An HDB keeps its data in sequential data sets called containers.
A new data set is created every time a request is submitted to load
data into the HDB.
Saving data in small data sets rather than one monolithic table
or data set makes management of the environment simpler:
- You can start using an HDB immediately without worrying whether
enough DASD space is available to hold many year's worth of data.
- DFHSM can migrate old data, ensuring only the most recent or required
data is retained online for immediate reporting, saving expensive
DASD resources.
- ABENDX37 conditions are avoided. In the event of a data set full
condition, CICS PA simply
closes the full data set and continues loading into a new one.
- Individual data sets can be loaded directly into a DB2® table or CSV extract data
set for further analysis.