Oracle 11g New Features

The new generation of Oracle will add many exciting new features. The following describes some new features of oracle11g.
1. Database Management
 
· Database Replay)
 
This feature can capture the load of the entire data and pass it to a test database created from the backup or standby database, and then replays the role of testing the system after optimization.
 
· SQL replay (SQL replay)
 
Similar to the previous feature. However, it only captures the SQL load, rather than all the load.
 
· Plan Management)
 
This feature allows you to fix the query plan of a specific statement, regardless of the statistical data or database version changes, it will not change her query plan.
 
· Automatic Diagnostic knowledge base (Automatic Diagnostic repository ADR)
 
When Oracle detects an important error, it will automatically create an event (incident), capture information related to the event, and automatically perform database health check and notify DBA. In addition, this information can be packaged and sent to the Oracle Support Team.
 
· Incident packaging service)
 
If you need to further test or retain relevant information, this feature can package information related to an event. You can also send the packaging information to the Oracle Support Team.
 
· Feature-based Patching)
 
During patching, this feature makes it easy for you to differentiate the features in the patch package that you are using and must play. Enterprise Manager (EM) enables you to subscribe to a feature-based patch service, so Enterprise Manager can automatically scan for the features you are using with patches.
 
· Auto SQL Tuning)
 
The 10 Gb automatic optimizer can write the optimization suggestions in the SQL profile. In the 11g environment, you can enable Oracle to automatically apply the profile with three times the original performance to SQL statements. Performance Comparison is completed by a new management task in the maintenance window.
 
· Access Advisor)
 
The 11g access recommender can provide partition recommendations, including recommendations for the new interval partition (interval partitioning. The interval partition is equivalent to the automatic version of range partitioning. When necessary, she can automatically create a partition of the same size. Range partitions and interval partitions can exist in a table at the same time, and range partitions can be converted to interval partitions.
 
· Auto Memory Tuning)
 
In 9i, automatic PGA optimization is introduced; in 10g, automatic SGA optimization is introduced. By 11 GB, all memory can be automatically optimized by setting only one parameter. As long as you tell oracle How much memory is available, she can automatically specify how much memory is allocated to PGA, How much memory is allocated to SGA, and how much memory is allocated to operating system processes. You can also set the maximum and minimum thresholds.
 
· Resource Manager)
 
The 11g resource manager can manage both the CPU and IO. You can set the priority, type, and ASM disk of a specific file.
 
· ADDM
 
ADDM is introduced at 10 Gb. In 11g, ADDM not only provides suggestions for a single instance, but also provides suggestions for the entire RAC (that is, the database level. In addition, you can add some direve ve instructions to ADDM to ignore information that you don't care about.
 
· AWR Baselines)
 
The AWR baseline is extended. You can automatically create baselines for some other features used. A weekly baseline is created by default.
 
2. PLSQL Section
 
· Result Set Caching)
 
This feature greatly improves the performance of many programs. In some MIS or OLAP systems, many queries such as "select count (*)" need to be used. Previously, to improve the query performance, we may need to use materialized views or query rewriting technologies. At 11 GB, we only need to add a/* + result_cache */prompt to cache the result set, which can greatly improve the query performance. Of course, in this case, we may have to worry about another issue: integrity. In oracle, consistent reading is used to ensure data integrity. Obviously, in this new feature, to improve performance, data is read from the cache result set rather than from the rollback segment. The answer to this question is that integrity can be fully guaranteed. Because the result set is independently cached, any other DML statements during the query will not affect the content of the result set, thus ensuring data integrity.
 
· Object dependency Improvement
 
If a function or view is dependent on a table before 11 GB, whether or not the function or view-dependent attributes are involved in the Structure Change of the table, changes the function or view to invalid. In 11g, this situation is adjusted: If the changed attribute of the table is irrelevant to the related function or view, the state of the related object will not change.
 
· Regular Expression Improvement
 
A regular expression is introduced in 10 Gb. This feature greatly facilitates developers. 11 GB, Oracle again improved this feature. A function named regexp_count is added. In addition, other regular expression functions are also improved.
 
· New SQL syntax =>
 
When calling a function, we can use => to specify data for specific function parameters. In 11g, this syntax can also appear in SQL statements. For example, you can write the following statement:
Select f (x => 6) from dual;
 
· For TCP packets (utl_tcp, utl_smtp ...) Support for fgac (fine grained access control) Security Control
 
· Read-only table)
 
Previously, we used triggers or constraints to implement read-only control on tables. 11G does not need to be so troublesome, you can directly specify the table as a read-only table.
 
· Improved trigger execution efficiency

· Intra-Unit inlining)
 
In C, you can use inline or macro implementation to inline small and frequently called functions. After compilation, the part that calls the inline function is compiled into the function body of the inline function, thus improving the function efficiency. This inline function can also be implemented in plsql of 11 GB.
 
· Set trigger sequence
 
Multiple triggers may exist on a table. In 11g, you can specify the trigger sequence, without worrying about the disordered sequence resulting in data confusion.
 
· Compound trigger)
 
This is a new trigger in 11g. She allows you to have declarative, before, after each, and after processes in the same trigger.
 
· Create an invalid Trigger (Disabled Trigger)
 
In 11g, developers can create an invalid trigger and compile the trigger as needed.
 
· Use sequence in non-DML statements)
 
In earlier versions, if you want to assign the sequence value to the variable, You need to implement it using a statement similar to the following:
Select seq_x.next_val into v_x from dual;
In 11g, the following statement can be implemented without such trouble:
V_x: = seq_x.next_val;
 
· Plsql_warning
 
In 11g. You can set plsql_warning = enable all to send a warning if "when others" has no error.
 
· Inheritance of PLSQL
 
You can use the super (similar to Java) keyword in the Oracle object type to implement inheritance.
 
· Improved compilation speed
 
Because the external C compiler is not used, the Compilation speed is improved.
 
· Improved the dbms_ SQL package
 
One of the improvements is that dbms_ SQL can receive clob larger than 32 KB. In addition, it supports user-defined types and bulk operations.
 
· Added the continue keyword.
 
You can use the continue keyword in the loop Statement of PLSQL (the function is the same as the continue keyword in other advanced languages ).
 
· New PLSQL data type -- simple_integer
 
This is an integer data type that is more efficient than pls_integer.
 
3. Other parts
 
· Enhanced Compression Technology
 
Up to 2/3 of the space can be compressed.
 
· High-Speed Promotion Technology
 
This greatly improves the speed of reading data from the file system.
 
· Enhanced DATA Guard
 
You can create snapshots of the standby database for testing. Combined with the database Replay Technology, stress testing can be performed to simulate the generation of system load.
 
· Online application upgrade
 
That is, hot Patching-you do not need to restart the database when installing and upgrading or patching.
 
· Database repair recommender
 
DBA can be guided during error diagnosis and solution implementation.
 
· Logical object partitioning
 
Partitions can be performed on logical objects, and partitions can be automatically created to facilitate management of ultra-Large Databases (Very Large Databases VLDBs)
 
· New High-Performance LOB infrastructure
 
· New PHP driver