Skip to main content
Technology areas
close
AI and ML
Application development
Application hosting
Compute
Data analytics and pipelines
Databases
Distributed, hybrid, and multicloud
Industry solutions
Migration
Networking
Observability and monitoring
Security
Storage
Cross-product tools
close
Access and resources management
Costs and usage management
Infrastructure as code
SDK, languages, frameworks, and tools
/
Console
English
Deutsch
Español
Español – América Latina
Français
Indonesia
Italiano
Português
Português – Brasil
עברית
中文 – 简体
中文 – 繁體
日本語
한국어
Sign in
AlloyDB for PostgreSQL
Start free
Overview
Guides
Reference
Resources
Technology areas
More
Overview
Guides
Reference
Resources
Cross-product tools
More
Console
Discover
Product overview
What's new in documentation
Security, privacy, risk, and compliance overview
What is AlloyDB AI?
Explore AlloyDB AI use cases and examples
Get started
Try it out
Free trial clusters
Free trial clusters overview
Create a free trial cluster
Create and query a database
Connect from Compute Engine
Connect from Cloud Shell
Perform a vector search
Plan and prepare
Choose a machine type
Choose a connectivity option
Create and manage
Create a cluster and its primary instance
Create and manage a database
Manage clusters
View the list of clusters
View the list of instances in a cluster
View cluster details
Delete a cluster
Create and manage instances
Create a read pool instance
Create a primary instance
View instance details
Configure database flags
Configure database extensions
Scale an instance
Start, stop, or restart instances
Delete an instance
Manage instance high availability
High availability overview
Fail over a primary or secondary instance manually
Test a primary instance for high availability
Reduce costs using basic instances
Manage cross-region replication
Overview
Work with cross-region replication
Organize resources using tags
Tags overview
Attach and manage tags
Manage your resources using Knowledge Catalog
Connect
Overview
Install psql
Connect using gcloud
Connect from your laptop
Connect from Cloud Run
Connect from Google Kubernetes Engine
Connect to an instance using public IP
Connect to an instance using private IP
Overview
Set up private services access
Overview
Enable private services access
Increase private services access address space
Remove a private services access allocated range
Set up Private Service Connect
Overview
Connect to an instance using Private Service Connect
Connect securely using the AlloyDB Auth Proxy
Overview
Connect using the Auth Proxy
Best practices for using the Auth Proxy
Troubleshoot Auth Proxy connections
AlloyDB language connectors
Overview
Connect using the AlloyDB language connectors
Connect to a cluster from outside a VPC
Configure managed connection pooling
Manage database connections with write endpoints
Develop database applications
Use the remote MCP server
Use saved queries
Overview
Create and manage saved queries
Access BigQuery data from AlloyDB
Choose how to access BigQuery data
Access to real-time data overview
Configure access to real-time data
Data synchronization overview
Sync BigQuery data to AlloyDB
External search
Access Elasticsearch data
Access Solr data
Access OpenSearch data
Full-text search
Full-text search overview
Create and manage a RUM index
Create and manage a BM25 index
Develop AI and ML applications
Build generative AI applications using AlloyDB AI
Configure user access to Agent Platform models
Natural language and agents
Conversational analytics
Natural language
Natural language overview
Generate SQL statements from natural language
Use natural language to generate SQL
Query database in natural language with QueryData
AI-powered search
Vector indexing
Vector indexing overview
Run a vector similarity search
Store vector embeddings
Create and manage vector indexes
Choose a vector index
Create indexes
Create a ScaNN index
Create an HNSW index
Create an IVF index
Create an IVFFLAT index
Tune vector query performance
Vector assist
Vector assist overview
Use vector assist
AlloyDB AI functions
AlloyDB AI functions overview
Query using AI powered SQL operators
Rank search results
Accelerate queries using optimized functions
Analyze text sentiment
Summarize content
Hybrid search
Run a hybrid vector similarity search
Inference
Forecasting with time series models
Perform time-series forecasting
Call models using model endpoints
Overview
Register a model
Manage model endpoints using AlloyDB Studio
Generate embeddings
Generate text embeddings
Generate multimodal embeddings
Generate and manage auto vector embeddings for large tables
Invoke predictions
Build realtime vector embedding pipeline for AlloyDB with Dataflow
Build AI applications using frameworks
Build LLM-powered applications using LangChain
Build LLM-powered applications using LlamaIndex
Use AlloyDB with agents
Best practices for securing agent interactions with MCP
Write SQL with Gemini assistance
Secure and control access
Grant AlloyDB access to other users
Use Secret Manager to handle secrets
Secure access
Secure connections using SSL mode
Control connector enforcement
Secure data with customer-managed encryption keys (CMEK)
Overview
Use CMEK
Parameterized secure views
Overview
Manage application data security
Use parameterized secure views
Organization policies
Overview
Use predefined organization policies
Use custom organization policies
Secure resources and data using VPC Service Controls
Manage database users
Overview
Built-in database authentication
IAM database authentication overview
Manage PostgreSQL users with built-in authentication
Manage IAM authentication
Manage password policies
Connect using an IAM account
Enforce data residency requirements
Best practices for securing your resources
Maintain
Manage maintenance updates
Overview
Manage cluster maintenance windows
Perform self-service maintenance
Manage AlloyDB maintenance updates using a staging cluster
Upgrade a major database version
Upgrade a cluster's major server version
Upgrade in place
Overview
Perform an in-place major version database upgrade
Maintenance changelogs
Overview
PostgreSQL 14 maintenance changelog
PostgreSQL 15 maintenance changelog
PostgreSQL 16 maintenance changelog
PostgreSQL 17 maintenance changelog
PostgreSQL 18 maintenance changelog
Migrate
Migration overview
Migrate using Database Migration Service
Database Migration Service overview
Use quick-start migrations
Migrate from Cloud SQL for PostgreSQL to AlloyDB
Migrate data from a vector database to AlloyDB
Back up and restore
Overview
Manage enhanced backups
Manage standard backups
Configure backup plans
Create an on-demand backup
View a list of backups
View backup details
Delete a backup
Cancel an in-progress backup
Restore a cluster
Use point-in-time recovery (PITR)
Restore from a backup
Import and export
Best practices for importing and exporting data
Import data
Import a CSV file
Import a DMP file
Import a SQL file
Cancel an import operation
Export data
Export a DMP file
Export a CSV file
Export a SQL file
Cancel an export operation
Monitor and optimize
Monitor and troubleshoot with AI assistance
Monitor the health of your clusters and instances
Sizing and deployment recommendations
Instance performance
Monitor instances
Database performance
Monitor and analyze query performance
About query insights
Improve query performance using query insights
Advanced query insights overview
Improve query performance using advanced query insights
Monitor active queries
Get and analyze execution plans
Analyze performance using pgbench
Benchmark OLTP performance on AlloyDB
Improve query performance
Index advisor overview
Use the index advisor
Use the index advisor with query insights
Optimize database performance by comparing snapshots
Improve INSERT SELECT performance with parallel SELECT
Find and fix inconsistent indexes
Find and fix foreign key violations
Create and manage named hints
Measure and address table and index bloat
Enable cache-aware costs for index scan query plans
Manage query plans
Optimize resources and isolate read queries with transparent query forwarding
Accelerate analytical queries using the columnar engine
About the columnar engine
Configure the columnar engine
Manage column store content using auto-columnarization
Manage column store content manually
Optimize queries with virtual columns for expressions
Monitor the columnar engine
Tune the columnar engine
Maintain freshness of column store data
Configure adaptive autovacuum
Monitor AlloyDB using the Database Insights MCP server
Apply recommendations
Enable backup plans
Enable database auditing
Enable high availability
Enable password policy
Enforce SSL/TLS encryption
Increase backup retention
Increase cluster storage quota
Optimize underprovisioned instances
Measure vector query recall
Optimize vector query performance
ScaNN Vector query performance overview
Best practices for tuning ScaNN indexes
Accelerate vector search with the columnar engine
Filtered vector search performance
Filtered vector search overview
Understand adaptive filtering
Activate adaptive filtering
Maintain vector indexes
Accelerate performance for queries with AI functions
Monitor access using audit logs
Monitor session and object access using pgAudit
About pgAudit
Enable pgAudit
Configure logging behavior
View audit logs
Disable pgAudit
Troubleshoot
Troubleshoot ScaNN indexes in the columnar engine
Troubleshoot backups and restores
Troubleshoot read pool queries
Troubleshoot replication issues
Troubleshoot ScaNN index errors
Troubleshoot auto embedding issues
Troubleshoot optimized functions
Troubleshoot an in-place major version upgrade
Database in-place major version upgrade errors
Troubleshoot network configuration errors
Troubleshoot AlloyDB maintenance
Troubleshoot the columnar engine
Troubleshoot issues with the BigQuery foreign data wrapper extension
AlloyDB Omni
