New Features

This section lists and briefly describes newly added features.

DBMS Engine

SQL Plan Management

– Prevents performance degradation after an execution plan change.

JSON type

– Stores and processes JSON data.

Fine-grained Auditing

Audits column- or row-level data that meets specific conditions in a table or view. The DBMS_FGA package supports this feature.

Flashback Database

– Rewinds the entire database to a specific point in time, without a backup/recovery operation.

– Saves a significant amount of time taken normally for retrieving and recovering backup files to return to a specific point in time.

TSC(Tibero Standby Cluster)

• Snapshot Standby

– Enables independent executions of DDL/DML statements from a certain point of time, while continuing to receive redo logs from a primary node to synchronize with it. When converting back to the primary node, however, all DDL/DML statements that have been executed will be rolled back.

• Multi-Node Standby (Multi-Node TSC)

– Ensures high availability between the primary and standby nodes. In read-only mode, queries are distributed across the nodes and this results in a greater query performance.

• Cascade Standby (Cascade TSC)

– Transfers redo logs from a standby to another standby, which reduces the network load on the primary node.

AWS QuickStarts

– Ensures observer availability.

Splitting Encryption Keys with HSM Devices

– Stores wallet-managed keys to a Hardware Security Module (HSM).

• The following HSM devices are supported.

D’Amo KMS (Penta Security)

Vormetric Data Security Manager (Thales)

Measuring and Recording Data for Cloud-Based Tibero Pricing

– Records data for charging cloud license fees.

GOST Encryption Algorithm

GOST as a Tibero encryption algorithm is added.

OLTP Compression

– Compresses tables where occur executions of typical DMLs instead of direct path Insert/Load, in an OLTP environment.

Index Prefix Compression

– Effectively manages the space used by indexes by eliminating duplicate key columns within an index leaf block. You can use this feature when creating or re-creating an index by adding the COMPRESS N clause.

Geometry KNN Computation

– Supports KNN computation.

TAS disk resizing

– Expands the size of an existing disk used in TAS.

TAS disk repair timer

– Automatically drops a failed disk in TAS after a certain period of time.

tbascmd and DDL for Checking TAS Metadata Integrity

A tbascmd command and DDL to check the integrity of metadata about files and disks managed in TAS are added.

Prioritizing Log Flush in SSVR

– Gives preference to log flush over any other jobs in SSVR.

GIS Geography Spatial Coordinates

– Added different spatial coordinate systems and computations that include these coordinates.

New GIS Geometry Features

– Added different types of spatial data including 2D and 3D as well as computations based on these types of data.

Label Security

– Provides row-level data access controls based on associated labels for individual table rows and users.

In-Memory Column Store

– Stores data in memory using a columnar format to optimize data scans on specific columns.

Utilities

Recovery Manager Tool(RMGR)

The following utilities have been added to RMGR.

• It provides compatibility for NetBackup, a backup/recovery solution by Veritas.

• It supports online backup on a standby node, and thereby recovery on a primary node.

The following options have been added or modified.

gateway for oracle 19c

– Added Gateway for Oracle 19c.

ESRI shapefile Conversion

– Added tibero2shp, which converts tables into the shapefile format.

Updated Features

This section lists and briefly describes modified and improved key features.

DBMS Engine

Default Character Set

– Changed the default character set of Tibero to UTF8 from MSWIN949.

DBMS_SCHEDULER package

– Improved the overall function of the DBMS_SCHEDULER package and added some of its detail features that were unsupported so far.

TAC Recovery Parallel Reconfiguration

– Redistributes internal resources that are individually managed by each instance for clustering, when a Tibero Active Cluster (TAC) instance goes down or a new instance is added to the active cluster.

– Modified so that multiple threads redistribute resources in parallel, to reduce the overall execution time. Previously, the resource redistribution operation was performed by a single thread.

New Features

This section lists and briefly describes newly added features.

DBMS Engine

TAC(Tibero Active Cluster)

• Online Undo tablespace replacement

Undo Tablespace can be replaced during online on TAC environment.

TSC(Tibero Standby Cluster)

• Non-stop standby deployment

– Added a function to build a standby DB using only archive logfiles without stopping the primary DB and backing up online redo logfiles.

• Standby forced reverse synchronization

The existing primary can clear the logs with differences without rebuilding, receive the data block image accessed by the log from the primary (existing standby), force sync, and become a standby.

• Online Undo tablespace replacement

Undo Tablespace can be replaced during online on TSC environment.

• Restrict primary connection IP from standby

– Restricts connection from Standby to only specific IPs of the Primary.

Statistical information collection

• [ DBA | ALL | USER ]_TAB_MODIFICATIONS

– Added view to check the amount of change in ROWS that has been DML proceeded to the table since the last statistical information collection.

• [ DBA | ALL | USER ]_TAB_STATISTICS

– Added a view that provides statistical information about the table.

• GRANULARITY

– Added GRANULARITY to DATABASE, DICTIONARY, SCHEMA, INDEX statistical information collection.

– Added TABLE statistics collection only on specified SubPartitions.

• Statistical information collection performance

– Collects statistics information using TEMPORARY TABLE.

System Package

• DBMS_PARALLEL_EXECUTE

– Added a package to provide the ability to update tables in parallel.

• DBMS_XMLDOM

– Supports DBMS_XMLDOM package on Solaris 5.11.

TAS

– Added TAS begin/end backup to ensure the consistency of the TAS metafile.

Frame

– Added a feature to clean up the session if no request from client for a certain period of time in a session where an active transaction exists.

Utilities

tbImport

– Added option to set INDEX PARALLEL DEGREE when importing.

Updated Features

DBMS Engine

Data Definition Language (DDL)

• – Improved Parallel INDEX build performance.

• – Improved DD LOCK contention between sessions.

• – Improved performance when using bulk SEQUENCE.

• – Improved usability by returning space in UNUSABLE INDEX.

• – Added the UPDATE GLOBAL INDEXES.

Optimizer

• – Activated INDEX SKIP SCAN when using the LIKE conditional statement.

• – Improved JOIN plan performance.

Executor

• – Improved BLOCK SAMPLING performance during TABLE FULL SCAN.

• – Improved Performance by performing a TABLE FULL SCAN of a size larger than a certain HRESHOLD using PGA.

CONTEXT INDEX

• – Improved CONTEXT INDEX creation performance.

• – Improved Search performance using CONTEXT INDEX.

Buffer Cache performance

• – Improved Buffer cache bucket contention.

TAC(Tibero Active Cluster)

• – Improved resource contention in the TAC environment.

• – Improved FULL SCAN performance in the TAC environment.

• – Stabilized the message processing logic between instances In the TAC environment.

TSC(Tibero Standby Cluster)

• – Added ASYNC LNW to transmit logs by referring to LOG_ARCHIVE_DEST.

TAS

• – Improved TAS to boot immediately without waiting for resync to complete If a new TAC Instance is booted while TAS is performing resync.

• – skipped rebalance on necessary When adding a DISK in a TASK environment consisting of external redundancy disk space.

• – Improved feature and stability of TBASCMD.

Controlfile

• – Improved a stability of controlfile related I/O logic.

TPR

• – Added SQL ID item output to the query unit information in the TPR report.

Label Security

• Renamed the Label Security system user from SYS to LB

Added or improved packages and procedures

• SA_POLICY_ADMIN

• SA_USER_ADMIN

• SA_AUDIT_ADMIN

Utilities

tbSQL

• – Expanded a range of SET INTERVAL support from natural numbers to real numbers with 3 decimal places.

• – Added a feature to specify the number of loops adding COUNT to LOOP.

Overview

This guide introduces new features and changes implemented to Tibero 7.

Release History

New Features

This section lists and briefly describes newly added features.

DBMS Engine

JSON Type

– Supports the JSON_MERGEPATCH and JSON_OBJECTAGG functions.

– Supports Function-based indexes for JSON_VALUE. (Options for JSON_VALUE must match.)

– Supports declaration of JSON type variables and an IN/OUT parameter in PSM.

SQL PROFILE

It allows SQL tuning when the user cannot modify an SQL query within an application.

SQL PLAN BASELINE

– Stores SQL plans to be used by the user without choosing the Optimizer for optimized plans.

Patch and rollback tasks when TAS meta files are added or removed

– Added DDLs to support a patch implementation that requires adding or removing TAS meta files, without reconfiguring TAS.

Tibero Recovery Catalog

It supports metadata management for multiple databases.

• Tibero Recovery Catalog stores metadata of different databases to physically separated locations. The database storing metadata is referred to as Catalog database (Catalog).

• When a database is registered in the catalog, tbrmgr scans the database's control file and extracts metadata, which is sent and stored to the catalog.

• It keeps data up-to-date.

• It updates metadata when backups are created or deleted using the rmgr client, or when the administrator requests resynchronization.

Copying files based on the redundancy number with ASCMD commands

– Added an option for the cp and cptolocal commands to specify a mirror of a target file through ascmd.

– The -redun option (set to a value between 0 and 2) copies a file according to the redundancy number.

MVIEW

– Added new MVIEW features.

• [DBA|ALL|USER]_MVIEW_DETAIL_RELATIONS : displays MVIEW-related objects information.

• [DBA|ALL|USER]_MVIEW_REFRESH_TIMES : displays refresh time of MVIEW.

CONTEXT INDEX

– Added views for CONTEXT INDEX.

• CTX_PREFERENCES

• CTX_PREFERENCE_VALUES

• CTX_STOPLISTS

Utilities

SET MARKUP for tbSQL

– Displays output of tbsql in html.

Client Drivers

JDBC 4.1 Standards

• It supports java.math.BigInteger and JDBC type BIGINT.

• It supports disconnection using the connection.abort function.

• It supports specifying schema using the connection.setSchema function.

• It supports specifying a timeout using the connection.setNetworkTimeout function.

• It supports closing statements upon the end of result sets using the statement.closeOnCompletion function.

• It supports specifying return types using a class as the argument of the Resultset.getObject function.

JDBC 4.2 Standards

• It supports java.time.LocalDate, LocalTime, LocalDateTime, OffsetTime, and OffsetDateTime types.

• It supports specifying types using SQLType as the argument of the setObject and registerOutParameter functions.

• It supports returning the affected row count value in the long data type using the Statement.executeLargeUpdate function.

Updated Features

This section lists and briefly describes modified and improved key features.

DBMS Engine

DBMS_VERIFY

It supports DBMS_VERIFY for local partitioned indexes.

SYSTEM VIEW

– Added TYPE column for the following views.

• [DBA|ALL|USER]_TAB_PRIVS

• [DBA|ALL|USER]_TBL_PRIVS

• [ALL|USER]_TAB_PRIVS_MADE

DBMS_UTILITY

– Added the MAXNAME_ARRAY type in the DBMS_UTILITY package.

DBMS_XMLGEN

– Added the SETBINDVALUE procedure in the DBMS_XMLGEN package.

Client Drivers

Default Character Set

– Changed the default character set for client environments from MSWIN949 to UTF8. (Set the TB_NLS_LANG environment variable.)

New Features

This section lists and briefly describes newly added features.

DBMS Engine

Database Replay

• Added the program to validate the effects on applications when updating or patching Tibero DBMS.

• This program captures the workload within the database at the message level to perform the same operations.

• Database Replay captures the database itself and replays its operation to assess effects without downtime. This process simulates database operations precisely, allowing an accurate reflection of the operational workload.

• This program is particularly effective for environments where procedures and functions are heavily used.

• DBreplay is performed in the process of the following steps: Database workload capture, Preprocess, Replay, and Result (optional).

Data Definition Language (DDL)

• Automates statistics collection for index creation Automatically collects index statistics when executing CREATE INDEX.

• Added the following ALTER TYPE statements

  • ALTER TYPE ADD METHOD

  • ALTER TYPE DROP METHOD

  • ALTER TYPE RENAME ATTRIBUTE

Utilities

gateway for oracle

– Added a memory allocator dump feature of the gateway used to link Tibero to Oracle database.

Updated Features

DBMS Engine

Data Definition Language (DDL)

• Improved the performance of the tablespace creation statement when the number of tablespaces is high.

• Improved the MOVE operation for LOB-type columns.

Optimizer

• Improved the DISPLAY_CURSOR function in the DBMS_XPLAN package.

Executor

• Executes DISTINCT AGGREGATION with less memory and at a faster speed.

• Reduces memory usage when specifying a partition or subpartition of a partitioned table in a query.

PSM

• Increases the maximum size of CHAR and VARCHAR data types used in PSM from 32767 to 65532.

• Improved the RESULT_CACHE function in PSM.

  • Allows sequence in the RESULT_CACHE function.

  • Displays an error when using REF CURSOR in the RESULT_CACHE function.

  • Displays an error when using RESULT_CACHE in the nested function.

  • Displays an error when using RESULT_CACHE in a function in anonymous block.

Recovery

• Modified to allow adding datafiles to a tablespace in hot backup status through BEGIN BACKUP.

Cache

• Enables V$THRINFO to display buffer handle usage by session and across the entire database instance in real time.

Stability

• Improved the stability of the logic related to CCC in a TAC environment.

Parameter

Modified Parameters

New Parameters

Known Issue

This section briefly describes known issues of Tibero 7.2.2.

Missing statistics of a partition with no data when creating a partitioned index

• Issue When creating a partitioned index in parallel, the collected statistics do not include the block count value for the index segment of no-data partition, and '0' is entered instead.

• Action Recollect the index statistics by using EXEC DBMS_STATS.GATHER_INDEX_STATS.

Missing subpartition processing in the EXPORT/IMPORT_STATS function of the DBMS_STATS package

• Issue When migrating statistics, subpartitions are not processed for the EXPORT_TABLE/INDEX_STATS and IMPORT_TABLE/INDEX_STATS functions.

• Action None: Plans to provide a patch.

Error with the LAST_ELAPSED_TIME value in the DISPLAY_CURSOR function of the DBMS_XPLAN package

• Issue When displaying LAST_ELAPSED_TIME in the DISPLAY_CURSOR function of the DBMS_XPLAN package, the subquery execution time is double-counted.

• Action Refer to AVG_ELAPSED_TIME instead of LAST_ELAPSED_TIME, taking care to note that the two have different meanings.

  Refer to TOT ELAPSED TIME after the query is hard-parsed and executed once.

Error with Wait Time items within Wait Events in TPR

• Issue Some Wait Time items, including TPR Wait Events by Wait Time, show values exceeding 100%.

• Action None: Plans to provide a patch.

Error caused by multi update on tables with LONG and LOB columns

• Issue Performing multiple updates on tables containing LONG and LOB columns causes an error due to missing row reset.

• Action Avoid using multiple operations. (by setting _ENABLE_TDD_MU=N and _TDD_MU_ENABLE_ALL_COL=N)

Error caused when enabling standby REDO thread

• Issue Enabling the standby REDO thread causes TBR-2400 error if the number of standby log files exceeds that of online log files.

• Action Add the same number of standby log files as online log files.