Orientation on Technology

The ad hoc demands placed on the data warehouse will often require the use of temporary tables to store interim results. Space will need to be set aside for this purpose. That space needs to be efficiently managed. The consequence of not managing it efficiently is that either the space required will be many times larger than necessary, or everything will grind to a halt. lf the system and database management tool allows this space to be efficiently monitored, and allocated space to be dated and tracked, the space can be kept to a minimum, saving many tens of gigabytes of disk space.

As ever, it is the size and complexity of the data warehouse that is the issue here. Large numbers of error log and trace files will be produced. These need to be regularly and automatically scanned for errors or problems. This is one of the functions of the event manager. These files also need to be managed, with old log and trace files being purged to save space and make it easier to deal with the current logs.

Logs that contain significant data should be archived. Logs such as the database log, which contains all major changes, should be kept to maintain a history. This could prove to be important in a crisis situation, where the thread of events over the last number of weeks or months can become important. Any log files that contain errors that have not yet been explained should also be kept until the cause of the error is understood. The tool should allow such indicated log files to be archived, while others are deleted.