Cassandra to GCS
real-time replication
TwinCommit replicates Apache Cassandra to Google Cloud Storage in real time by reading the Cassandra transaction log and applying every committed change exactly once and in commit order. Use it to land live Cassandra data in GCS as open Delta Lake or Iceberg tables, with no triggers or capture tables on the source.
Cassandra
log reader
commit-ordered · exactly-once
GCS
native writer
How capture works on Cassandra
TwinCommit tails the CDC commit-log segments on every node, deduplicates replica writes, and orders mutations by write timestamp per partition.
Prerequisites
- ✓cdc_enabled: true in cassandra.yaml
- ✓WITH cdc = true on replicated tables
- ✓Agent sidecar or shared volume access to cdc_raw on each node
How apply works on GCS
Changes land as Apache Iceberg, Delta Lake or Parquet tables with one atomic commit per micro-batch. Iceberg tables register in the BigLake metastore so BigQuery can query them in place.
Requirements
- ✓Storage Object Admin on the bucket
- ✓BigLake Admin for catalog registration (optional)
- ✓Workload identity or a service-account key
Setup
Set up Cassandra → GCS in four steps
- 1
Prepare Cassandra
Run twincommit check. It verifies cdc_enabled: true in cassandra.yaml and the other prerequisites, and prints the exact fix for anything missing.
- 2
Point at GCS
Add a target block for GCS. 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: cassandra
contact_points: [cass-1, cass-2, cass-3]
cdc_dir: /var/lib/cassandra/cdc_raw
target:
type: gcs
bucket: acme-lake
path: replica/{schema}/{table}
format: iceberg
catalog: { type: biglake, project: acme-analytics }
tables:
- source: sales.orders
- source: sales.customersCassandra to GCS replication FAQ
How does TwinCommit replicate Cassandra to GCS?
+
TwinCommit reads changes from the Cassandra transaction log, buffers them per transaction, and applies each committed transaction to GCS in commit order. Changes land as Apache Iceberg, Delta Lake or Parquet tables with one atomic commit per micro-batch. Iceberg tables register in the BigLake metastore so BigQuery can query them in place.
Does Cassandra to GCS replication need triggers or capture tables?
+
No. TwinCommit reads the Cassandra log directly and creates no objects in the source database.
Is Cassandra to GCS replication exactly-once?
+
Yes. TwinCommit commits its checkpoint in the same GCS transaction as the data, so a crash or restart never loses or duplicates a transaction.
What Cassandra versions are supported?
+
Apache Cassandra 4.0, 4.1, 5.0 · DataStax Enterprise 6.8+ · Astra DB (target) · ScyllaDB (target).
Stream Cassandra to GCS today.
Download the free Developer edition, or have one of our replication engineers walk you through a migration plan.