Skip to content
Databricks logo

Databricks

Infraestructura de Datos · www.databricks.com

¿Es tu herramienta? Reclama esta ficha →

Resumen

Databricks is a cloud-based data and AI platform built by the original creators of Apache Spark, Delta Lake, MLflow, and Unity Catalog. It combines data engineering, SQL analytics, machine learning, and generative AI on a single "lakehouse" architecture that runs on AWS, Azure, or GCP.

El problema que resuelve Databricks

Data teams building analytics and AI products typically end up assembling a separate data warehouse, a Spark cluster, an ML platform, and a BI connector layer, each with its own security model and copies of the data. Databricks solves that by putting data engineering pipelines, SQL analytics, governance, and ML/AI model training and serving on one lakehouse, so teams work against a single governed copy of the data instead of reconciling it across systems.

Contexto para decidir

Usa estos puntos para comprobar si el producto encaja con tu operación, no solo con la lista de funciones.

  • Precio de entrada publicado: Free. Confirma límites de usuarios, uso y funciones por plan.
  • Despliegue: cloud. Comprueba requisitos de seguridad, residencia de datos y acceso para todos los equipos que lo utilizarán.
  • Integraciones verificadas: Fivetran, dbt, Alation, Microsoft Power BI, Tableau, Rivery. Valida el sentido de sincronización y los límites del plan elegido.
  • La ficha se comprobó por última vez el 31/7/2026; los precios y las funciones pueden cambiar.

Cómo evaluar Databricks

Una ficha ayuda a crear una lista corta; una prueba con el flujo real del equipo decide si la herramienta encaja. Usa esta lectura junto con los datos estructurados y confirma cualquier cambio con el proveedor.

Encaje de flujo

El registro la considera especialmente adecuada para Data engineering and data platform teams building large-scale ETL/ELT pipelines, Organizations wanting a single governed platform for both BI/SQL analytics and ML/AI, Enterprises with dedicated cloud/data infrastructure teams to manage usage-based costs. Comprueba que ese contexto coincide con el volumen, los roles y los procesos que debe soportar tu equipo.

Preguntas del piloto

  • ¿Puede Databricks completar el flujo crítico sin trabajo manual fuera de la herramienta?
  • ¿Las conexiones registradas (Fivetran, dbt, Alation, Microsoft Power BI) cubren el sentido de sincronización, los permisos y el volumen que necesitamos?
  • ¿Qué límites de usuarios, uso, almacenamiento, soporte o seguridad aparecen después del precio inicial?

Evidencia y vigencia

Esta ficha se comprobó el 31/7/2026. La fecha indica cuándo se revisó el registro, no una garantía de que el proveedor no haya cambiado sus condiciones después.

Ideal para

  • Data engineering and data platform teams building large-scale ETL/ELT pipelines
  • Organizations wanting a single governed platform for both BI/SQL analytics and ML/AI
  • Enterprises with dedicated cloud/data infrastructure teams to manage usage-based costs

No encaja si

  • Small teams or startups without data engineering expertise or budget to monitor consumption-based billing
  • Companies wanting a simple, flat-fee, fully self-hosted/on-premises deployment

Por qué está listada

  • Category-defining lakehouse platform combining data engineering, warehousing, and AI/ML in one system
  • Built by the original creators of Apache Spark, Delta Lake, MLflow, and Unity Catalog
  • Used by 20,000+ organizations including a majority of the Fortune 500

Precios

Free Edition

Free

Free access to core Databricks tools for individual learning, prototyping, and small-scale exploration.

  • Notebooks (SQL/Python/R)
  • Limited serverless compute
  • Unity Catalog basics
  • Community support

Standard compute (Jobs)

$0 per DBU

Pay-as-you-go pricing for automated Jobs compute, the lowest-cost general-purpose workload type.

  • Automated/scheduled workloads
  • Per-second billing
  • Autoscaling clusters
  • Delta Lake storage

SQL Serverless

$1 per DBU

Serverless SQL warehouses for BI dashboards and ad hoc analytics.

  • Serverless SQL warehouses
  • Instant elastic scaling
  • BI tool connectors
  • Query result caching

Enterprise

$0 per DBU (Jobs, Enterprise tier)

Adds enhanced security, compliance controls, and premium support on top of usage-based compute pricing.

  • Unity Catalog governance
  • HIPAA/PCI-eligible workspaces
  • Private connectivity (PrivateLink)
  • Premium support SLAs
  • Committed-use discounts

Funciones

Workflow automationDatabricks Workflows/Lakeflow Jobs natively orchestrate notebooks, pipelines, and ML tasks with dependencies and retries.
Pre-built connectorsNative ingestion via Auto Loader plus Partner Connect integrations, but not as broad a native connector catalog as dedicated iPaaS/ELT tools.
Data pipelines / ETLDelta Live Tables / Lakeflow Declarative Pipelines is a core product for building and managing ETL pipelines.
Scheduling & triggersWorkflows support cron scheduling, file-arrival triggers, and event-based triggers.
Data governance & lineageUnity Catalog provides centralized governance, lineage, and cataloging across all workspaces.
Role-based access controlUnity Catalog supports role-based, row/column-level, and attribute-based (ABAC) access control.
AI featuresMosaic AI, Databricks Assistant, AI/BI Genie, and Model Serving provide built-in generative AI and ML tooling.
Public APIExtensive REST APIs are documented for workspaces, jobs, clusters, and Unity Catalog objects.
Self-hosting / on-premSaaS-only; compute runs inside the customer's cloud account, but there is no on-premises or self-hosted deployment option.

Integraciones

FivetrandbtAlationMicrosoft Power BITableauRiveryLabelboxProphecyArcionConfluent (Apache Kafka)LookerQlikApache AirflowCollibraInformatica

Seguridad y cumplimiento

SOC 2 Type IIISO 27001HIPAAPCI DSSGDPR

Pros y contras

Pros

  • Strong at unifying data engineering, data science, and BI/analytics under one governed platform
  • Delta Lake and Unity Catalog provide robust data reliability, lineage, and access control
  • Scales well for very large datasets and complex ML/AI workloads
  • Native integrations across major BI tools (Tableau, Power BI, Looker) and ingestion tools (Fivetran, dbt)

Contras

  • Steep learning curve, especially for teams new to Spark or lakehouse concepts
  • Usage-based DBU pricing is complex and can scale unpredictably
  • Non-technical users find the interface and setup less approachable than simpler BI tools
  • Managing and forecasting compute cost requires ongoing FinOps discipline

Qué muestra el registro

4.6/5
658 reviews agregadasÚltima comprobación 2026-07-31

Reviewers praise Databricks for unifying data engineering, analytics, and ML/AI workflows, while citing a steep learning curve and unpredictable usage-based costs as the main drawbacks.

Resumen y puntuación agregados de plataformas públicas de reseñas. Enlazamos a las reseñas originales en lugar de reproducirlas — lee la fuente antes de decidir.

Reseñas de usuarios

Escritas por cuentas de Audyense · moderadas antes de publicarse

Todavía no hay reseñas de usuarios.

¿Has usado Databricks? Sé el primero en contarles a otros compradores qué funcionó de verdad.