Citus Use Cases

Delivering Powerful Analytics on Real-time Data

Citus delivers value in a variety of use cases across multiple industries. These range from customer facing real-time dashboards in ad-tech companies or SaaS analytics providers to behavioral analytics in mobile / web analytics platforms to real-time geospatial analytics and operational analytics in retailers or telco operators. Citus makes all of these use cases available by providing:

  • Seamless horizontal scaling by sharding your data across multiple Postgres hosts
  • Intelligently parallelizing your queries to distribute workloads
  • Multiple executor types to optimize for your workload
  • Integrating with native PostgreSQL tools across the PostgreSQL ecosystem.

Scaling multi-tenant SaaS applications

Citus makes it easy for you to continue scaling your data for a multi-tenant application by seamlessly sharding your data. Because Citus extends Postgres you do not have to pull large tables out or entirely re-architect to continue scaling when you start to outgrow a single node which can happen as early as 100 GB.

With Citus your entire dataset can easily be split up across multiple physical machines allowing you to scale memory and compute, and your tenants still reside on the same node. This provides performant joins and the full breadth of SQL when interacting with a single tenant as most B2B applications do.

  • Scale to millions of tenants
  • Full transaction support when interacting with a single tenant
  • Full SQL coverage when querying a single tenant

Real-time Analytics Dashboards

Citus combines fast reads and writes with powerful analytics in one database which allows you to implement a cost effective architecture to power real-time dashboards. You can provide insights which include both your real-time data and your historical data.

Citus has a high data ingestion rate so you can ingest terabytes of data every day, in real-time as your data is generated. Citus uses parallel query processing to provide very fast query responses, which enables real-time insights on real-time data.

  • Optimized for event/time-series data
  • Sub-second queries with parallelism
  • Real-time data ingest
  • High concurrency
For further reading take a look at how Cloudflare provides real-time dashboards to their customers across terabytes of data.

Behavioral Analytics

Citus enables interactive cohort and funnel analytics on real-time event data at scale. It can serve instant responses to concurrent queries with on-the-fly segmentation and funnel step definitions.

It simplifies and speeds up behavioral analytics through use of advanced analytics functions built on PostgreSQL’s powerful expressiveness. Citus also allows you to update your data in real-time so your end users always see fresh insights into your current data.

  • Optimized for clickstream, users, events data
  • Advanced funnel, cohort and session analytics libraries
  • Real-time data ingest
For further insights listen to how Heap powers behavioral analytics for thousands of businesses:

Operational Analytics

Citus enables you to scale out real-time operational analytics with the power of a relational database.

You can combine powerful analytics with fast inserts/updates for reduced complexity and cost while utilizing powerful and familiar PostgreSQL expressiveness and tools.

  • Hybrid operational and analytics workloads
  • Horizontal scalability on commodity hardware
  • Standard PostgreSQL drivers and tools
For further reading take a look at how MixRank deployed a consolidated analytics platform to support B2B sales teams with real-time insights generated from 160TB of data.

Fast geospatial analytics

Citus allows you to utilize the power of hundreds or even thousands of CPU cores to analyze your geospatial data with human real-time responsiveness.

As your data grows, you can horizontally scale out on commodity servers and achieve analytics performance that is orders of magnitude higher than what is possible with single threaded geospatial databases today.

  • Optimized for geofencing
  • Support for PostGIS
  • Parallelized query processing
Watch how a large retailer analyzes large geospatial data from sensors in its truck fleet with high degrees of flexibility and responsiveness: