TUC provides functionality that allows you to monitor your DB2 datasets disk storage and performance, alert and notify critical conditions, and automate corrective actions to avoid availability issues.
The monitoring allows you to:
- Get an early warning when the object is about to reach its maximum size.
- Detect lack of free space in your SMS storage groups and volumes.
- Prevent failures on out of storage or hitting the extents limit.
- Trigger REORG or statistics collection based on growth.
- Report space usage per database for capacity planning, including flat files.
Collecting Growth Statistics
TUC can collect growth statistics and project when an object is about to reach its maximum available space. An early warning can be issued to avoid an outage by planning to enlarge the object, add partitions, compress or delete data. By default, statistics are collected only for objects that grew in more than 20MB since last collection. An early warning is issued when the object is about to reach its maximum available space within the next 7 days. An early warning is also issued for fast growing tables that might reach the maximum available space within hours due to a run-away process. To get these warnings you will need to run the collect job frequently enough.
SMS Storage Groups Free Space
DB2 volumes might need more attention to avoid allocation failures due to lack of free space in the DASD pool. This is important to allow unexpected growth but also for online REORG since shadow datasets are allocated by online REORG. The shadow datasets need at least the same space as the datasets in use. Space is released as soon as online REORG switches to the shadow datasets but the pool must have enough free space to allow allocating the shadow datasets while the datasets in use are still allocated. TUC has an efficient fast process that can quickly provide an alert if the free space in the DB2 volumes goes below the specified limit. The process can run once a day or even multiple times a day to issue an early warning if space needs to be released or added, for example, by adding more volumes to the pool or dropping unused objects.
Accurate calculation of the required space, minimizes wasted space, reduces extents and prevents going out of space. To better manage your allocated space, you can estimate the maximum number rows expected in each table. TUC automatically corrects the estimated number of rows when the actual number of rows in the table exceeds the estimate. TUC calculates space using the provided estimates and alters space allocation and compression prior to REORG.
Compressed tables require less DASD space, less log space, less backup space and shorter backup time. TUC automates the process of identifying tables with compression benefits.
Free Space Management
TUC calculates and alters the required allocated space prior to reorg. Since reorg reinstate free space, it is also required to adjust free space as needed. The calculation of space is also increasing free space by 50 percent if the object was selected for reorg because of overflows in tablespaces or splits in indexes. Free space should be sufficient to allow update activity for at least 6 months before the object number of overflows or splits exceed the defined threshold. Therefore, free space is not increased if the object was last reorged more than 6 months ago or if the object was already reorged in the recent week. Increase in free space (higher PCTFREE and lower FREEPAGE), also means that more space is allocated. TUC calculates accurately the extra space required for the free space increase.
PCTFREE is increased by 50 percent up to the maximum of 40 for tablespaces and 10 for indexes. FREEPAGE is always 100/PCTFREE.
Space usage reports can be helpful in planning future capacity based on growth. TUC maps all the DB2 linear VSAM datasets and all flat files, including datasets created by UNLOAD or DISCARD. Reports can show how much space is allocated for each database and what is the rate of growth.