Db2 to Databricks
real-time replication
TwinCommit replicates IBM Db2 to Databricks in real time by reading the Db2 transaction log and applying every committed change exactly once and in commit order. Use it to feed Databricks with live Db2 data for analytics, AI and reporting, with no triggers or capture tables on the source.
Db2
log reader
commit-ordered · exactly-once
Databricks
native writer
How capture works on Db2
TwinCommit reads the Db2 recovery log through the native log read API and decodes row images itself. No SQL Replication, no Q Replication, no CD tables, no triggers and no capture agents are installed on the source.
Prerequisites
- ✓Archive logging enabled (LOGARCHMETH1)
- ✓DATA CAPTURE CHANGES on each replicated table
- ✓A primary key or unique index on each replicated table
- ✓DBADM (or SYSADM) for the capture user
How apply works on Databricks
Micro-batches are staged as Parquet in your cloud storage and applied with Delta MERGE through a SQL warehouse. Deletes, updates and ordering are preserved per key.
Requirements
- ✓A SQL warehouse or all-purpose cluster
- ✓A Unity Catalog volume or external location for staging
- ✓A service principal with MODIFY on the target schema
Setup
Set up Db2 → Databricks in four steps
- 1
Prepare Db2
Run twincommit check. It verifies archive logging enabled (logarchmeth1) and the other prerequisites, and prints the exact fix for anything missing.
- 2
Point at Databricks
Add a target block for Databricks. TwinCommit can generate target DDL with twincommit ddl -apply.
- 3
Snapshot
twincommit snapshot copies each table under a brief share lock and records the exact log position.
- 4
Stream
twincommit run streams every committed change from that position, exactly once and in commit order.
source:
type: db2
database: PRODDB
user: ${SRC_USER}
password: ${SRC_PASSWORD}
target:
type: databricks
host: adb-1234567890.12.azuredatabricks.net
warehouse: replication-wh
catalog: main
schema: replica
tables:
- source: SALES.ORDERS
- source: SALES.ORDER_LINES
exclude_columns: [INTERNAL_NOTES]Db2 to Databricks replication FAQ
How does TwinCommit replicate Db2 to Databricks?
+
TwinCommit reads changes from the Db2 transaction log, buffers them per transaction, and applies each committed transaction to Databricks in commit order. Micro-batches are staged as Parquet in your cloud storage and applied with Delta MERGE through a SQL warehouse. Deletes, updates and ordering are preserved per key.
Does Db2 to Databricks replication need triggers or capture tables?
+
No. TwinCommit reads the Db2 log directly and creates no objects in the source database.
Is Db2 to Databricks replication exactly-once?
+
Yes. TwinCommit commits its checkpoint in the same Databricks transaction as the data, so a crash or restart never loses or duplicates a transaction.
What Db2 versions are supported?
+
Db2 LUW 11.1, 11.5, 12.1 · Db2 for z/OS 12 and 13 · Db2 on Cloud (target).
Use cases for this route
Related replication routes
Stream Db2 to Databricks today.
Download the free Developer edition, or have one of our replication engineers walk you through a migration plan.