Data Catalog is the metadata management service in Google Cloud. This article covers what it does, the kinds of metadata it tracks, the search and tagging features that matter for the Associate Cloud Engineer exam, and how questions about it tend to show up.
It does not cover every Dataplex feature, the data lineage product, or the deeper data mesh architecture patterns that build on top of Data Catalog. Data Catalog is now folded into the Dataplex umbrella, and you may see either name in the documentation. For ACE purposes, the concept is the same. The goal here is to give you what you need for the exam, not a full governance reference.
Data Catalog is a fully managed metadata service. Think of it as a search engine and central index for the data sitting across your GCP services. BigQuery datasets and tables, Cloud Storage buckets and objects, Pub/Sub topics, and other resources all generate metadata. Data Catalog collects that metadata in one place so you can find what you have without manually opening every service.
This matters because once an organization has more than a handful of projects, nobody knows where everything lives. A new analyst joins and needs to find the customer transactions table. A compliance team needs to locate every dataset that contains personal information. Without a catalog, the answer is "ask around and grep through the console." With Data Catalog, the answer is one search.
The course lists the kinds of metadata Data Catalog pulls in. Table schemas from BigQuery. Object metadata from Cloud Storage. Topic configurations from Pub/Sub. Column families from Bigtable. Cluster metrics. Pipeline templates. The basic idea is that anywhere data lives in GCP, the metadata about that data flows into the catalog.
The most important thing to understand is the distinction between data and metadata. Data Catalog does not store your actual data. It stores information about your data. Schema, ownership, location, descriptions, tags. The data itself stays in BigQuery or Cloud Storage or wherever it already lives. The catalog is the index, not the warehouse.
Search is the headline feature. You can query the catalog for column names, table names, dataset names, or descriptions across every project the catalog has access to. Canonical examples are the example of a regulatory compliance audit. Instead of opening every BigQuery dataset to look for tables that contain personal information, you search the catalog for everything tagged with a personal information label and you get a single list back.
This is the use case that matters most for the Associate Cloud Engineer exam. If a question describes a scenario where someone needs to find data across multiple projects, or audit what sensitive data exists, or discover what assets a team owns, Data Catalog is the answer.
Beyond the metadata that GCP automatically pulls in, you can add your own metadata through tags. A tag template defines a set of fields. Retention period, sensitivity level, environment, owner, anything you want. Once a template exists, you apply tags based on it to specific resources.
The reason this matters beyond simple labeling is that Data Catalog policy tags can plug into BigQuery and BigLake to enforce column-level security. You define a policy tag like "Confidential PII" and apply it to specific columns. Then you grant read access to that policy tag only to certain users or groups. Anyone querying the table without that grant cannot see those columns. This is the foundation for fine-grained governance and is the most important piece of Data Catalog that goes beyond simple search.
Data Catalog is not a heavily tested topic on the Associate Cloud Engineer exam, but when it shows up, the patterns are recognizable.
If you see a question about discovering data assets across multiple projects, finding datasets by column name or tag, or auditing what data exists in an organization, the answer is Data Catalog. The other GCP services that hold data, BigQuery, Cloud Storage, Pub/Sub, do not provide cross-service search by themselves. That is the whole point of having a separate catalog service.
If you see a question about column-level security in BigQuery, especially one mentioning policy tags, Data Catalog is in the answer because that is where policy tags live. The taxonomy is defined in Data Catalog and consumed by BigQuery for access enforcement.
If you see a question about cataloging metadata for governance or compliance, Data Catalog is the answer. The exam treats it as the governance and discovery layer, separate from the storage and analytics services that actually hold the data.
Data Catalog is the metadata index for GCP. It pulls schemas, configurations, and other metadata from across services into a single searchable place. Tag templates let you add custom metadata, and policy tags integrate with BigQuery for column-level access control. The data itself never leaves the underlying services. The catalog is just the index on top.
For the Associate Cloud Engineer exam, the cue is "discovery, search, or governance across multiple data services." When you see that phrasing, Data Catalog is the answer.
My Associate Cloud Engineer course covers Data Catalog in the data and analytics section, alongside the BigQuery security model and how policy tags actually get enforced at query time.
