Unanswered: DPF to single partition conversion in db2 10.5 fp7 + AIX
Would like to thank you all in advance !!
We have got an requirement to convert our multipartition database to a single partition database.
What all measures to take care to perform the activity, is there any process other than export and dump tables.
What can be considered as pre-requisites.
Explain the motivation. Post whether adaptive-row-compression is available for the single partition environment or whether the single-partition environment is BLU . What's the total number of terabyes to migrate? Is the new environment physically different from the DFP environment? Is there IP-connectivity between them (load from cursor). Presumably you've already proved that the single-table limits won't be exercised by the merge from multiple partitions to single partitions. For the DDL, adjustments will be needed to remove the distribute-by clause plus any tablespace alterations necessary.
Thanks for response, Motive behind this is to establish HADR for the system
we have to migrate around 4 TB of data and will be having adaptive row compression. In terms of resources, both servers will be same.
Rest is in line.
Strange motive, I would not like to have to re-establish HADR on an 4-TB compressed database. Anyway, your choice. If the single-node instance and DFP-instance can be both UP at the same time, with both databases concurrently activated, then you might consider remote-load-from-cursor parallelised (via scripting) to arrange the data transfer, having already ensured that no single-table-limits are broken by the data. This presumes the SAN allocations are sufficient for both databases to coexist for a period on different LUNs, and the prereq is also that you've successfully migrated the DDL to the single node instance. Be aware that if any code-objects exist that use DFP-specific syntax then that code will need adjusting also.