New Features › CA Datacom/DB Enhancements › DBUTLTY

DBUTLTY

The new DBUTLTY VERINDEX function ensures the integrity of the CA Datacom/DB data rows and their corresponding keys in the index. This function is designed to use when you question the integrity of the index with the data, but the table cannot be taken offline. VERINDEX executes while the table is open for processing.

Note: For more information, see the CA Datacom/DB DBUTLTY Reference Guide.

DBUTLTY Functionality Provided for Online Reorganization Capabilities

The following are new DBUTLTY functions have been provided as part of the new online reorganization process. The online reorganization process is covered within the larger data reorganization discussion in the CA Datacom/DB Data Reorganization Green Book which can be found at:

www.ca.com/greenbooks

DATANE Native Efficiency Report

The main function of the Native Efficiency Report (NER) is to build a simple report that shows the sequential access efficiency for a given data table. You can use this report to analyze the sequential effectiveness (native key order) of a data table and whether data reorganization is needed.

The NER requires the MUF to be up and the selected data table to be available for processing. The NER uses highly efficient read-only index commands to limit the effect on other processing in the MUF.

For detailed instructions about executing the DBUTLTY DATANE function, see Appendix A: Implementing and Using DBUTLTY REPORT=DATANE in the CA Datacom/DB Data Reorganization Green Book.

DATASP Data Area Space Report

The main function of the Data Area Space Report (DATASP) is to build a simple report that shows the detailed space usage in a selected data area. This space usage provides insight into the number of rows per block, bytes per block, compression efficiency and so on. The DATASP report provides a simple data area map that can help you determine:

• available space in a data area
• amount of compression
• an estimate of rows that can be added (based on calculated average row size and available space)

The DATASP runs outside of the MUF and does not require that the MUF availabity. The DATASP report uses a highly efficient I/O process that has no effect on the MUF processing.

For detailed instructions about executing the DBUTLTY DATASP function, see Appendix B: Implementing and Using DBUTLTY REPORT=DATASP in the CA Datacom/DB Data Reorganization Green Book.

OLREORG Online Data Table Reorganization

The online reorganization (OLREORG) function was introduced as part of DBUTLTY in CA Datacom/DB. This function uses a patented process to reorder the table rows within the selected table into improved native key order. This improved native key order enhances the efficiency of applications that must access the data table in native key sequence.

OLREORG runs within the MUF while the data table is open for user access. This allows you to improve the sequential efficiency of the data without taking the data table offline.

Restart/recovery fully protects OLREORG within the MUF. Therefore, you can cancel the OLREORG process at any time and the amount of completed work is kept. Similarly, any MUF abend or LPAR failure terminates the OLREORG, but the reordering work that is done in the data table is kept.

You can implement the OLREORG process with three different modes:

• Reference group within existing data blocks (REFGROUP=2 or more). In this mode, the OLREORG process selects a set of blocks (the reference group) and moves rows between the blocks to improve the sequential order of the rows in those blocks. Then it moves onto the next set of blocks and repeats. The end goal of this reorganization is to improve native order efficiency within the existing data blocks. While moving rows to improve order, some blocks may be emptied and placed on the free block list.
• Reference group with additional data blocks (REFGROUP=2 or more). In this mode, the OLREORG process follows the same process as noted previously. However, if the order of the rows in the reference group cannot be improved because no free space is available in the selected blocks, the OLREORG process is allowed to acquire an empty block from the data area and use it to move the rows into sequential order and improve the efficiency. It then moves onto the next set of blocks and repeats. The end goal of this reorganization is to improve native order efficiency within the existing data blocks where possible and to use a limited number of new blocks to improve order for groups that cannot be improved otherwise. While moving rows to improve order, some blocks may be emptied and placed on the empty block list. In many cases, these newly emptied blocks make up for the empty blocks that are used during the move process. For OLREORG REFGROUP to use empty blocks for the move process, the DSOP option must be set to option 4 or option 5. For more information, see the following Data SPACE Option changes.
• Reference group with full reordering of data rows (REFGROUP=0). In this mode, the OLREORG process utilizes empty data blocks to copy the data rows into native sequence order. At the end of the REFGROUP=0 reorganization, the data rows are native key sequence within their data blocks providing a sequential efficiency similar to a full offline reorganization (BACKUP/LOAD). For OLREORG REFGROUP to use empty blocks for the move process, the DSOP option must be set to option 4 or option 5. For more information, see Data SPACE Option 4 - Random Reuse for OLREORG and Data Space Option 5 - Sequential Reuse for OLREORG changes.

For detailed instructions about executing the DBUTLTY OLREORG function, see the CA Datacom/DB Data Reorganization Green Book.

Data Space Option 4 - Random Reuse for OLREORG

To support the use of empty blocks during the OLREORG reference group reorganizations, the data area must have its data space option set to DSOP=4 (or DSOP=5).

DSOP=4 provides the same data space reuse functionality as DSOP=1. Additionally, DSOP=4 creates and maintains an empty data block map that is used for the OLREORG processing.

For detailed information about the DSOP=4 setting, see the CA Datacom/DB Data Reorganization Green Book.

Data Space Option 5 - Sequential Reuse for OLREORG

To support the use of empty blocks during the OLREORG reference group reorganizations, the data area must have its data space option set to DSOP=5 (or DSOP=4).

DSOP=5 provides the same data space reuse functionality as DSOP=2. Additionally, DSOP=5 creates and maintains an empty data block map that the OLREORG processing uses.

For more detailed information about the DSOP=5 setting, see the CA Datacom/DB Data Reorganization Green Book.