| In Oracle Database 10g Release 2, the Data Pump technology has been enhanced to provide greater manageability and flexibility with your export and import options. Following is a short summary of the new 10g Release 2 features with Oracle Data Pump: Default Data Pump directory object. Starting with 10g Release 2, you no longer have to manually create the directory object prior to using Data Pump. 10g Release 2 will create a default Data Pump directory object (DATA_PUMP_DIR) in $ORACLE_BASE/admin/DB_UNIQUE_NAME as long as ORACLE_BASE is defined during the 10g Release 2 database creation or upgrade. If for some reason ORACLE_BASE is not found, the default directory object will be created in $ORACLE_HOME/admin/DB_UNIQUE_NAME. If there is a problem with both of these locations, then Oracle will create the default directory object in $ORACLE_HOME/rdbms/log. In any case, all Data Pump users must have the appropriate permissions to the default directory object that Data Pump uses. Access to the default DATA_PUMP_DIR is granted automatically to the EXP_FULL_DATABASE and IMP_FULL_DATABASE roles. Expanded support for 10g Enterprise Manager. Starting with 10g Release 2, you can use the EM Database Control GUI to monitor all Data Pump jobs, including those created by the expdp and impdp utilities as well as the DBMS_DATAPUMP package. Within the 10g Release 2 EM, you can view the current status of the job and also change the status to EXECUTE, SUSPEND, or STOP. New export command-line parameters. Starting with 10g Release 2, you can use the COMPRESSION parameter to indicate whether the metadata should be compressed in the export dump file to reduce space consumption. The possible values are METADATA_ONLY (default) or NONE. If this parameter is used, the metadata is automatically uncompressed during the import operation. Also new with 10g Release 2 is your ability to use the SAMPLE parameter to specify a percentage of data to be sampled and unloaded from the source database. You can indicate the chosen percentage (.000001 to 100) by adding the number preceding the schema and table, such as SAMPLE="PROXY"."USERS":51. If you specify a schema, you must also specify a table, but you can specify a table without a schema because Data Pump will assume the table is part of the current user's schema. If no table is specified with the SAMPLE parameter, then the percentage will apply to the entire export operation. New export interactive-mode parameters. Starting with 10g Release 2, you can use the FILESIZE parameter to set the default file size (in bytes) for subsequent ADD_FILE commands. In 10g Release 1, the size of files added using ADD_FILE was unlimited. New import command-line parameters. Starting with 10g Release 2, you can use the ENCRYPTION_PASSWORD parameter to specify a password key for accessing encrypted column data from a dump file. ENCRYPTION_PASSWORD cannot be used in conjunction with the NETWORK_LINK parameter New DBMS_DATAPUMP API calls. Starting with 10g Release 2, you can use the WAIT_FOR_JOB API to provide specifics for completion time for Data Pump operations as well as the GET_DUMPFILE_INFO API to provide more detail on dump-file sets such as version, creation date, and character set.
|
