management Tools• Reducing Errors and Lowering Costs
• To Modernize I/T Infrastructure • Highest QoS and Flexibility, Lower Cost
Oracle 10g Grid
Real Application
Clusters
Automatic Storage
Management
Application Server
Clusters
Grid Control
What a year it has been –New Product Releases
• Jan’06: Oracle Application Express
• Feb’06: Oracle Database Express Edition
• Mar’06: Oracle SQL Developer
Secure Enterprise Search
• Apr’06: Oracle Secure Backup
• May’06 Oracle Warehouse Builder
• Jul’06 Oracle Developer Tools for .NET
• Aug’06 Oracle Content Database
Oracle Records Database
Oracle Database Vault
<Insert Picture Here>
“The art of progress is to preserve order amid change and to preserve change amid order.”
Alfred North Whitehead:Principia Mathematica
Oracle Database 11g New Change Assurance Capabilities
Capture andReplay Workload
Set Up TestEnvironments
Perform Online Application Upgrades
Detect and TunePerformance Changes
Manage ExecutionPlan Evolution
Package Incidentsfor Support
Online ‘Hot’ Patching
Oracle Database 11gBeta announcement feedback
• CIO Magazine: 'Oracle 11g Database in beta testing‘
• CRN: 'Oracle Trots out Laundry List of 11g Database Perks‘
• Computerworld: 'Oracle lays out 11g in broad brushstrokes‘
• eWeek: 'Oracle Readies Beta of Next Generation 11g‘
• CRN: 'Oracle Opens Beta for 11g Database‘
• InternetNews: 'Oracle Talks Up 11g at Openworld‘
• The Register: 'Oracle releases 11g Database beta‘
• ZDNET: 'Oracle announces Database 11g beta'
Agenda
• OOW Highlights
• Database New Features
• Data Warehousing New Features
• Oracle Business Intelligence
• Oracle Enterprise Linux
• Service Oriented Architecture
Oracle is Ready to Scale Today- 20 Years of Investment
• Scalable Execution• Scale-up on Large SMPs
• Scale-out on Clusters
• Parallel Everything
• Query, DML, DDL, Loads, Unloads
• Non-escalating Row Locks
• Read Consistency
• Scalable Storage• Table Partitioning
• Automatic Storage Management
• Bigfile Tablespaces
• Corruption Detection
• Scalable Availability• RAC
• Flashback
• Data Guard
• RMAN
• Online Table Redefinition
• Read-only Tablespaces
• Scalable Management• Grid Control
• Automatic Tuning of Highly Complex Applications
• Database Resource Manager
<Insert Picture Here>
Sneak Preview of Upcoming Innovations
• Scalable Execution
• Scalable Storage
• Scalable Availability
• Scalable Management
Leader in Industry Benchmarks
OracleSAP Business Information Warehouse
(BIW) 2-tier
OracleTPC-H@10000GB
Non-Clustered
OracleSAP Transaction Banking (TRBK)
OracleTPC-H @3000GB
OracleTPC-H@1000GB
OracleTPC-H@300GB
OracleTPC-C Clustered
World Record
Leadership
Benchmark
As of October 23, 2006: Source: www.tpc.org & www.sap.com/benchmark.SAP TRBK Standard Application Benchmark: Sun Fire E6900 DB Server (8 1.5 GHz US-IV+ processors, 16 cores, 16 threads, 56 GB memory): 10,012,000 Day posts/hr, 6,664,000 Night bal accs/hr, Solaris 10, Oracle 10g, SAP Account
Management 3.0 (64-bit) Cert #2006018.The two-tier SAP Business Information Warehouse 3.5 Standard Application Benchmark suite performed on 2/28/06, by Fujitsu Siemens Computers in Paderborn, Germany, was certified on 3/14/06 with the following data. The scenario for 32GB main memory which corresponds to 467,200,000 records in fact
table was used. Load Phase - Average throughput total step 1+2 (rows/hour): 53,255,652. Query Navigation Steps: 377,280. The software configuration for all steps of the SAP BW Benchmark: Operating system central server: SUN Solaris 10. RDBMS: Oracle 10g. Platform Release: SAP NetWeaver '04. Configuration: Central server: Fujitsu
Primepower 850, 16 processors/16 Cores/16 threads, SPARC64 V, 2.16 GHz, 128 KB (D) + 128 KB(I) L1 cache, 4 MB L2 cache, and 32 GB main memory, Cert # 2006014.
Leader in Real-World Scalability
• Winter TopTen™- survey of largest and most heavily used databases in the world• Sizes are table sizes after compression. Indexes and RAID NOT
counted
• The World’s Largest Commercial Database • Yahoo @ 100TB
• The World’s Largest Linux Data Warehouse • Amazon.com @ 24.7TB
• The World’s Largest Linux OLTP systems• Amazon.com @ 4TB
• The World’s Largest Scientific database • Max Planck @ 222TB
• The World’s Largest Unix OLTP database• US Patent & Trademark Office @ 16.4TB
• Oracle runs nine of the ten largest UNIX OLTP systems
• Oracle runs 100 Percent of all Linux DSS and OLTP systems
Source: WinterCorp 2005 TopTen™ Program, www.wintercorp.com
Upcoming Scaling Milestones
• First Petabyte (1000 TB) Database is coming soon
• Traditional size leaders were Telcos, Retailers, Web
• New leaders are• Science
• High energy physics, astronomy
• Medicine
• Genomic and protein data
• Health Care
• Country-wide centralized health records with x-rays etc.
• Law Enforcement
• Many new areas make heavy use of LOB data
Upcoming Scaling Milestones
• First 1000 Processor Database is coming soon
• Biggest sites today in the 100 to 300 processor range
• Small clusters of large SMPs
• e.g. 2x64 or 4X32
• Large clusters of small SMPs
• e.g. 32x4
• Multicore doubles and then quadruples these
• First Terabyte buffer cache is coming soon
Server Results Cache
join
join
T1 GBY
join
T2 T3
GBY
join
T4 join
T5 T6
join
join
T1 GBY
join
T2 T3
GBY
join
T4 join
T5 T6
Q1: build query cachejoin
join
T1 GBY
join
T2 T3
GBY
Q2: Use it transparently
cache
• Automatically Caches results of queries, query blocks, or pl/sql function calls
• Cache is shared across statements and sessions on server
• Significant speed up for read-only / read-mostly data
• Buffer cache taken to the next level – Fully Consistent
OCI Consistent Client Cache
• Caches query results on client (e.g. App Server)
• Improves performance for read-mostly tables
• Faster response time - eliminates network round trips
• Saves database CPU,
• Users cheaper RAM on AS boxes
• Stays consistent with server
• Cache proactively refreshed when result set changes
• Like Cache Fusion between server and client
Application Server
Database
Consistent Caching
OCI Consistent Client Cache
• Unified interface with Server result cache
• Per process cache shared across multiple sessions/threads
• Works with all 11g OCI-based clients• Including ODP.Net, JDBC Type 2, PHP, ODBC
• DBA just sets size of client cache to activate
• Client side parameter to overwrite client cache size
alter table DEPT result_cache
Real PL/SQL Native Compilation
• Through 10gR2, PL/SQL compiled to a native DLL is significant faster than PL/SQL compiled for interpretation by the PVM
• Oracle translates PL/SQL source to C code and leaves the last step to a 3rd party C compiler
• But… some customers forbid deploying a C compiler on a production box
Real PL/SQL Native Compilation
• In 11.1, Oracle translates PL/SQL source directly to the DLL for the current hardware
• Moreover, Oracle does the linking and loading so that the file system directories are no longer needed
• So PL/SQL native compilation will work out of the box –and without compromising compiler regulations
• Only one parameter remains: the on/off switch, PLSQL_Code_Type
“Native” PL/SQL and JAVA Compilation
• 100+% faster for pure PL/SQL or Java code
• 10% – 30% faster for typical transactions with SQL
• PL/SQL• Just one parameter - On / Off
• No need for C compiler
• No file system DLLs
• Java
• Just one parameter – On / Off
• JIT “on the fly” compilation
• Transparent to user (asynchronous, in background)
• Code stored to avoid recompilations
Auto-Learning Optimizer
• This features completely automates SQL tuning
• This is the next generation result of the 10g Optimizer_Auto_Learn effort and Automatic Tuning Optimizer (ATO)
• The information that populates this comes from three automatic maintenance tasks: Automatic Optimizer Statistics gathering, Automatic Segment Advisor data gathering, and SQL Tuning Advisor data gathering.
Optimizer Statistics
• Faster
• Order of magnitude faster optimizer stats computations
• Efficient incremental maintenance of partitioned table stats
• Better
• Correlated stats (Number of Distinct Values – NDV and histograms) on multiple columns
• Safer
• Stats gathered can remain private in "statistics workspace" until they are ready to publish for general use
SQL Pivot
• Rotate rows into columns and vice versa
• Common operation for crosstab queries
1000Q4Jeans
100Q3Jeans
500Q3Jeans
1000Q1Jeans
1000Q2Shoes
2000Q1Shoes
AMOUNTQUARTERPROD
1000600Null1000Jeans
NullNull10002000Shoes
Q4Q3Q3Q1PROD
SELECT * from tSELECT * from tSELECT * from tSELECT * from tPIVOT (sum (amount)PIVOT (sum (amount)PIVOT (sum (amount)PIVOT (sum (amount)
FOR quarter inFOR quarter inFOR quarter inFOR quarter in‘‘‘‘Q1Q1Q1Q1’’’’, , , , ‘‘‘‘Q2Q2Q2Q2’’’’, , , , ‘‘‘‘Q3Q3Q3Q3’’’’, , , , ‘‘‘‘Q4Q4Q4Q4’’’’););););
DML Trigger enhancements
• DML triggers on tables are transparently faster• 25% speed-up for the firing update with a row-level trigger that
does SQL
• New compound trigger type• Encapsulates logic for each of the four timing points (before/after
statement/row)
• Allows safe and easy coding of the ancillary package paradigm (advocated by authors like Feuerstein and Tom Kyte)
• Create a trigger as ‘disabled’
• Specify mutual firing order for triggers on the same table
• Oracle Database 9i table compression works for DW• Compression for bulk load operations
• Direct Path Load
• Create Table As Select (CTAS) …
• Compression is now viable for OLTP• Regular (non-bulk) INSERT, UPDATE, DELETE
• Add, Drop Column
• Compression benefits• Reduced disk space consumption – typically 2x to 3x
• Reduced I/O
• Enhanced cache efficiency
• Better read performance
Compression for OLTP
Securing Files with Relational Data
• Many applications have both files and relational data
• e.g. Document Management, Medical, CAD, Imaging
• Simplicity and performance of file systems makes it attractive to store file data in file systems, while keeping relational data in DB
• This split compromises security, robustness, and management
• Disjoint security and auditing models
• Backup and recovery are fragmented
• Search across relational data and files is difficult
• Space management is complicated
• Separate interfaces and protocols
• “Two data managers for one application is one too many”
• The application must patch over the gap
Oracle SecureFilesConsolidated Secure Management of Data
• SecureFiles is a new database feature designed to break the performance barrier keeping file data out of databases
• Similar to LOBs but much faster, and with more capabilities
• Transparent encryption, compression, de-duplication, etc.
• Preserves the security, reliability, and scalability of database
• Superset of LOB interfaces allows easy migration from LOBs
• Enables consolidation of file data with associated relational data
• Single security model
• Single view of data
• Single management of data
High Performance
• Filesystem-like performance
• Preliminary performance compared to Linux filesystem
• No network, no logging, single stream of file reads or writes
• Remember that LOBs and SecureFiles offer a lot more functionality than vanilla files
0
20
40
60
80
100
120
0.1 1 10 100
File Read Performance
File Size (MB)
Speed (MB/sec)
0
20
40
60
80
100
120
0.1 1 10 100
File Write Performance
File Size (MB)
LOBs
Linux Files
SecureFiles
LOBs
Linux Files
SecureFiles
Database and File System Interfaces
• SecureFiles can be accessed by both database clients and file system clients
• Database clients use extended LOB interfaces
• JDBC, ODBC, OCI, .NET, PL/SQL
• 11g has a highly optimized streaming protocol for SecureFiles
• File system clients use the file system protocols implemented inthe XML DB repository
• FTP access
• WebDav Access
• Mountable NFS V4 file system (preview only)
Unified Security Model
• Unified security and identity model for relational and file data• Huge reduction in attack footprint
• And better security features
• Encryption on the network
• Transparent data encryption (TDE) secures stored data• Existing applications require no changes
• Backups, log files, etc. are also encrypted
• Compression is applied before encryption to minimize size
• Fine-Grained Auditing for content based auditing of access
• Label Security restricts access based on• Level (e.g. confidential)
• Compartment (e.g. finance)
• Group (e.g. japan)
Storage Optimized
• De-duplication
• Multiple copies of identical SecureFiles are stored only once
• Great for document management, testing, product “homes”
• Significantly better performance for copy operations
• Compression• Intelligent – Automatically determines if data is compressible
• 2x to 3x compression for typical files
• Improves performance by reducing IO, buffer cache requirements, redo generation and encryption overhead
• Transparent to applications
Designed from Scratch
• SecureFiles is a major rearchitecture of how the database handles file data
• Not an incremental improvement to LOBs
• Entirely new:
• Disk format
• Network protocol
• Versioning and sharing mechanisms
• Caching and locking
• Redo and undo algorithms
• Space and memory management
• Cluster consistency algorithms
The Best of Files and Databases
• SecureFiles have all the leading-edge file system capabilities
• Deduplication, Encryption, Compression, Logging
• SecureFiles have advanced DB capabilities not in file systems
• Transactions, Read Consistency, Flashback
• Readable Standby, Consistent Backup, Point in Time Recovery
• Fine Grained Auditing, Label Security
• XML indexing, XML Queries, XPath
• Real Application Clusters
• Automatic Storage Management
• Partitioning and ILM
• Search across meta-data and file content
• Capabilities go far beyond any other database or file system
• Having the best of both worlds removes the need to compromise
Out of the Box Benefits
• Existing LOB applications benefit from
• Better performance and scalability
• Security
• Optimized Storage
• Immediate benefits for XML DB, Spatial, Intermedia, OLAP, Content DB, etc.
• File system based applications can switch to SecureFiles using Oracle’s native implementation of industry standard files protocols
• FTP, WebDAV, NFS V4 (preview)
• Easy conversion from LOBs to SecureFiles using Online Table Redefinition
Partitioning Enhancements
• Composite List-Hash Partitioning• Enables to hash subpartition a list partitioned object, for example to enable
partition-wise joins.
• Composite List-List Partitioning• Enables to logically list partition along two dimensions, e.g. list partition by
country_id and list subpartition by sales_channel
• Composite List-Range Partitioning• Enables to logically range subpartition within a given list partitioning
strategy, e.g. list partition by country_id and range subpartition by order_date
• Interval Partitioning• Define the interval criteria and the database will know when to create new
partitions for new or modified data. e. g. such as adding new daily partitions
• Virtual Column-Based Partitioning
Predictive Analytics
• DETECT - Could be used to quickly find expense transactions that seem unusual.
• PREDICT with EXPLANATION – Could provide a detailed IF/Then rule, e.g If AGE > 50 AND INCOME < $72,000, THEN CUSTOMER will BUY, with PROBABILITY = 95%.
• GROUP – Could be used to quickly find customers who share similar purchase, demographic, and behavioral patterns
• To reduce development coding, this work can be also done within an Excel spreadsheet
Availability Focus Area
• Oracle database development launched the “Database Focus Area” in 1998 (8i release)
• Database availability was primarily provided by underlying hardware/OS platform
• Goal was to abstract the database from the limitations of the hardware platform• Pool and virtualize underlying hardware resources
• Create fault-tolerance using low cost hardware
• Create innovative solutions to long standing availability problems• e.g. Human error, corruptions, etc.
Established a New Standard for High Availability
Old Standard –HA Components at High Cost
Single
SMP Server
Idle Cluster
Failover Server
Single
Storage Array
Premium Priced
Clusterware & Volume Manager
Premium Priced
DB Backup Module
No Good Solution for:• Human Errors• Data Changes• Software Upgrades• Corruption Protection
Premium Priced
Remote Mirrorin
g
Idle DR Systems
New Standard –Oracle Maximum Availability Architecture
• Best availability solution
• For every outage category
• At lowest cost • Low cost grid servers and storage
• No idle hardware
• Bundled clusterware, volume manager, and DR
• Integrated, automated, and simple to use
• High availability for the masses
Best Availability AND Lowest Cost
We Wrote the Book on Best Practices –Maximum Availability Architecture (MAA)
• Operational Practices are key
• Technology alone is not enough
• MAA is a blueprint for achieving HA & DR
• Tested, validated, and documented best practices
• Database, Storage, Cluster, Network
• 25 person year effort
• http://otn.oracle.com/deploy/availability
Maximum Availability = Unbreakable Architecture + Best Practices
MAA
Tolerate, and
From O
utages
Prevent,
Recover
Best Computer Failure Protection- at Lowest Cost
Real Application Clusters
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
RAC Enables High Availability
• RAC pools standard low costservers
• Great for Scalability
• Perfect for Availability• No Idle Resources
• Failover to any other node
• Integrated clusterware, file system, volume manager
• Application notification
• Thousands of production customers
Designed to Tolerate Failure of Low Cost Servers
Database
Storage
Next Release RAC Sneak Preview
• Oracle focus is on continued maturity and stability
• Fine tuning performance, scaling, failover, management
• Still the only vendor to deliver an OLTP cluster solution
• Faster and more robust handling of instance crashes and hangs
• Most of this also in the 10.2.0.3 patchset
• Transaction tagging
• Quickly identify which transactions committed before failure
• Seamless integration with XA and Microsoft Transaction Server
• Optimized Cache Fusion protocols
Optimized Cache Fusion Protocol
• Next generation of Cache Fusion protocols.
• Highly optimized protocols for handling common internode operations
• Reduction in cluster messaging wait time by 50% to 90% in many applications
Database
Optimized Cache Fusion Protocol
• Locality Optimized Fusion Protocol (10gR2)• Oracle detects when most segment accesses are coming
from a single instance
• Optimizes access by that instance
• Reader Optimized Fusion Protocol• Highly read-intensive segments are automatically converted
to a reader optimized messaging protocol
• Long Query Optimized Fusion Protocol• After all modified cache buffers at start of query are written
to disk, no more need for RAC communication
• Direct reads for non-parallel table scans
• Update Optimized Fusion Protocol• Update block in parallel to readers releasing the block
Best Data Protection- at Lowest Cost
Continuous Data Continuous Data
ProtectionProtection
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
Trade cheap disk spacefor expensive downtime
Best Storage Failure Protection- at Lowest Cost
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
Storage Failure
Human Error
Corruption
Site Failure
Four Data Failure Types
Data Mirroring with ASM
• ASM mirrors data across low costmodular storage arrays
• Automatically remirrors when disk or array fails
• ASM is still free
• Next Release Sneak Preview
• Automatically repair corrupt blocks
from mirror copy
• Temporarily freeze I/O to unresponsive storage and replay
writes when it returns
Designed to Tolerate Failure of Low Cost Storage
Database
Storage
Best Human Error Protection- at Lowest Cost
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
Storage Failure
Human Error
Corruption
Site Failure
Human Error
• Goal is to quickly analyze and repair
• For Localized damage
• Need surgical analysis and repair
• Example – deleted wrong order
• For Widespread damage
• Need complete back-out to avoid long downtime
• Example – batch job deletes this month’s orders
Other Downtime
Single Biggest Cause of Downtime
Human
Error
Other
Causes
Investigate Errors using Time Navigation
• Flashback Query
• Query all data at point in time
Tx 1
Tx 2
Tx 3
select * from Emp AS OF ‘2:00 P.M.’ where …
select * from Emp VERSIONS BETWEEN‘2:00 PM’ and ‘3:00 PM’ where …
select * from FLASHBACK_TRANSACTION_QUERYwhere xid = ‘000200030000002D’;
� Flashback Transaction Query
– See all changes made by a transaction
� Flashback Version Query
– See all versions of a row between times
– See transactions that changed the row
Flashback Database
• A new strategy for point in time recovery
• Flashback Log captures old versions of changed blocks
• Think of it as a continuous backup
• Replay log to restore DB to time
• Restores just changed blocks
• It’s fast - recover in minutes, not hours
• It’s easy - single command restore
Flashback Database to ‘2:05 PM’
“Rewind” button for the DatabaseData Files FlashbackLog
New BlockVersion
Disk Write
Old BlockVersion
Flashback Error Correction
• Correct errors at any level
• Database• Flashback Database - restore
database to time
• Table• Flashback Table - restore
contents of tables to time
• Flashback Drop - restore dropped table
• Row• Flashback Query - restore
individual rows
Order
Database
Customer
Restore Points (10gR2)
• 10gR2 allows the declaration of named restore points
• Easy way to return to a known time
• Guaranteed restore points ensure that flashback logs are retained
• 10gR2 can flashback through:• Previous database recovery and open resetlogs
• Switchover to standby
• Previous Flashback DB
• Flashback and incremental backup apply can be combined to maintain a reporting database
• Similar to split mirror and merge mirror
Create Restore Point Known_good_point; …..
Flashback Database to Restore Point Known_good_point;
Revolution in Recovery
• Flashback Revolutionizes Error Recovery
• Operates on just changed data
• Time to correct error equals time to make error
• Minutes instead of hours
• Flashback is Easy• Single command instead of complex procedure
• Very low performance overhead – less than 2%
Correction Time = Error Time + f(DB_SIZE)
0
10
20
30
40
50
60
70
80
Time To Recover (minutes)
Flashback
Traditional Recovery
Next Release Flashback
• Flashback Transaction• Push button back-out of a transaction and all conflicting
transactions
• Optimized Flashback Database
• Even lower overhead for inserts and direct loads
Flashback Data Archive
• Long term retention - years
• Automatically stores all changes to selected tables in Flashback Data Archive• Archive cannot be modified
• Old data purged per retention policy
• View table contents as of any time using Flashback SQL
• Uses• Change Tracking
• ILM
• Long term history
• Auditing
• Compliance
ORDERSORDERSORDERSORDERS
User Tablespaces
Flashback Data Archive
ArchiveTables
Oracle DatabaseOracle DatabaseOracle DatabaseOracle Database
Changes
Total Recall
Select * from ordersSelect * from ordersSelect * from ordersSelect * from ordersAS OFAS OFAS OFAS OF
‘Midnight 31‘Midnight 31‘Midnight 31‘Midnight 31----DecDecDecDec----2004’2004’2004’2004’
Best Corruption Protection- at Lowest Cost
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
Storage Failure
Human Error
Corruption
Site Failure
RMAN - Best Tool for Repairing Corruptions
• RMAN’s deep integration with the database engine makes it the best tool for DB backup & recovery
• Smart
• Sophisticated backup and recovery strategies
• Fast
• Optimized backup to disk for fastest recovery
• No extra redo during backup
• Block level incremental backup
• Reliable
• Block contents validated during backup
• Easy
• Simple management with Enterprise Manager
• Supports over 20 Media Managers
• Veritas, Legato, Tivoli, HP, Oracle Secure Backup, etc.
• Still FREE
Tape LibrariesTape Libraries
Oracle DatabaseOracle Database
RMAN Engineinside Oracle DB
RMAN is used at thousands of enterprise sites
Enterprise Manager & 3rd Party Tools
Media Manager(Veritas, Legato, etc.)
Automated Disk-Based Backup
• RMAN fully automates DB backup and recovery to disk
• Set and Forget
• Nightly incremental backup rolls forward recovery area backup
• Changed blocks are tracked in production DB
• Full scan is never needed
• Dramatically faster
• Blocks validated to prevent corruption of backup copy
• Use low cost ATA disk array for recovery area
Two Independent Disk Systems
Flash RecoveryArea
Nightly ApplyValidated
Incremental
ArchiveTo Tape
DatabaseArea
Oracle SecureBackup
Oracle Secure Backup
• Media manager for RMAN backup to tape of Oracle9i and 10g databases
• Backup file system files• Oracle Home, App Server, etc.
• Centralized management of data center backups
• Scalability to low 100’s of servers
• Best integrated backup of Oracle Databases
• End-to-End Encryption of database backups
RMAN
Databases
Linux, UnixWindows,
Filers
File Systems
Supports popular tape libraries & drives
Oracle Secure Backup -World’s Fastest Database Backup
• Oracle Secure Backup is fast
• 10% to 25% faster than competition
• Tight integration with Database optimizes performance
• Direct calls into DB engine
• Unused Block Elimination skips all unused space
• Next release even faster by not backing up Undo tablespace dataSpeed of Backup
Leading Vendor
Oracle Secure Backup - Lowest Price
• Oracle Secure Backup price is just $3000 per tape drive
• Backup to virtual tape device (disk) is Free
• Low end bundle - one Free direct attached tape drive per database server
Feature
Leading
Vendor OracleTape Drive 3,000$ 3,000$
Client Host 1,000$ Free
Oracle Agent 6,000$ Free
Central Server 1,000$ Free
SAN Backup 3,000$ Free
Encryption 2,000$ Free
Media Server 2,000$ Free
NFS Filer-NDMP 6,000$ FreePrice
Leading Vendor
Investigation Lengthens Outages
• In an outage, uncertainty and confusion are common
• Largest part of downtime is:• Investigating the problem
• Choosing the right solution
Total Downtime
Time to Investigate& Time to plan
Time to Repair
Data Repair Advisor
• Analyzes failures based on symptoms• e.g. “Open failed” because datafiles missing
• Intelligently determines repair strategies• Aggregates failures for efficient repair
• e.g. for many bad blocks restore entire file
• Presents only feasible repair options
• Are there backups?
• Is there a standby database?
• Ranked by repair time and data loss
• Can automatically perform repairs
Database Backup & Recovery Improvements
• Intra-file parallelism for backup and restore of Bigfiles
• RMAN duplicate database works across the network
• Instantiate standby or test database
• No extra temporary copy
• Integration with Windows Volume Shadow Service (VSS) Backup Interface
• For tape or snapshot backups on Windows
Best Disaster Protection- at Lowest Cost
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
Storage Failure
Human Error
Corruption
Site Failure
Best Site Failure Protection- at Lowest Cost
• Best Recovery Point Objective (RPO) - Zero data loss
• Best Recovery Time Objective (RTO) – Seconds to failover
• Reporting on standby site using Logical Standby
• Best Management on Failure - no need to reinstantiate primary after failover
• Compared to remote mirroring, Data Guard provides better:• Data protection, network usage, hardware utilization - for free
ProductionDatabase
StandbyDatabase
Reporting on
Logical Standby
Data Guard
Redo Shipping
with Real Time Apply
Standby Database Provides Fault Tolerance from Unplanned Failures
• On unplanned failure, failover to standby first
• Then take your time to diagnose and repair the failure
• Use for cluster failure, corruption, full storage failure, operator error, network failure
Computer Failures
Data Failures
UnplannedDowntime
Storage Failure
Human Error
Corruption
Tolerate using
Data Guard
Automatic Failover to Standby (10gR2)
• Fast-Start Failover implements automatic failover to standby database• Triggered by failure of Site, Hosts, Storage, or Network
• Works with and supplements RAC server failover
• Failover occurs in seconds• Comparable to cluster failover
• Original site automatically rejoins the configuration after recovery
Production Database
SynchronousRedo Shipping
Automatic Failover Physicalor Logical
Standby DB
Physical Standby with Real-Time Query
Physical Standby
Database
Primary
Database
Real-time
Queries
Continuous Redo Shipment and Apply
Concurrent Real-Time
Query
• Read-only queries on physical standby concurrent with redo apply• Supports RAC on primary / standby
• Queries see transactionally consistent results
• Immediate appeal to the many users of physical standby
• DR with real time query is unique in the industry – no idle resources
• Handles all data types, but not as flexible as logical standby
Snapshot Standby – Leverage Standby Database for Testing
• Convert Physical Standby to Snapshot Standby and open for writes by testing applications
• ALTER DATABASE CONVERT TO
SNAPSHOT STANDBY;
• Discard testing writes and catch-up to primary by applying logs
• ALTER DATABASE CONVERT TO
PHYSICAL STANDBY;
• Preserves zero data loss • But no real time query or fast failover
• Similar to storage snapshots, but:• Provides DR at the same time
• Single copy of storage, good for VLDB
• No idle resources
Physical Standby Apply Logs
Snapshot Standby Perform Testing
Open Database
Back out Changes
Continuous Redo Shipping
Data Guard Improvements
• Fast-Start Failover time cut in half
• Fast-Start Failover for Maximum Performance (Async) Mode• Automatic failover for long distance standby
• May incur small data loss – administrator can limit
• Fast Redo Transport
• Streaming asynch network protocol
• Redo Transport Compression for gap fetching
• Fast Incremental Backup on Physical Standby DB
• Tracks changed blocks on standby database
Data Guard Improvements – Cont
• New parallel media recovery doubles redo apply performance
• SYS user no longer required for redo shipping• Use strong authentication (certificates)
• RMAN understands Data Guard configurations• Tracks correct file names and original location of backups
• Gives preference to local backups
SQL Apply and Streams Improvements
• Support XMLType data type (CLOB)
• Support Transparent Data Encryption (TDE)
• Apply performance improved up to 50%
• SQL Apply additional PL/SQL Package support • Support Fine Grained Auditing
• Support DBMS_RLS (Virtual Private Database)
Which Standby to Use
• Physical Standby is the base Disaster Recovery configuration
• Three ways to overcome limitations• Logical Standby
• Snapshot Standby
• Streams
PhysicalStandby
+Supports all data types
+Fast Automatic Failover
+Application Transparent
+High Performance Apply
- No rolling upgrades+Real-time Query
- No updates allowed+Disaster Recovery
LimitationsBenefits
Which Standby to Use
• Reporting with Updates – Logical Standby
• Allows updates to reporting tables, adding indexes, materialized views, auditing, etc.
• Limitations - data types, slower apply
• Testing – Snapshot Standby
• Full updates to all tables
• Limitations - slow failover, no real-time reporting
• Multi-Site Updates – Streams
• Active-active updates on primary and ‘standby’, heterogeneous hardware
• Limitations - loses application transparency, data types, apply performance
• Logical and Streams allow Rolling Upgrades
LogicalStandby
Reporting with Updates
SnapshotStandby
Testing
StreamsMulti-Site Updates
All can achieve zero data loss Disaster Recovery
Test
DatabaseProduction
Database Redo Shipment and Apply
Replay Workload on Snapshot Standby
CAPTURERe
c REPLAY
Best Data Protection at Lowest Cost
Achieve any Level of Availability using Low Cost Hardware
Dramatic Advances in Ease of Use
ASM MirroringStorage Failure
Protection
FlashbackHuman Error
Protection
Data Guard / StreamsSite Failure Protection
Flash Recovery
AreaCorruption
Protection
Oracle Secure BackupArchive to Tape
Online ReconfigurationOnline Reconfiguration
Best Online System Changes
Online UpgradeOnline Upgrade
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime
Online Reconfiguration –Capacity on Demand
• CPU
• Add/remove CPUs on SMP online
• Cluster Nodes
• Add/remove cluster nodes online
• No data movement needed
• Memory
• Grow and shrink shared memory and buffer cache online
• Auto tuning of memory online
• Disk
• Add/remove disks online
• Automatically rebalance
Database
Storage
Online Patch Upgrade using RAC
Initial RAC Configuration Clients on A, Patch B
EmergencyOne-offPatches
OperatingSystemUpgrades
Upgrade Complete
HardwareUpgrades
Clients Clients
Clients on B, Patch A
Patch
1 2
34
A
B PatchB
Critical Patch Updates
Online Database Upgrades using SQL Apply
Major Release
Upgrades
Patch SetUpgrades
Cluster Software &
Hardware Upgrades
Initial SQL Apply Config
ClientsRedo
Version X Version X
1
BA
Switchover to B, upgrade A
Redo
4
Upgrade
X+1X+1
BA
Run in mixed mode to test
Redo
3
X+1X
A B
Upgrade node B to X+1
Upgrade
LogsQueue
X
2
X+1
A B
Support for Online Hot Patching
• Oracle already supports rolling patches between nodes in a cluster
• Patch a running Oracle instance with no downtime
• Many one-off patches can be online patched
• Subset of RAC online upgradeable patches
• Great for diagnostic patches
• Enable, disable and de-install one-off patches with no downtime• Integrated with OPatch and inventory
• Initially available on Linux and Solaris
• Long term goal is online patching of Critical Patch Updates (CPUs).
Rolling Database Upgrades Using Transient Logical Standby
• Logical standby allows rolling upgrades but has data type restrictions
• Temporarily convert physical standby to logical to perform a rolling upgrade
• Data type restrictions are limited to short upgrade window
• No need for separate logical standby for upgrade
Physical
Logical
Upgrade
PhysicalAlso possible in 10gR2 with more manual steps
Best Data Changes
Computer Failures
Data Failures
System Changes
Data Changes
UnplannedDowntime
PlannedDowntime Online Application Online Application
UpgradesUpgrades
Online Table & Index Redefinition
• All index changes can be done online
• Create new index, move index, defragment index
• Tables can be Reorganized & Redefined online
• Table contents are copied to a new table
• Defragments and allows changing location, table type, partitioning
• Contents can be transformed as they are copied
• Can change columns, types, sizes - specified using SQL “Select”
Source Table
Update Tracking
Transform CopyTable
TransformUpdates
GUI interface to make it
Simple
Result Table
Continuous Queries & Updates
Store Updates
Online Redefinition Enhancements
• In 10gR2 Online Redefinition supports
• Online Move of a partition of a partitioned table
• AQ tables
• Clustered Tables
• Materialized View Logs
• Abstract Data Types (Objects)
Online Application Upgrade
• New tools to enable Online Redefinition of Applications
• Application developers use these tools in their upgrade scripts
• Key Oracle Fusion Applications initiative
• Available to all Oracle ISVs and developers
How did we do it?
• Oracle Database 11g brings the Edition, the Editioning View, and the CrossEdition Trigger
• Code changes are installed in the privacy of a new edition.
• Data changes are made safely by writing only to new columns or new tables not seen by the old Edition
• An Editioning View exposes a different projection of a table into each edition to allow each to see just its own columns
• A CrossEdition trigger propagates data changes made by the old edition into the new edition’s columns, or (in hot-rollover) vice-versa
Edition-Based Redefinition
• Editions provide versioning for database code objects
• PL/SQL, triggers, views, synonyms, etc.
• Patch and upgrade scripts make changes in a New Edition that is invisible to current production users
• New Edition is published for production use after scripts are complete, and new code is tested
sql> create Edition “New_Edition” as child of “Old_Edition”;
sql> alter session set Current_Edition = “New_Edition”;
… Load new versions of non-table objects
In-Place Redefinition Improvements
• Fast add column with default value
• Does not need to update all rows to default value
• Invisible Indexes prevent premature use of newly created indexes
• Online index build with NO pause to DML
• No recompilation of dependent objects when
• Columns added to tables
• Procedures added to packages
• Easier to execute table DDL operations online
• Option to wait for active DML operations instead of aborting
Summary
• Best solution for every outage• Best Clustering – Real Application Clusters
• Best Human Error Protection - Flashback Error Correction
• Best Backup and Recovery – Recovery Manager
• Best Disaster Protection – Data Guard
• Best Uptime for System Changes – Online Patching and Upgrades
• Best Uptime for Data Changes – Redefinition and Editions
• Lowest Cost• Low cost grid servers and storage
• No idle hardware
• Bundled clusterware, volume manager, file system and DR
• Integrated, automated, and simple to use
Best Availability AND Lowest Cost