Side by side
dbmazz vs TiCDC
Both stand on change data capture software. Which one you can start free with, which one lets you read the code, and what each costs once you pay.
Free plan LinuxMacSelf-hosted
dbmazz provides a focused path for streaming database changes, with a one-job-per-instance limit and a defined set of stable sinks. Teams should also account for its license restriction and any separately charged BYOC support or connectivity.
Visit dbmazz Free plan Open source APILinuxMac+1
TiCDC provides incremental TiDB replication with filtering, multiple sink types, and cluster-to-cluster options. Plan for its at-least-once delivery behavior, table index requirements, and documented unsupported cases before relying on a replication path.
Visit TiCDCThe facts
| dbmazz | TiCDC | |
|---|---|---|
| Free plan | Yes | Yes |
| Open source | Not published | Yes |
| Runs on | Linux, Mac, Self-hosted | API, Linux, Mac, Self-hosted |
| Lowest paid plan | Free or on request | Free or on request |
| Plans published | 2 | 1 |
| Maker | Not published | Not published |
| Website | github.com | docs.pingcap.com |
| On change data capture software | #5 of 23 | #1 of 23 |
dbmazz: what it does well
- Streams inserts, updates, and deletes through logical replication.
- Stable sinks include StarRocks, PostgreSQL, and Snowflake.
- Automatic setup can create replication slots and sink tables.
- Self-hosted engine and BYOC plans are free.
- CLI supports Linux and macOS on amd64 and arm64.
What to know first
- Each instance handles one replication job.
- S3/Iceberg is listed as roadmap, not stable.
- BYOC support and optional PrivateLink are charged separately.
- The license restricts offering it as a managed service.
TiCDC: what it does well
- Replicates to databases, Kafka, cloud storage, and NFS.
- Supports bidirectional TiDB cluster replication.
- Filters databases, tables, DMLs, and DDLs.
- Nodes can be added or removed dynamically.
What to know first
- At-least-once delivery can repeat statements or Kafka messages.
- Tables generally need a valid index.
- Large transactions have partial support and out-of-memory risk.
- RawKV-only clusters are unsupported.