DataTerminology.comNEW
An integrated set of services to ingest, store, process, govern, and serve data for analytics and applications.
DESCRIPTION
A data platform provides core capabilities for data ingestion, storage, transformation, governance, and consumption. It unifies batch and streaming, supports multi-tenant access, and enforces policies. Modern platforms favor cloud-managed services, scalable compute, and declarative workflows. They expose self-service interfaces for analysts, scientists, and engineers. Lakehouse architectures increasingly combine warehousing performance with flexible data lake storage under unified governance.
ALTERNATIVE DEFINITIONS
Google Cloud
This guide describes the end-to-end data lifecycle on Google Cloud from ingestion, storage, processing, and analytics to visualization.
Outlines platform capabilities across the full lifecycle.
Microsoft Azure Architecture Center
The Azure Data Architecture Guide presents a structured approach for designing end-to-end data platforms and analytics solutions on Azure.
Presents a structured approach to designing data platforms.
Databricks
A data platform is a set of integrated technologies that collect, store, process and analyze data.
Concise definition emphasizing integrated technologies and capabilities.
RELATED TERMS
Data Lake Data Warehouse Data Lakehouse Orchestration Feature Store Data Discovery Platform Streaming Platform Governance Extract-Transform-Load ELT
COMPETING TERMS
Best-of-breed Toolchain
RELATIONS
- Narrower: Customer Data Platform
KNOWLEDGE AREAS
Platform & Tooling Data Engineering Cloud Data Architecture Data Services
TARGET AUDIENCE
Data Engineer Data Architect Cloud Architect MLOps Engineer DevOps Engineer Chief Data Officer
EXTERNAL SOURCES
- Wikipedia - Platform (computing)
- Microsoft - Designing data platforms on Azure
- Google Cloud - Architectures for data and analytics
- AWS - AWS data lake reference solution
- Databricks - Lakehouse platform perspective
TOP QUESTIONS
What capabilities should a modern Data Platform include?
Ingestion, storage, transformation, orchestration, governance, lineage, observability, and serving for BI/ML are table stakes.
Category: Requirements
How does a Data Platform differ from a Data Warehouse?
A platform is broader, encompassing ingestion, processing, governance, and multiple storage paradigms including warehouses and lakes.
Category: Misunderstanding
What is a Lakehouse in the Data Platform context?
A Lakehouse unifies lake storage with warehouse performance under one governance model.
Category: Example
How do you measure Data Platform success?
Developer productivity, time-to-insight, reliability SLOs, cost efficiency, and adoption across domains indicate success.
Category: Metrics
What is a common mistake when building a Data Platform?
Over-customizing core components before establishing governance, standards, and self-service abstractions.
Category: Mistakes
How should multi-tenancy be handled?
Use isolation, quota management, governance guardrails, and cost allocation by domain or product.
Category: Best practices
What role does orchestration play?
It coordinates data flows, dependencies, and SLAs across batch and streaming workloads.
Category: Relationship
How does a Data Platform support ML?
Feature stores, experiment tracking, model registry, and model serving integrate with core data services.
Category: Application
What is the trade-off between platform unification and best-of-breed?
Unification simplifies governance and operations; best-of-breed may optimize specific workloads but increases integration complexity.
Category: Opinion