Input sources for apply requests
Related topics
This topic describes the SQL and logical log input sources that High-speed Apply Engine uses.
Input sources provide the database changes to be applied. High-speed Apply Engine supports multiple input types, including SQL statements and logical log files created by BMC AMI Log Master for Db2. These input sources allow the product to apply large volumes of changes generated by change capture, migration, or recovery processes.
For more information about specifying the input sources in a configuration or an apply request, see Defining configurations or Apply requests.
SQL statements
You can specify an input file that contains SQL statements. The SQL can be generated by BMC AMI Log Master for Db2 or other standard SQL generators.
SQL input can include data manipulation language (DML) or data definition language (DDL) statements. High-speed Apply Engine can process any DML or DDL statements. To provide restart capability, SQL input must not contain more than one SQL statement on the same line.
The BMC AMI Log Master for Db2 product generates a separate template file that High-speed Apply Engine can use to efficiently process SQL input as static SQL against Db2. The template file contains one entry for each distinct statement type in the SQL input. If an apply request includes a template file, High-speed Apply Engine processes the template file before it begins processing SQL input.
Logical log
High-speed Apply Engine can read logical log input produced by either BMC AMI Log Master for Db2 or an application program that conforms to the published format. BMC AMI Log Master for Db2 produces logical logs that let you specify the types of operations that you need to perform. The logical log files provide information that High-speed Apply Engine can use to update the target tables.
Logical log input consists of the following files:
- Logical log control file—Contains information about the format and content of the logical log. Specify the control file as the input for High-speed Apply Engine.
- Logical log data file—Contains actual change information (updates, deletes, inserts, or DDL objects) in logical log format.
- (Optional) Large volume VSAM files—Contain data from Db2 columns that hold large volumes of data (such as XML columns or LOB columns). To apply changes to large volume columns, the appropriate VSAM files must be available to High-speed Apply Engine along with the control file and the data file.
- (Optional) Logical log XMLSTRING control file—Contains the string IDs and string data that Db2 uses to encode the data in XML columns. If the logical log includes XML data and the target Db2 subsystem is different from the subsystem where the logical log was generated, High-speed Apply Engine needs the string IDs and data to serialize your XML data correctly.
High-speed Apply Engine can often process a logical log file faster than an SQL file that contains the same database changes. This improvement occurs because logical log input is already in host variable format. High-speed Apply Engine can also avoid parsing SQL syntax.
You can create logical logs to perform data migration, UNDO SQL functions, or REDO SQL functions.
For more information about logical logs, see Logical log files.
Logical log input considerations
Be aware of the following points about logical log input:
- High-speed Apply Engine reads logical log input only for apply requests for Db2 on mainframe targets.
- Logical log input consists of the following files:
- The logical log control file contains information about the format and content of the logical log.
- The logical log data file contains actual change information (updates, deletes, inserts, or DDL objects) in logical log format.
- (Optional) The large volume VSAM files contain data from Db2 columns that hold large volumes of data (such as XML columns or LOB columns). The product or program that generates the logical log must generate these files to enable High-speed Apply Engine to apply data to large volume columns.
- (Optional) The logical log XMLSTRING control file contains the string IDs and string data that Db2 uses to encode the data in XML columns. High-speed Apply Engine needs this file when the logical log includes XML data and the Db2 subsystem of the target tables is different than the subsystem where the logical log was generated. The product or program that generates the logical log must obtain this information from the Db2 catalog table SYSIBM.SYSXMLSTRINGS. To apply the changes to a different target subsystem, High-speed Apply Engine uses the string IDs and data to serialize your XML data correctly.
- The published format for logical log input does not define an order for the records in either the logical log control file or the logical log data file.
- High-speed Apply Engine does not use all of the records in logical log input. It reads the following records in either the logical log control file or the logical log data file:
- CLUS LOB/XML Cluster Data Set Record
- DCCL Clustering Index Column Description Record
- DCDX Clustering Index Description Record (deprecated)
- DCL2 DB2 Column Description Record Part Two
- DCLI DB2 Column Information Record
- DCOL DB2 Column Description Record (deprecated)
- DCXI Clustering Index Information Record
- DDSM Data Sharing Member Record
- DLCI Logical Log Column Information Record
- DLCO Logical Log Column Description Record (deprecated)
- DLDS Logical Log Data Set Record
- DLGE Log Information End Record
- DLGI Log Information Record
- DLOB LOB Column Description Record
- DSS1 SysStrings Base Record
- DSS2 SysStrings TransTable Record
- DTBC Table Record Count Record
- DTBI Table Information Record
- DTBL Table Description Record (deprecated)
- DTBR Table RBA Record (deprecated)
- DTS2 Table Space Description Record Part Two (deprecated)
- DTSI Table Space Information Record
- DTSP Table Space Description Record (deprecated)
- DXML XML Column Description record
- DXSF XMLSTRING Data Set Name Record
- LLDF Data Change Record (in logical log data file)
- XPTH Current Path Record
- XTYP Type Record