Databricks: Features, Pricing and Alternatives (2026)

Vendor

Databricks, Inc.

Price

Freemium

Platforms

AWS, Azure, Google Cloud, SaaS

Licence

Commercial platform, usage-based (DBUs); built on open source projects such as Apache Spark, Delta Lake and MLflow

Works with

AWSMicrosoft AzureGoogle CloudDelta LakeApache IcebergParquetApache SparkMLflowPower BI (via Fabric mirroring)

At a glance

SQL support	Databricks SQL (ANSI mode) on SQL warehouses and Spark SQL on clusters
Storage and compute architecture	Open table format files in your cloud object storage; separate clusters, serverless compute and SQL warehouses
Workload types	Data engineering, streaming, SQL analytics and BI, machine learning and AI
Integrations	Delta Lake, Apache Iceberg, Spark, MLflow, Fabric mirroring of Unity Catalog, JDBC and ODBC
Deployment	Managed platform on AWS, Azure (Azure Databricks) and Google Cloud; classic compute in your own cloud account
Pricing model	DBUs per second by compute type, tier, cloud and region; cloud VM costs extra for classic compute

Short answer

Databricks suits teams whose data work centres on engineering, streaming and machine learning, who write Python and Spark as much as SQL, and who want tables to stay in open formats in their own storage. Its serverless SQL warehouses make it a workable BI back end too, but there are more compute types, tiers and cost lines to understand than in a warehouse-first service.

How we know: This profile is research-based: architecture, compute types, governance, billing model and free options were checked against Databricks' documentation and pricing page, Microsoft Learn and Microsoft's Azure Retail Prices API in November 2026. We have not run Databricks or measured its performance or cost.

Overview

Databricks describes its product as a Data Intelligence Platform built on the lakehouse model: data lives as files in cloud object storage, organised into tables by an open table format. Delta Lake, which Databricks originally developed, is the default format, and managed Iceberg tables are also supported. Unity Catalog governs tables, volumes, functions and models in a catalog.schema.object namespace.

Compute comes in several forms: all-purpose clusters for notebooks, job compute, Lakeflow pipelines, serverless compute and SQL warehouses (serverless, pro or classic) for SQL and BI. The platform runs on AWS, on Microsoft Azure as Azure Databricks (sold and billed by Microsoft) and on Google Cloud.

Key features

SQL warehouses in serverless, pro and classic types, with the Photon vectorised engine
Unity Catalog for privileges, row and column filters, attribute-based policies, lineage and auditing
Delta Lake tables plus managed Iceberg tables and an Iceberg REST Catalog API
Apache Spark in notebooks and jobs in Python, SQL, Scala and R
Lakeflow pipelines for batch and streaming ingestion and transformation
MLflow, Feature Store and Model Serving for the machine learning lifecycle
Serverless compute for notebooks, jobs and pipelines, run in Databricks' compute plane

Pros and cons

Pros

Spark, streaming, SQL and ML on the same governed tables
Data stays in open Delta Lake or Iceberg format in your own storage
Unity Catalog governs tables, files, functions and models with lineage
Choice of serverless compute or compute inside your own cloud network
Free Edition for learning, plus a 14-day trial of the full platform

Cons

Many compute types, tiers and runtimes to choose and govern
For classic compute, cost is split between the Databricks bill and the cloud provider bill
Per-DBU list prices are not shown as static values on the Databricks pricing page
Free Edition cannot be used commercially and is limited to serverless compute and one 2X-Small SQL warehouse

Pricing

Databricks bills compute in Databricks Units (DBUs) per second, pay-as-you-go or through committed-use contracts. The DBU rate depends on the product and compute type (jobs, all-purpose, SQL classic, pro or serverless, and others), the tier, and the cloud and region. With classic compute the virtual machines run in your own cloud account and your cloud provider bills them separately; for serverless compute Databricks runs the compute and the DBU charge covers it. Databricks' own price tables load dynamically, so we could not quote them. Azure Databricks prices are set by Microsoft: the Azure Retail Prices API listed Premium rates in East US of USD 0.70 per DBU-hour for Serverless SQL, USD 0.55 for all-purpose compute and USD 0.30 for jobs compute (November 2026). Microsoft Learn states that new Azure Databricks workspaces must use the Premium tier since 1 April 2026 and that remaining Standard tier workspaces are being upgraded to Premium in a phased rollout from 1 October 2026, so plan on Premium rates there. A 14-day free trial is available, and Free Edition is free but limited and may not be used commercially.

Pricing checked on the vendor's official page on 7 November 2026. Prices change; confirm before buying.

Who it is best for

Data engineering and data science teams that want Spark, streaming, SQL and ML on open-format tables, on AWS, Azure or Google Cloud.

Who should look elsewhere

Analyst-led teams that only need managed SQL warehousing with the least to configure: see Snowflake vs Databricks and Databricks vs BigQuery.
Microsoft and Power BI shops that prefer one SaaS capacity: see Databricks vs Microsoft Fabric.

Verdict

Our take

Databricks is the more direct fit when engineering and ML sit at the centre of your data work and open formats matter. It asks more of you in configuration and cost tracking than a warehouse-first service, so price a representative pipeline and dashboard workload, including cloud charges for classic compute, before deciding.

Frequently asked questions

Is Databricks a data warehouse?

Databricks calls itself a lakehouse platform, but Databricks SQL provides serverless, pro and classic SQL warehouses for SQL and BI. Tables are stored as Delta Lake or Iceberg files in cloud object storage and governed by Unity Catalog.

What is a DBU?

A Databricks Unit is a normalised unit of processing power billed per second. The rate depends on the compute type, tier, cloud and region. With classic compute your cloud provider also bills the virtual machines; with serverless compute the DBU charge covers the compute.

Is Databricks Free Edition free for work?

No. Databricks states that Free Edition accounts may not be used for commercial purposes. It is serverless only, with limits such as one 2X-Small SQL warehouse and one workspace. For commercial evaluation there is a 14-day free trial.

Sources

Checked November 2026.

How we research reviews and what star ratings mean: our editorial method.