iTera'sReorganize While Active Eliminates Downtime From File Reorganizations Each time a record is deleted from a file on the iSeries, the disk space that the deleted record occupied isn't normally reused until a file reorganization is performed. IBM ships a utility (RGZPFM) to reorganize your files; however, when performing a reorg with this tool, no users or processes can access the file. As a result, large, heavily used files can take several hours to reorganize. This of course means significant system downtime and a major cause of business disruption. With iTera's Reorganize While Active product, you can eliminate this downtime and reap the system performance benefits of regularly reorganized files.
Exclusive File Analysis Tool Before beginning a file reorganization with Reorganize While Active, use iTera's file analysis tool to quickly see the files with the most deleted records and the amount of disk space that will be recovered when the reorg process runs.
You Choose the Best Way to Reorganize!
"In-Place" The "in-place" method can be used for files with many kinds of attributes, and reclaims all of the deleted records right from the production file - no copy or synchronization is necessary. When using the in-place method to reorganize, you have the option to keep the records in the same sequential order when the file is reorganized (same as IBM's RGZPFM command), or you can reorganize the file "randomly" which moves the records from the end of the file toward the beginning. If your files are being journaled, you can further protect data integrity by adding commitment control functionality. Exclusive allocation of the file is required for a few seconds to complete the process, but only at a time that is conventient.
"Mirrored File" The "mirrored file" method makes a copy of the file in a new library and performs the reorganization as it is being copied. This copy is then kept "in sync" with changes that occur in the production file until a convenient time when the synchronized copy can replace the production file. The Mirrored-File method is typically used when triggers or referential constraints are used, if data warehousing is active, or multiple members are being reorganized.
