IBM Db2 data sources
Overview
The IBM Db2 virtualization solution consists of the Delphix Continuous Data Engine, and the connector for IBM Db2 which must be uploaded and installed into the Delphix Continuous Data Engine. This solution ingests source data using the staging architecture, which copies the production database from the source environment into a staging environment. The Delphix Continuous Data Engine captures the persistent storage layer of the staging database, which then allows Delphix Continuous Data Engine to create virtualized copies in target environments.
![db2 overview.png](../../__attachments/99453477/db2%20overview.png?inst-v=9f5cd9ef-94fe-4677-b3c4-6c5d1617af5c)
IBM Db2 connector architecture
IBM Db2 connector supports ingestion of both single partition (non-DPF) AND multi-partition (logical) also known as DPF (Database Partition Feature) databases.
IBM Db2 architecture for single partition (non-DPF) databases:
![Db2 connector architectue.png](../../__attachments/99453477/Db2%20connector%20architectue.png?inst-v=9f5cd9ef-94fe-4677-b3c4-6c5d1617af5c)
IBM Db2 architecture with multi-partition database (DPF):
A multi-partition, or partitioned, database is a database with more than one partition. Depending on your hardware environment, there can be several configurations in which you can partition your database. During DPF dSource ingestion, the production database can be physically partitioned across multiple hosts. However, for the dSource ingestion or VDB provisioning process, the staging and target hosts will have to be configured using logical partitions only.
![multi-partition architecture.png](../../__attachments/99453477/multi-partition%20architecture.png?inst-v=9f5cd9ef-94fe-4677-b3c4-6c5d1617af5c)
The IBM Db2 connector supports both staging pull and staging push.
The following specific ingestion methods are supported:
Staging Pull: This ingestion method allows Delphix Continuous Data Engine to execute the necessary operations to prepare the data content in the Staging database. Staging pull supports the below ingestion methods for both DPF and non-DPF databases.
HADR (High Availability Disaster Recovery)
Backup only
Backup with Archived logs
Rollforward to Point-in-Time (RFPIT)
Staging Push: This ingestion method requires you to execute the necessary operations to prepare the data content in the staging database for both DPF and non-DPF databases.
Limitations of the IBM Db2 solution
Delphix Continuous Data Engine and the IBM Db2 connector should not be considered a replacement for backup and recovery of your source instances and databases. The Delphix Continuous Data Engine solution is intended for very different purposes, primarily to create non-production database copies.
The following are not supported:
Architecture other than 64-bit OS
Windows OS type
Db2 Versions below 10.5 i.e (10.1, 9.8 etc)
Db2 Offline backup at staging for staging push
Migrating existing VDBs to a different host
Virtual to Physical (V2P) VDB operation
Db2 Pure Scale databases
Multi-node configuration of dSource or VDB databases
Support
License for the IBM Db2 connector is included with the standard Delphix License Agreement, and the solution is supported by Delphix Technical Support. For any questions, bugs, or feature requests, contact us via Delphix Support or Delphix Community portals.