Managing data redistribution on database partition groups
In this recipe, we will perform data redistribution on database partitions 3
and 4
. For the method, we will use the default uniform distribution.
Getting ready
Now that we have added partitions 3
and 4
to partition group NAVDATAGRP
and the corresponding containers to the NAV_TBLS
and NAV_INDX
table spaces, the next logical step should be to redistribute data on all existent partitions.
How to do it...
Internally, DB2 9.7 uses a distribution mechanism based on a partition map, which is actually an array containing 4,096 entries for backward compatibility. It can be extended to 32768 by setting the DB2_PMAP_COMPATIBILITY
registry variable to OFF
. The distribution map is generated and associated internally with a database partition group when is created. Every partition defined on a database partition group is mapped inside this array in a round-robin fashion.
We will generate a distribution map once before data redistribution for database...