Diagram and Query Google AlloyDB's HTAP Architecture

AlloyDB speaks PostgreSQL, so DbSchema connects with the standard PostgreSQL JDBC driver it already bundles — and models the schema, from partitioned tables to pgvector columns, with diagrams, documentation, and Git versioning on top.

Connect with the bundled PostgreSQL driver — no extra setup
Diagram partitioned tables, foreign tables, and extensions
Run OLTP and analytical queries over the same connection
Reach AlloyDB Omni on-premises with the identical setup

available for

ISO-27001 certified

Free Community Edition No signup required 15-day Architect trial included

Free Download

DbSchema modeling a Google AlloyDB schema as an ER diagram

Trusted by teams at

The Columnar Engine Is Automatic
Schema Clarity Is Not

AlloyDB accelerates analytics with an in-memory columnar engine and keeps transactions fast on row storage — all behind the PostgreSQL interface your team already knows. What it cannot automate: how the schema fits together, what drifted between environments, and what a migration will actually do.

DbSchema covers that side — a design model with diagrams, published documentation, and SQL you review before it runs.

Query, Compare & Publish
the AlloyDB Schema

Postgres-grade tooling, pointed at Google's HTAP service.

Write HTAP queries against Google AlloyDB in the DbSchema SQL editor

One Editor for Both Sides of HTAP

Write transactional queries and analytical aggregations against the same connection — with pgvector operators for similarity search and autocompletion that reads your schema.

Compare AlloyDB environments and generate migration SQL

Environment Drift, Itemized

Compare the model with development and production instances, then generate the migration SQL you approve difference by difference.

Publish HTML documentation of a Google AlloyDB schema

Schema Answers Without Console Access

Publish interactive HTML5 documentation — tables, extensions, comments — that teammates browse in a browser, no Google Cloud login required.

Migrations Reviewed as Pull Requests
Before They Touch the Primary

Keep the AlloyDB schema as a model file in your repository. Branch it and pull-request design changes with the application code, then let DbSchema produce the SQL and carry it through development, staging, and production.

Git-based schema synchronization workflow for Google AlloyDB

Read Git Workflow Docs

Reaching AlloyDB over JDBC

DbSchema uses the standard PostgreSQL JDBC driver (org.postgresql.Driver), which it already includes — AlloyDB's protocol compatibility means no additional driver setup. Because instances have no public IP by default, start the AlloyDB Auth Proxy or configure Private Service Connect so JDBC traffic can reach the instance from your machine.

The URL takes the form jdbc:postgresql://alloydb-instance:5432/mydb on port 5432, where the host is the primary's private IP or the address the Auth Proxy exposes; authenticate with a PostgreSQL-compatible username and password, or IAM database authentication. For high availability, point DbSchema at the cluster endpoint, which routes to the current primary — and AlloyDB Omni on-premises connects with the same driver configuration. Details are on the AlloyDB JDBC driver page.