Partitioning HDAM databases of partitioning process

BMC AMI Partition Facility for IMS can use one of the following methods to partition HDAM databases:

• A standard HDAM randomizing algorithm (a single-stage randomizer)
  When using a single-stage randomizer, BMC AMI Partition Facility for IMS determines the number of blocks in the entire database and passes this information to the randomizer. The randomizer examines the root segment key and selects the root segment pointer address, known as the Root Anchor Point (RAP) to determine the storage location of each root segment in the database. BMC AMI Partition Facility for IMS determines where to partition the database, based on the number of blocks in the root addressable area (RAA) of each partition.

• A DEDB randomizing algorithm (a two-stage randomizer)
  A two-stage randomizer randomizes within a particular partition to keep all records for specific criteria (such as geographic area) within the same partition. The randomizer determines into which partition/area to place a root segment. The routine then determines the location of the root segment within that partition/area.

• Partition Selection Routine
  The routine selects the partition.

• Highkeys
  If a highkey is defined for each partition, BMC AMI Partition Facility for IMS can use the key ranges to determine the partition boundaries.
  When not using a partition selection routine or highkeys, BMC AMI Partition Facility for IMS uses the name of the randomizer load module to determine which interface to use. If the load module name begins with DBF, BMC AMI Partition Facility for IMS uses the Fast Path DEDB interface. For any other load module name, BMC AMI Partition Facility for IMS uses the standard HDAM randomizer.

Randomizing HDAM databases