Top 10, no – make that 11, things about Oracle Database 11g Release 1 Thomas Kyte

<Insert Picture Here>

Top 10, no – make that 11, things about Oracle Database 11g Release 1Thomas Kytehttp://asktom.oracle.com

The Beginning...

• Data Model with Structure• Data Independent of Code• Set-oriented• 1977 the work begins

““A A Relational Model

Relational Model

forfor

Large Shared

Large Shared

Databanks”

Databanks”

E.F. Codd - 1970

First RDBMS: Version 2 June 1979

• FIRST Commercial SQL RDBMS• Impressive First SQL

• Joins, Subqueries • Outer Joins, Connect By

• A Simple Server• No transactions, ‘Limited’ Reliability

• Portability from the Start• Written in Fortran• But multi-platform – PDP11, Dec VAX

Oracle7.3 February 1996

• Partitioned Views• Bitmapped Indexes• Asynchronous read ahead for

table scans• Standby Database• Deferred transaction recovery

on instance startup• Updatable Join View• SQLDBA no longer shipped.• Index rebuilds• DBV introduced• Context Option• PL/SQL - UTL_FILE

Spatial Data Option Tablespaces changes -

Coalesce, Temporary Permanent,

Trigger compilation, debug Unlimited extents on STORAGE

clause. Some init.ora parameters

modifiable - TIMED_STATISTICS HASH Joins, Antijoins Histograms Oracle Trace Advanced Replication Object

Groups

Oracle Database Innovation

Audit Vault Database Vault Grid Computing Self Managing Database XML Database Oracle Data Guard Real Application Clusters Flashback Query Virtual Private Database Built in Java VM Partitioning Support Built in Messaging Object Relational Support Multimedia Support Data Warehousing Optimizations Parallel Operations Distributed SQL & Transaction Support Cluster and MPP Support Multi-version Read Consistency Client/Server Support Platform Portability Commercial SQL Implementation

1977 2007

30 years of sustained innovation …

… continuing with Oracle Database 11g

#1 Encrypted

Tablespaces

Encrypted Tablespaces

• Oracle Database 10g Release 2 introduced column encryption• Could not range scan• Primary/foreign key issues

• Tablespace encryption Removes those limitations

• Many encryption algorithms• 3DES168• AES128• AES192• AES256


ops$tkyte%ORA11GR1> create tablespace encrypted

2 datafile '/…/encrypted.dbf' size 10m

3 ENCRYPTION default storage( encrypt );

Tablespace created.

ops$tkyte%ORA11GR1> create tablespace clear

2 datafile '/…/clear.dbf' size 10m;

Tablespace created.


ops$tkyte%ORA11GR1> create table t

2 tablespace clear

3 as

4 select * from all_users;

Table created.

ops$tkyte%ORA11GR1> create index t_idx

2 on t(lower(username))

3 tablespace clear;

Index created.


ops$tkyte%ORA11GR1> alter system checkpoint;

System altered.

$ strings /…/clear.dbf | grep -i ops.tkyte

OPS$TKYTE from the table

ops$tkyte from the index


ops$tkyte%ORA11GR1> alter table t move

2 tablespace encrypted;

Table altered.

ops$tkyte%ORA11GR1> alter index t_idx rebuild

2 tablespace encrypted;

Index altered.


ops$tkyte%ORA11GR1> alter system checkpoint;

System altered.

$ strings /…/encrypted.dbf | grep -i ops.tkyte

[This space intentionally left blank]


ps$tkyte%ORA11GR1> set autotrace traceonly explain

ops$tkyte%ORA11GR1> select * from t where lower(username) like 'ops$%';

Execution Plan

-------------------------------------------------------------

| Id | Operation | Name | Rows | Bytes |

-------------------------------------------------------------

| 0 | SELECT STATEMENT | | 2 | 112 |

| 1 | TABLE ACCESS BY INDEX ROWID| T | 2 | 112 |

|* 2 | INDEX RANGE SCAN | T_IDX | 1 | |

-------------------------------------------------------------

Predicate Information (identified by operation id):

---------------------------------------------------

2 - access(LOWER("USERNAME") LIKE 'ops$%')

filter(LOWER("USERNAME") LIKE 'ops$%')

#2 Cache More

Stuff

Cache More

• Everyone knows the fastest way to do something is – to not do it• Client Side Cache• Server Results Cache (JIT-MV)• PL/SQL Function results cache

Cache More

ops$tkyte%ORA11GR1> /*

ops$tkyte%ORA11GR1> drop table t;

ops$tkyte%ORA11GR1> create table t as select * from all_objects;

ops$tkyte%ORA11GR1> */

ops$tkyte%ORA11GR1> update t set object_type=object_type where rownum=1;

1 row updated.

Cache More

ops$tkyte%ORA11GR1> set autotrace traceonly

ops$tkyte%ORA11GR1> select /*+ result_cache */

2 owner,

3 object_type,

4 count(*) cnt

5 from t

6 group by owner, object_type

7 order by owner, object_type

8 /

249 rows selected.

Cache More

Execution Plan

--------------------------------------------------------------------------------------------------

| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |

--------------------------------------------------------------------------------------------------

| 0 | SELECT STATEMENT | | 58742 | 1606K| 284 (2)| 00:00:04 |

| 1 | RESULT CACHE | 5cwffcum1ajfud1088m1m73f81 | | | | |

| 2 | SORT GROUP BY | | 58742 | 1606K| 284 (2)| 00:00:04 |

| 3 | TABLE ACCESS FULL| T | 58742 | 1606K| 280 (1)| 00:00:04 |

--------------------------------------------------------------------------------------------------

Result Cache Information (identified by operation id):

------------------------------------------------------

1 - column-count=3; dependencies=(OPS$TKYTE.T); parameters=(nls); name="select /*+ result_cache */

owner,

object_type,

count(*) cnt

from t

group by owner, object_type

order by"

Cache More

Statistics

----------------------------------------------------------

0 recursive calls

0 db block gets

1005 consistent gets

0 physical reads

0 redo size

6922 bytes sent via SQL*Net to client

596 bytes received via SQL*Net from client

18 SQL*Net roundtrips to/from client

1 sorts (memory)

0 sorts (disk)

249 rows processed

Cache Moreops$tkyte%ORA11GR1> /

249 rows selected.

Statistics

----------------------------------------------------------

0 recursive calls

0 db block gets

0 consistent gets

0 physical reads

0 redo size




0 sorts (memory)

0 sorts (disk)

249 rows processed

Cache More

ops$tkyte%ORA11GR1> select * from (

2 select /*+ result_cache */

3 owner, object_type, count(*) cnt

4 from t

5 group by owner, object_type

6 order by owner, object_type

7 )

8 where cnt > 100

9 /

38 rows selected.

Cache More

Statistics

----------------------------------------------------------

0 recursive calls

0 db block gets

0 consistent gets

0 physical reads

0 redo size




0 sorts (memory)

0 sorts (disk)

38 rows processed

Cache More

SQL> create or replace

2 function not_cached

3 ( p_owner in varchar2 )

4 return number

5 as

6 l_cnt number;

7 begin

8 select count(*)

9 into l_cnt

10 from t

11 where owner = p_owner;

12 dbms_lock.sleep(1);

13 return l_cnt;

14 end;

15 /

Function created.

SQL> create or replace

2 function cached

3 ( p_owner in varchar2 )

4 return number

5 result_cache

6 relies_on(T)

7 as

8 l_cnt number;

9 begin

10 select count(*)

11 into l_cnt

12 from t

13 where owner = p_owner;

14 dbms_lock.sleep(1);

15 return l_cnt;

16 end;

17 /

Function created.

Cache More

SQL> exec dbms_output.put_line( not_cached( 'SCOTT' ) );

6

PL/SQL procedure successfully completed.

Elapsed: 00:00:01.93


6


Elapsed: 00:00:01.29


6


Elapsed: 00:00:01.07

Cache More

SQL> exec dbms_output.put_line( cached( 'SCOTT' ) );

6


Elapsed: 00:00:01.09


6


Elapsed: 00:00:00.01


6


Elapsed: 00:00:00.01

Cache More

SQL> update t set owner = initcap(owner) where rownum = 1;

1 row updated.

SQL> commit;

Commit complete.


6


Elapsed: 00:00:01.25


6


Elapsed: 00:00:00.01

Cache More

SQL> exec dbms_output.put_line( cached( 'SYS' ) );

29339


Elapsed: 00:00:01.21

SQL> exec dbms_output.put_line( cached( 'SYS' ) );

29339


Elapsed: 00:00:00.01


6


Elapsed: 00:00:00.00

#3 Standby

Just got better

Standby Database

• Logical Standby was…• Limited in type support• But was always open for business

• Physical Standby was…• Easy• But considered “not useful day to day”

Standby Database

• Logical Standby has…• XMLType support• DBMS_RLS & DBMS_FGA support• TDE support

Active Data Guard: Develop & Test on Standby DB

• Use physical standby databasefor development & testing• Preserves zero data loss in test/dev mode

• Flashback DB to back-outchanges & use as standby

ProductionDatabase

Standby Database

Developers,Testers

Eliminates cost of idle DR system

Active Data Guard: Report & Backup from Standby DB

• Offload reporting to standby• Simultaneously available for recovery

• Offload backups to standby• Complete database and fast incremental

backups

ProductionDatabase

Standby Database

Reporting

Backups

Improves performance on production database

Active Data Guard – More than a Standby

Simultaneous readand recovery

High ROI

Automated

Disaster and performance protection

Use daily in testingand production

Recoverymode only

Low ROI

Manual intensive

Disasterprotection only

Used in disaster only

#4 Real

Application

Testing

Real Application Testing –Database Replay

• Recreate actual production database workload

• Capture production workload incl. concurrency

• Replay workload in test with production timing

• Analyze & fix issues before production

MiddleTier

Capture DB Workload

Storage

OracleDB servers Replay DB

Workload

ProductionEnvironment

Test (RAC)Environment`

#5 Smaller

more secure

DMP files

Datapump

• COMPRESSION• ALL, DATA_ONLY, METADATA_ONLY, NONE

$ expdp / directory=tmp dumpfile=uncompressed.dmp compression=NONE schemas=ops\$tkyte

Export: Release 11.1.0.6.0 - Production on Friday, 21 September, 2007 12:23:26

. . exported "OPS$TKYTE"."BIG_TABLE" 24.57 MB 250000 rows

. . exported "OPS$TKYTE"."T" 6.791 MB 67945 rows

$ expdp / directory=tmp dumpfile=compressed.dmp compression=ALL schemas=ops\$tkyte

Export: Release 11.1.0.6.0 - Production on Friday, 21 September, 2007 12:23:58

. . exported "OPS$TKYTE"."BIG_TABLE" 3.110 MB 250000 rows

. . exported "OPS$TKYTE"."T" 762.1 KB 67945 rows

$ ls -l /tmp/*compressed.dmp

-rw-r----- 1 ora11gr1 ora11gr1 4124672 Sep 21 12:24 /tmp/compressed.dmp

-rw-r----- 1 ora11gr1 ora11gr1 33136640 Sep 21 12:23 /tmp/uncompressed.dmp

Datapump

• ENCRYPTION• All• Data_only• Metadata_only• None• Encrypted_columns_only

• PARTITION_OPTIONS• None• Departition• Merge

• REUSE_DUMPFILES

• Ability to use DML error logging features • DATA_OPTIONS parameter

#6 Virtual

Columns

Virtual Columns

• Create Table

• Alter Table Add Column

• Are ‘column expressions’• Expressions involving other columns in table• Constants• Deterministic functions

• Ease of use and Optimizer enhancement

Virtual Columns

ops$tkyte%ORA11GR1> /*

ops$tkyte%ORA11GR1> create table emp

ops$tkyte%ORA11GR1> as

ops$tkyte%ORA11GR1> select all_objects.*, object_id sal,

round(dbms_random.value( 1000, 100000 )) comm

ops$tkyte%ORA11GR1> from all_objects

ops$tkyte%ORA11GR1> /

ops$tkyte%ORA11GR1>

ops$tkyte%ORA11GR1> */

ops$tkyte%ORA11GR1> exec dbms_stats.gather_table_stats( user, 'EMP' )


Virtual Columns

ops$tkyte%ORA11GR1> select avg( sal+comm ) avg_comp, avg(sal) avg_sal, avg(comm) avg_comm from emp;

AVG_COMP AVG_SAL AVG_COMM

---------- ---------- ----------

85376.9437 34821.6827 50555.261

ops$tkyte%ORA11GR1> select

count(case when sal+comm > 85376.9437 then 1 end) above_comp,

2 count(case when sal > 34821.6827 then 1 end) above_sal ,

3 count(case when comm > 50555.261 then 1 end) above_comm

4 from emp;

ABOVE_COMP ABOVE_SAL ABOVE_COMM

---------- ---------- ----------

33957 33830 34036

Virtual Columns

ops$tkyte%ORA11GR1> select * from emp where sal > 34821.6827;

--------------------------------------------------------------------

| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time

--------------------------------------------------------------------

| 0 | SELECT STATEMENT | | 34673 | 3893K| 309 (1)| 00:0

|* 1 | TABLE ACCESS FULL| EMP | 34673 | 3893K| 309 (1)| 00:0

--------------------------------------------------------------------


---------- ---------- ----------

33957 33830 34036

Virtual Columns

ops$tkyte%ORA11GR1> select * from emp where comm > 50555.261;

--------------------------------------------------------------------


--------------------------------------------------------------------



--------------------------------------------------------------------


---------- ---------- ----------

33957 33830 34036

Virtual Columns

ops$tkyte%ORA11GR1> select * from emp where sal+comm > 85376.9437;

--------------------------------------------------------------------


--------------------------------------------------------------------



--------------------------------------------------------------------


---------- ---------- ----------

33957 33830 34036

Virtual Columns

ops$tkyte%ORA11GR1> ALTER TABLE emp ADD (comp AS (sal+comm));

Table altered.

ops$tkyte%ORA11GR1> exec dbms_stats.gather_table_stats( user, 'EMP', method_opt=> 'for columns comp' );


Virtual Columns

ops$tkyte%ORA11GR1> select * from emp where sal+comm > 85376.9437;

--------------------------------------------------------------------


--------------------------------------------------------------------



--------------------------------------------------------------------


---------- ---------- ----------

33957 33830 34036

#7 Partitioning

just got better

Partitioning

• Composite Completely

• Virtual Column Partitioning

• Partition by Reference

• Interval Partitioning

Range List Hash

Range 9i 8i

List

• Partition (or index) on virtual (computed) columns • New composite partitioning

Range List Hash

Range 11g 9i 8i

List 11g 11g 11g

JANFEB

>5000

1000-

5000

ORDERS

RANGE-RANGEOrder Date by

Order Value

USA EUROPE

>5000

1000-

5000

ORDERS

LIST-RANGERegion by

Order Value

USA EUROPE

Gold

SilverORDERS

LIST-LISTRegion by

Customer Type

Enhanced Partitioning

Partitioningby REFERENCE

Table ORDERS

Jan 2006

... ...

Feb 2006

Table LINEITEMS

Jan 2006

... ...

Feb 2006

• RANGE(order_date)• Primary key order_id

• RANGE(order_date)• Foreign key order_id

• Partitioning key inherited through PK-FK relationship

• Avoids redundant storage, maintenance of order_date

New “INTERVAL” partitioning

• Automatically creates a new partition when data outside the existing range is first inserted • E.g., monthly partitions, automatic

new partition first day of the month

• Composite partitioning: interval, interval-list, interval-hash, and interval-range

• Automates partition management

ORDERS

Jul Aug Sep

Sep 1 2007

PartitioningAutomation

Partitioning

ops$tkyte%ORA11GR1> create table audit_trail

2 ( ts timestamp,

3 data varchar2(30)

4 )

5 partition by range(ts)

6 interval (numtodsinterval(1,'day'))

7 store in (users, example )

8 (partition p0 values less than

9 (to_date('22-sep-2007','dd-mon-yyyy'))

10 )

11 /

Table created.

Partitioning

ops$tkyte%ORA11GR1> select partition_name, tablespace_name, high_value

2 from user_tab_partitions

3 where table_name = 'AUDIT_TRAIL';

PARTITION_ TABLESPACE HIGH_VALUE

---------- ---------- ----------------------------------

P0 USERS TIMESTAMP' 2007-09-22 00:00:00'

Partitioning

ops$tkyte%ORA11GR1> insert into audit_trail

2 select sysdate+rownum, 'x'

3 from all_users

4 where rownum <= 5

5 /

5 rows created.

Partitioning

ops$tkyte%ORA11GR1> select partition_name, tablespace_name, high_value

2 from user_tab_partitions

3 where table_name = 'AUDIT_TRAIL';

PARTITION_ TABLESPACE HIGH_VALUE

---------- ---------- ----------------------------------

P0 USERS TIMESTAMP' 2007-09-22 00:00:00'

SYS_P66 EXAMPLE TIMESTAMP' 2007-09-23 00:00:00'

SYS_P67 USERS TIMESTAMP' 2007-09-24 00:00:00'


SYS_P69 USERS TIMESTAMP' 2007-09-26 00:00:00'


6 rows selected.

#8 The long

awaited pivot

Pivot

ops$tkyte%ORA11GR1> select deptno,

2 sum( decode( job, 'CLERK', sal ) ) clerk,

3 sum( decode( job, 'SALESMAN', sal ) ) salesman,

4 sum( decode( job, 'MANAGER', sal ) ) manager,

5 sum( decode( job, 'ANALYST', sal ) ) analyst,

6 sum( decode( job, 'PRESIDENT', sal ) ) president

7 from emp

8 group by deptno

9 order by deptno

10 /

DEPTNO CLERK SALESMAN MANAGER ANALYST PRESIDENT

---------- ---------- ---------- ---------- ---------- ----------

10 1300 2450 5000

20 1900 2975 6000

30 950 5600 2850

Pivot

ops$tkyte%ORA11GR1> select *

2 from (select deptno, job, sal

3 from emp ) e

4 pivot( sum(sal) for job in

5 ( 'CLERK', 'SALESMAN', 'MANAGER',

6 'ANALYST', 'PRESIDENT' ) )

7 order by deptno

8 /

DEPTNO 'CLERK' 'SALESMAN' 'MANAGER' 'ANALYST' 'PRESIDENT'

---------- ---------- ---------- ---------- ---------- -----------

10 1300 2450 5000

20 1900 2975 6000

30 950 5600 2850

#9 Flashback

Data Archive

Flashback Data ArchiveTotal Data Recall

• Tamper-proof data archive• Efficient storage and

retrieval of undo • Keep data for months,

years, decades!• Fast access to even very

old data• View data, versions of

rows as of any time

• Control data retention time, purging of data

User Tablespaces

Flashback Data Archive

Oracle 11g Database

Select * from orders AS OF ‘Midnight 31-Dec-2004’

Changes ArchiveArchiveTablesTables

#10 Finer

Grained

Dependency

Tracking

Finer Grained Dependency Tracking

• Fewer Invalidations• Add to a specification – so what• Add/Drop/Modify a column – so what• Holds true with view modifications too• Change a synonym pointer – so what• Replace a procedure – so what


ops$tkyte%ORA10GR2> create or replace package p1

2 as

3 function f return number;

4 end;

5 /

Package created.


2 as

3 g_global number := p1.f;

4 end;

5 /

Package created.


ops$tkyte%ORA10GR2> select object_name, status

2 from user_objects

3 where object_name like 'P_';

OBJECT_NAME STATUS

------------------------------ -------

P1 VALID

P2 VALID



2 as


4 procedure p;

5 end;

6 /

Package created.


2 from user_objects


OBJECT_NAME STATUS

------------------------------ -------

P1 VALID

P2 INVALID



2 as


4 procedure p;

5 end;

6 /

Package created.


2 from user_objects


OBJECT_NAME STATUS

------------------------------ -------

P2 VALID

P1 VALID


ops$tkyte%ORA10GR2> create table t ( x int, y int );

Table created.

ops$tkyte%ORA10GR2> create or replace procedure p

2 as

3 begin

4 for x in ( select x, y from t )

5 loop

6 null;

7 end loop;

8 end;

9 /

Procedure created.

ops$tkyte%ORA10GR2> select status from user_objects where object_name = 'P';

STATUS

-------

VALID


ops$tkyte%ORA10GR2> alter table t add z int;

Table altered.


STATUS

-------

INVALID


STATUS

-------

VALID


ops$tkyte%ORA10GR2> alter procedure p compile;

Procedure altered.

ops$tkyte%ORA10GR2> alter table t drop column z;

Table altered.


STATUS

-------

INVALID


STATUS

-------

VALID

#11 OLTP Table

Compression

• Oracle 9i compresses data only during bulk load; useful for DW and ILM

• Oracle 11g compresses w/ inserts, updates

• Trade some cpu for disk & i/o efficiency

• Compress large application tables

• Transaction processing, data warehousing

• Compress all data types: structured, unstructured

• Savings cascade to all db copies: test, dev, standby, mirrors, archiving, backup, etc.

Oracle AdvancedCompression

<Insert Picture Here>

AQ&

Documents

Top 10, no – make that 11, things about Oracle Database 11g Release 1 Thomas Kyte