AlloyDB Python Connector

The AlloyDB Python Connector is the recommended way to connect to AlloyDB from Python applications. It provides:

• Secure connections — TLS 1.3 encryption and identity verification, independent of the database protocol
• IAM-based authorization — controls who can connect to your AlloyDB instances using Google Cloud IAM
• No certificate management — no SSL certificates, firewall rules, or IP allowlisting required
• IAM database authentication — optional support for automatic IAM DB authentication

Supported drivers: pg8000 (sync) · psycopg (sync) · asyncpg (async)

Quickstart

Sync (pg8000 + SQLAlchemy)

Install:

pip install "google-cloud-alloydb-connector[pg8000]" sqlalchemy

Connect:

import sqlalchemy
from google.cloud.alloydbconnector import Connector

INSTANCE_URI = "projects/MY_PROJECT/locations/MY_REGION/clusters/MY_CLUSTER/instances/MY_INSTANCE"

with Connector() as connector:
    pool = sqlalchemy.create_engine(
        "postgresql+pg8000://",
        creator=lambda: connector.connect(
            INSTANCE_URI,
            "pg8000",
            user="my-user",
            password="my-password",
            db="my-db",
        ),
    )

    with pool.connect() as conn:
        result = conn.execute(sqlalchemy.text("SELECT NOW()")).fetchone()
        print(result)

Sync (psycopg + SQLAlchemy)

Install:

pip install "google-cloud-alloydb-connector[psycopg]" sqlalchemy

Connect:

import sqlalchemy
from google.cloud.alloydbconnector import Connector

INSTANCE_URI = "projects/MY_PROJECT/locations/MY_REGION/clusters/MY_CLUSTER/instances/MY_INSTANCE"

with Connector() as connector:
    pool = sqlalchemy.create_engine(
        "postgresql+psycopg://",
        creator=lambda: connector.connect(
            INSTANCE_URI,
            "psycopg",
            user="my-user",
            password="my-password",
            db="my-db",
        ),
    )

    with pool.connect() as conn:
        result = conn.execute(sqlalchemy.text("SELECT NOW()")).fetchone()
        print(result)

Async (asyncpg + SQLAlchemy)

Install:

pip install "google-cloud-alloydb-connector[asyncpg]" sqlalchemy

Connect:

import asyncio
import sqlalchemy
from sqlalchemy.ext.asyncio import create_async_engine
from google.cloud.alloydbconnector import AsyncConnector

INSTANCE_URI = "projects/MY_PROJECT/locations/MY_REGION/clusters/MY_CLUSTER/instances/MY_INSTANCE"

async def main():
    async with AsyncConnector() as connector:
        pool = create_async_engine(
            "postgresql+asyncpg://",
            async_creator=lambda: connector.connect(
                INSTANCE_URI,
                "asyncpg",
                user="my-user",
                password="my-password",
                db="my-db",
            ),
        )

        async with pool.connect() as conn:
            result = await conn.execute(sqlalchemy.text("SELECT NOW()"))
            print(result.fetchone())

        await pool.dispose()

asyncio.run(main())

Prerequisites

1. Enable the AlloyDB API in your Google Cloud project: console.cloud.google.com/apis/api/alloydb.googleapis.com

2. Grant IAM permissions: your principal needs the AlloyDB Client role (or equivalent) on the instance.

3. Set up credentials using Application Default Credentials (ADC):

   gcloud auth application-default login

Note: By default the connector uses private IP. Run your code from an environment connected to the VPC that hosts your AlloyDB instance, or see Configuring AlloyDB Connectivity for other options.

Configuration

Connector Lifecycle

Create one connector per application and reuse it for the lifetime of the process. Each connector maintains a background refresh cycle that keeps connection credentials warm, so creating one per request would waste resources and cause unnecessary latency.

The recommended approach is a context manager — close() is called automatically, even if an exception is raised:

# Sync
with Connector() as connector:
    ...

# Async
async with AsyncConnector() as connector:
    ...

For long-lived applications (e.g. a web server) where the connector outlives any single block, call close() explicitly at shutdown:

# Sync
connector = Connector()
...
connector.close()

# Async
connector = AsyncConnector()
...
await connector.close()

IP Address Type

Connect over private IP (default), public IP, or Private Service Connect (PSC):

# At the Connector level (applies to all connections)
connector = Connector(ip_type="PUBLIC")

# Or per connection
connector.connect(INSTANCE_URI, "pg8000", ..., ip_type="PSC")

Valid values: "PRIVATE" (default), "PUBLIC", "PSC".

IAM Database Authentication

Skip the password and authenticate using your IAM identity instead. First, enable IAM auth on your instance and create an IAM database user.

connector.connect(
    INSTANCE_URI,
    "pg8000",  # or "psycopg" (sync) / "asyncpg" (async)
    user="service-account@my-project.iam",  # omit .gserviceaccount.com suffix
    db="my-db",
    enable_iam_auth=True,
)

For IAM user accounts, use the full email address as user.

Lazy Refresh (Cloud Run, Cloud Functions)

In serverless environments where CPU may be throttled between requests, use refresh_strategy="lazy" to fetch connection info on demand instead of running a background refresh cycle:

connector = Connector(refresh_strategy="lazy")

Debug Logging

import logging

logging.basicConfig(format="%(asctime)s [%(levelname)s]: %(message)s")
logging.getLogger("google.cloud.alloydbconnector").setLevel(logging.DEBUG)

Import Paths

This package supports two equivalent import paths:

from google.cloud.alloydbconnector import Connector  # preferred
from google.cloud.alloydb.connector import Connector  # also supported

The first is preferred to avoid namespace collisions with the google-cloud-alloydb package.

Support Policy

Major Version Lifecycle

This project uses semantic versioning:

• Active — receives all new features and security fixes. New major versions are guaranteed active for a minimum of 1 year.
• Deprecated — receives security and critical bug fixes only, for 1 year after deprecation.
• Unsupported — any major version deprecated for ≥1 year.

Supported Python Versions

Follows the Python Version Support Policy used by Google Cloud Libraries for Python. Changes in supported Python versions are treated as minor changes and listed in the release notes.

Release Cadence

This project targets a minimum monthly release cadence. If no new features or fixes have been added, a new PATCH version with the latest dependencies is released.

Contributing

We welcome outside contributions. Please see our Contributing Guide for details on how best to contribute.