Starburst Data Platform

Overview

Starburst can query more than 50 sources in one SQL statement without copying their data. Its connectors span data lakes, warehouses, streaming systems, relational databases, and other sources, making the platform relevant to teams whose data is spread across different systems. Federation reduces the need to make copies for analysis; it does not replace the systems that store the data.

The platform combines SQL analytics, ACID transactions, streaming ingestion, and a governance catalog. Connector features including parallelism, table statistics, dynamic filtering, and pushdown are intended to improve query execution. Their practical value depends on whether the connectors suit the sources a team needs to query.

Key features

Deployment and governance

Starburst supports on-premises, cloud, and hybrid environments. Its Galaxy service is available on AWS, Google Cloud, and Microsoft Azure, which gives cloud users several deployment choices while leaving room for organizations that need on-premises or hybrid setups.

Galaxy supports SSO, SCIM, service principals, role- and attribute-based access control, row filters, column masking, and end-to-end encryption. Starburst says Galaxy does not store customer data. It also says it independently verifies security, privacy, and compliance controls; its security page displays HIPAA and GDPR badges. These controls suit organizations with governance requirements, though the range of controls does not by itself establish that a particular deployment meets an organization’s obligations.

Galaxy lists a 99.5% uptime SLA, 24/7 expert support, and a 30-minute response time. Those commitments may matter to teams operating business-critical workloads; the listed uptime guarantee is not the highest level offered by Starburst’s plans.

Pricing

Starburst is freemium. The free plan is 0.00 USD per free, billed Free forever, and allows up to three clusters with standard cluster execution mode for ad hoc queries. That is a useful entry point for limited or exploratory work, but the cluster cap and ad hoc execution mode constrain its fit for broader production needs.

The 30-day trial includes $500 in Galaxy compute resources and Enterprise features. At the end of the trial, the account downgrades to the free tier. The trial gives teams a defined window and resource allowance to evaluate the service, but it is not an ongoing substitute for a paid plan.

Pro is the step up for teams that need flexible execution, ingestion, or cluster management. Enterprise adds autoscaling and more advanced access controls, while Mission-Critical is aimed at teams that need stronger support and governance integrations. All three paid prices are starting rates per credit, rather than a fixed total, so expected credit usage matters when comparing plans. AIDA token usage is billed separately on Enterprise and Mission-Critical.

Platforms

Starburst supports API, Linux, self-hosted, and web environments. Galaxy’s availability on AWS, Google Cloud, and Microsoft Azure gives cloud teams a choice among those providers; the wider platform also supports on-premises and hybrid deployment.

Who it's for

Starburst is a strong fit for organizations that need governed SQL access to data across multiple sources and deployment environments. Its feature set and plan tiers address needs ranging from ad hoc queries to advanced autoscaling, governance integrations, and support. Smaller teams with limited query needs can start on Free, while organizations with substantial usage or operational requirements should weigh credit consumption and the differences between paid tiers. It is less suited to buyers seeking a simple fixed-price service or a free plan without a cluster cap.

Pros and cons

• Pros: Query federation across 50+ sources without copying data can simplify analysis across a varied data estate.
• Pros: On-premises, cloud, and hybrid deployment options suit organizations with mixed infrastructure.
• Pros: Galaxy combines access controls, masking, encryption, and stated support and uptime commitments.
• Cons: Free is limited to three clusters and standard execution for ad hoc queries.
• Cons: Paid plans start at per-credit rates, making costs dependent on usage rather than a fixed subscription total.
• Cons: AIDA token usage is an additional charge on Enterprise and Mission-Critical.

Alternatives

For a broader comparison, browse Data Lakehouse Platforms.

• Oracle Autonomous AI Lakehouse is another freemium option with an always-free plan for an unlimited time, subject to capacity limits; consider it if that free usage model is a priority.
• Starburst Galaxy is a related freemium option with a free tier capped at three clusters and paid plans starting at $0.50 per credit; consider it if you are evaluating the Galaxy offering specifically.
• Amazon SageMaker Autopilot is a paid service with a free trial and pay-as-you-go pricing; consider it if that pricing approach better suits your use case.
• Apache Hudi is a free, open-source data lakehouse platform for Linux and self-hosted use; consider it if those deployment options and open-source availability are priorities.
• Bauplan offers a free Sandbox with a shared environment, public datasets or uploads, full CLI, SDK and API, and community support on Slack; consider it if that sandbox model matches your needs.
• Databricks Notebooks has a free edition with one serverless workspace and limited compute size and usage; consider it if a notebook workspace is the requirement.
• Google Cloud Lakehouse for Apache Iceberg offers a metadata storage free tier of 1 GiB per month; consider it if that specific metadata storage allowance is useful.
• IOMETE has a free self-hosted, on-premises plan capped at 100 vCPUs with core features and community support; consider it if those deployment and capacity terms fit.

Verdict

Choose Starburst if your organization needs governed SQL queries across distributed data and wants the flexibility to operate across cloud, on-premises, or hybrid environments. Federation across 50+ sources is its clearest reason to choose it; credit-based paid pricing and the free plan’s three-cluster cap are the main reasons to look elsewhere.