GCP Datastream cost: per GB of change data processed

Quick answer

Datastream bills per GB of data processed, covering both ongoing change-data-capture (CDC) from source databases and the initial backfill of existing data. Cost scales with the change rate of your sources and the size of backfills, so streaming only the tables and columns you need, and backfilling selectively, are the levers. The destination (BigQuery, Cloud Storage) and any transformation add their own cost.

Google Cloud Datastream is a serverless change-data-capture (CDC) service that replicates data from source databases to destinations like BigQuery and Cloud Storage in near real time. It is priced per gigabyte of data processed, so cost tracks how much change data flows and how much existing data you backfill.

CDC and backfill, per GB

Component	Bills for
Change data capture (CDC)	Per GB of changes processed
Backfill	Per GB of existing data replicated initially
Destination + transform	BigQuery/Cloud Storage and processing cost

Ongoing CDC bills per gigabyte of change data captured from the source (inserts, updates, deletes), so a high-write source generates more. The initial backfill of existing data bills per gigabyte too, so backfilling large tables is a one-time cost proportional to their size. The destination service and any transformation add their own charges.

What drives the bill

Source change rate drives ongoing CDC cost, a write-heavy database generates a lot of change data. Backfill size drives the initial cost, replicating large historical tables. Streaming tables and columns you do not need, and backfilling more history than required, are the common wastes.

Controlling Datastream cost

Stream only the tables and columns you actually need at the destination rather than everything, backfill selectively (only the history you require), and be mindful of the destination and transformation cost downstream. Because it is per gigabyte processed, reducing the change data and backfill volume is the lever, the same do-only-what-you-need discipline as other data-movement services like DataSync.

FAQ

How is GCP Datastream priced?

Per gigabyte of data processed, covering both ongoing change-data-capture (CDC) from source databases and the initial backfill of existing data. The destination service (BigQuery, Cloud Storage) and any transformation add their own cost. Datastream cost scales with source change rate and backfill size.

How do I reduce Datastream cost?

Stream only the tables and columns you actually need at the destination rather than everything, backfill selectively (only the history you require), and account for the destination and transformation cost downstream. Reducing the change-data and backfill volume is the main lever, since billing is per gigabyte processed.

What drives Datastream cost?

Source change rate drives ongoing CDC cost, since a write-heavy database generates more change data, and backfill size drives the initial cost of replicating existing data. Streaming tables and columns you do not need, and backfilling more history than required, are the common wastes that inflate the per-gigabyte bill.

Does the initial backfill cost extra in Datastream?

Yes. The initial backfill of existing data bills per gigabyte, like ongoing CDC, so replicating large historical tables is a one-time cost proportional to their size. Backfilling only the tables and history you actually need, rather than everything, controls this initial cost.

Is Datastream cheaper than self-managed CDC?

It depends on scale and operational load. Datastream removes the burden of building and running CDC pipelines yourself, billing per gigabyte processed. For teams without CDC expertise or with variable replication needs, the managed serverless model often wins on total cost; at very high steady volume, a self-managed pipeline could be cheaper per gigabyte.

Does C3X estimate Datastream cost?

Datastream cost is driven by change-data and backfill volume, which are usage inputs. C3X prices the surrounding infrastructure, including the BigQuery or Cloud Storage destination, and you model your source change rate and backfill size to estimate the replication charges.

What to do next

Estimate the infrastructure around your replication before you build it. C3X reads your Terraform and prices your resources against a live catalog. Start with the quickstart.