GCP Dataplex cost: data compute units for tasks and management

Quick answer

Dataplex bills per data compute unit (DCU) hour for its processing, data discovery, quality checks, and tiering tasks, plus shuffle storage for those tasks, and metadata for its catalog. Cost scales with how much processing you run, so scheduling discovery and quality tasks to the frequency your data actually changes, and right-sizing DCUs, are the levers.

Google Cloud Dataplex is a data fabric for managing, governing, and organizing data across a data lake and warehouse. Its cost comes mainly from the processing it runs, discovery, data quality, and tiering tasks, billed in data compute units, so the frequency and scope of that processing drives the bill.

Data compute units and storage

Component	Bills for
Processing (discovery, quality, tiering)	Per DCU-hour
Shuffle storage	Per GB for task shuffle
Catalog metadata	Metadata storage and operations

Dataplex tasks, automatically discovering data, running data quality checks, and tiering, consume data compute units billed per hour, plus shuffle storage for the processing. The catalog that holds metadata adds its own charge. The processing is the main cost, and it scales with how often and how broadly you run these tasks.

What drives the bill

Running data discovery and quality tasks frequently across large data estates is the main driver. Scanning everything hourly when data changes daily wastes DCU-hours, and over-provisioning task compute processes the same data with more units than needed. Scoping and scheduling the tasks is where the saving is.

Controlling Dataplex cost

Schedule discovery and data quality tasks to match how often your data actually changes rather than a frequent fixed clock, scope tasks to the data assets that need them, right-size the DCUs allocated to tasks, and use incremental processing where supported so tasks handle only new or changed data. The do-only-what-you-need discipline mirrors Glue: process the data that changed, at the frequency it changes, with right-sized compute.

FAQ

How is GCP Dataplex priced?

Per data compute unit (DCU) hour for its processing tasks, data discovery, quality checks, and tiering, plus shuffle storage for those tasks and metadata storage for its catalog. The processing is the main cost, scaling with how often and how broadly you run the tasks.

What drives Dataplex cost?

Running data discovery and quality tasks frequently across large data estates. Scanning everything hourly when data changes daily wastes DCU-hours, and over-provisioning task compute processes the same data with more units than needed. Task scope and schedule are where the cost is decided.

How do I reduce Dataplex cost?

What is a data compute unit in Dataplex?

The billing unit for Dataplex processing, representing compute allocated to tasks like discovery, data quality, and tiering, billed per DCU-hour. More DCUs process data faster or handle more at higher cost, so right-sizing DCUs to the task's real needs controls the processing cost.

Does Dataplex catalog cost extra?

The catalog that holds metadata about your data assets adds metadata storage and operation charges on top of the processing DCU cost. For most deployments the processing tasks dominate, but the catalog contributes, so managing metadata scope alongside task scheduling keeps the total proportional.

Does C3X estimate Dataplex cost?

Dataplex cost is driven by DCU-hours for tasks and metadata, which are usage inputs tied to task frequency and scope. C3X prices the surrounding GCP infrastructure, and you model your data-management task activity to estimate the Dataplex charges.

What to do next

Estimate the infrastructure around your data governance before you deploy. C3X reads your Terraform and prices your resources against a live catalog. Start with the quickstart.