Search and analytics at cloud scale

Rockset is a real-time search and analytics database designed to serve millisecond-latency analytical queries on event streams, CDC streams, and vectors

Hero ImageHero Image

Cloud-native efficiency

Compute-compute diagram
Compute-compute separation
Compute-storage separation
Autoscaling in the cloud
Fully-managed cloud service
Rockset isolates streaming ingest compute from query compute for predictable performance even in the face of high-volume writes or reads. Avoid compute contention, overprovisioning, and replicas.
Read our architecture whitepaper ->
Compute-compute diagram

Real-time streaming ingest

Schemaless Ingest

Rockset automatically and continuously infers schema based on fields and types, even if types are mixed. Ingest semi-structured data or nested objects and arrays and execute relational SQL queries over these constructs.

Built-In Connectors

Ingest data with native connectors for streams, databases, data lakes, and cloud data warehouses. Ingest from anywhere using the write API or CDC templates. Ingest millions of events per second and query within 2 seconds. No ETL tools required.

Ingest Transformations

Pre-aggregate and transform data at ingest time using SQL. Reduce the cost of storing and querying high-volume streaming data by up to 100x, and eliminate the need for separate pipelines to process streaming data.

Real-Time Speed

Ingest streaming data, index it, and query it with end to end latency less than a second.

In-Place Updates

Insert, update, and delete data in place without expensive merge operations. Avoid the slow, error prone workarounds for append only databases.
Data Sources

Millisecond-latency search & analytics

Converged Index

Rockset stores all data in a Converged Index, which combines row, column and inverted indexes. This enables fast, compute-efficient queries, regardless of the access pattern or shape of the data.

Full-Featured SQL

Run SQL search, aggregations and joins directly on semi-structured data. Rockset uses SQL as its native query language.

High-Performance Joins

Rockset supports multiple strategies to optimize join performance, so users do not have to denormalize data or perform application-side joins.

Vector Search

Fast KNN search with metadata filters to deliver relevant results with just milliseconds of latency.
Image of a query in a console

Developer productivity

Query Lambdas

Query Lambdas are named, parameterized SQL queries that can be executed from dedicated REST endpoints. Enforce version control and integrate them into CI/CD workflows.

Resource Automation

Automatically deploy Rockset integrations, collections, resources, and third-party dependencies using the Rockset Terraform Provider. Use the dbt adapter to load data and create collections using SQL SELECT statements.


JavaScript user-defined functions (UDFs) can perform operations beyond built-in SQL functions. They run in an isolated, sandboxed environment and provide support for advanced math functions, transpositions, and more.


Rockset has client libraries for Node.js, Java, Go, Python, and the Rockset CLI that wrap the Rockset API. Client libraries can be used to programmatically insert, update, and query data from your application's code.
Image of code in a consoleImage of code in a console

Production ready

Image of latency graphs


  • Continuous backups with data stored on SSDs for performance and cloud storage for durability¬† Disaster recovery with
  • Multi-region hot-cold and hot-hot deployment options
  • Enterprise SLAs


  • SOC 2 Type II
  • GDPR
  • CCPA


  • Single Sign On with Okta and custom SSO
  • Data Masking
  • Data encrypted at rest and in transit
  • Role Based Access Controls (RBAC)
  • IP Allowlisting
  • AWS Privatelink
  • Virtual Private Rockset in your VPC


Performance metric endpoint