• Postgres at Any Scale

    You can start with a single Citus node. When you need to scale, it’s simple to add nodes, rebalance shards, & distribute tables.

  • Parallelized Performance

    Speed up queries by 20x to 300x (or more) through parallelism, keeping more data in memory, higher I/O bandwidth, and columnar compression.

  • Postgres elephant outline icon
    True Postgres

    Citus is an extension (not a fork) to the latest Postgres versions, so you can use your familiar SQL toolset & leverage your Postgres expertise.

  • managed database service cloud icon
    Fully-Managed on Azure

    Focus on your application & forget about your database. Run your app on Citus in the cloud with Azure Cosmos DB for PostgreSQL.

  • Global Replication

    Multi-region read replicas give you globally scalable applications via low-latency reads across geographies & multi-region disaster recovery.

  • Perfect for Multi-Tenant SaaS

    Many SaaS apps—especially B2B—are multi tenant. So the apps have a natural dimension on which to distribute data: just shard by tenant_id.

Features in Citus on Azure

Azure Cosmos DB for PostgreSQL

What’s New

Citus 11.3

Citus 11.3 is out! Find the noisy neighbor in your cluster thanks to insights on CPU usage and query counts gained with tenant monitoring for SaaS applications.

PostgreSQL 15

Within 1 week of the PostgreSQL 15 release, we made PostgreSQL 15 generally available in Azure Cosmos DB for PostgreSQL in all Azure regions, for both new clusters and for upgrades from existing clusters.


Use the pgvector extension to enable the use of embeddings such as ones used in OpenAI. Store index and query embeddings to unlock full power of ML models for your private data stored in your Citus cluster.

Programmatic management

Now you can use REST APIs, Bicep, ARM templates, and Terraform to perform all cluster management operations such as cluster provisioning, compute and storage scaling, and/or add-in nodes.

Easy to Scale

Single node = basic tier

Basic tier has been renamed to “single node”. With a single Azure Cosmos DB for PostgreSQL node, you can use a distributed data model and shard Postgres from the start— and then easily add nodes later, when needed. Or use single node as a handy way to try out Citus in the cloud.

Sharding Postgres

Transparently shard Postgres tables across multiple nodes to give your application more memory, compute, and disk. Which gives you parallelism, high performance, and a way to scale. You can also start small and shard Postgres on a single Azure Cosmos DB for PostgreSQL node.

Easy to scale up nodes

Compute goes up to 104 vcores per node. It’s easy to increase or decrease compute/memory—and increase storage—on your Azure Cosmos DB for PostgreSQL coordinator & worker nodes.

Easy to add nodes

Easy to add nodes to your Azure Cosmos DB for PostgreSQL cluster with just a few clicks in the portal. Adding nodes is a completely online operation with zero application downtime.

Online shard rebalancer

When adding nodes to an Azure Cosmos DB for PostgreSQL cluster, the shard rebalancer enables you to redistribute shards across nodes to better balance the data distribution and the performance of your Citus database. Online and non-blocking.

Global read replicas

Read-only replicas of Azure Cosmos DB for PostgreSQL clusters are available globally, in any supported Azure region. Can be used for DR.

Managed PgBouncer

Outsource PgBouncer administration to Azure. Allow many clients to connect, while limiting the number of active connections and keeping the Azure Cosmos DB for PostgreSQL coordinator node running smoothly.

Managed Database Service

High Availability

High availability (HA) in Citus avoids database downtime by maintaining standby replicas for all nodes in the Azure Cosmos DB for PostgreSQL cluster, with streaming replication & automated failover.

Automated backup & restore

Automated backups are stored in multiple Azure availability zones. Point-in-time restore is available for any point in time in the last 35 days. Learn more about backup and restore in the Azure docs.

Geo-redundant DR

If something goes wrong with your primary cluster, or with an entire region, you can open a read replica for writes as part of your disaster recovery.

Automated monitoring & alerts

Automated monitoring, configurable metrics dashboard—plus alert setup through Azure Monitor just like with other Azure services.

Security controls

Security features in Azure Cosmos DB for PostgreSQL include: Encryption at rest & in flight, firewall rules, and 2FA. For networking, you can use public access with public IPs, private access with private endpoints—or enable both together.


Certifications achieved with Azure Cosmos DB for PostgreSQL: HITRUST, HIPAA, SOC 2 Type 2, ISO 27001, FedRAMP High, & more.


Pay-as-you-go pricing, or prepay with reserved capacity discounts. Learn more on the Azure Cosmos DB for PostgreSQL pricing page.

Read the docs for Citus on Azure, available as Azure Cosmos DB for PostgreSQL 

See What Citus on Azure Can Do

Watch a short demo on how to shard Postgres on a single node, so you can start small and seamlessly add nodes to your Citus cluster in the future. Umur also demos the Citus columnar feature, showing >5X compression.

Video thumbnail: UK Covid-19 dashboard / 24X compression demo

Getting Started with Citus on Azure

  • If you want to jump right in, use the free trial to try out Azure Cosmos DB for PostgreSQL.

  • Azure Cosmos DB for PostgreSQL is powered by Citus open source. The Citus ability to distribute tables enables you to build highly scalable relational apps, whether on a single node or a distributed cluster. This Quickstart shows you how to create an Azure Cosmos DB for PostgreSQL cluster via the Azure portal.

  • If you don’t yet have an Azure subscription, the first step is to create a free account. After you have an account and have read the docs, you can provision a cluster. Then you can choose your distribution key, run the create_distributed_table command, ingest your data, and assess how Azure Cosmos DB for PostgreSQL performs for your workload.