DataTerminology.comNEW
Modern architecture that unifies data lake storage with warehouse-like governance and performance for analytics and AI.
DESCRIPTION
Lakehouse Architecture integrates open data lake storage with transactional consistency, governance, and SQL performance typical of warehouses. It relies on open table formats, ACID storage layers, and decoupled compute, enabling BI and ML on the same data. Common implementations use Delta Lake, Apache Iceberg, or Apache Hudi over cloud object stores. The approach reduces data duplication, supports schema evolution, and simplifies data management. It is widely adopted for flexibility, cost efficiency, and interoperability across tools.
ALTERNATIVE DEFINITIONS
Databricks
A lakehouse is a new open architecture that combines the best elements of data lakes and data warehouses.
Emphasizes openness and combination of lake and warehouse capabilities.
Dremio
A data lakehouse is an open data management architecture that combines the flexibility of data lakes with the data management features and performance of data warehouses.
Highlights openness and management features on lake storage.
Snowflake
A data lakehouse is a modern architecture that offers the scalability of data lakes and the robust data management and performance of data warehouses.
Positions lakehouse as combining scalability and robust management/performance.
Wikipedia
A data lakehouse is an architecture that provides data lake features, such as low-cost storage of large amounts of data, together with data warehouse features such as schema management and ACID transactions.
Mentions ACID transactions and schema management on lake storage.
RELATED TERMS
Delta Lake Apache Iceberg Apache Hudi Parquet ELT Medallion Architecture Decentralized Data Architecture SQL Analytics
COMPETING TERMS
KNOWLEDGE AREAS
Data Architecture Data Engineering Data Warehousing & BI Cloud Platform & Tooling
TARGET AUDIENCE
Data Architect Data Engineer Analytics Engineer BI Specialist Cloud Architect Chief Data Officer
EXTERNAL SOURCES
- Wikipedia - Data lakehouse
- Delta Lake - Open-source ACID lake storage layer
- Apache Iceberg - Open table format for lakes
- Apache Hudi - Streaming data lake platform
- Databricks - Lakehouse architecture overview
TOP QUESTIONS
Why is Lakehouse Architecture gaining adoption?
It unifies BI and ML on the same data, reduces duplication, and leverages open formats and cloud object storage for scalability and cost efficiency.
Category: Relevancy
What are core components of a Lakehouse Architecture?
Cloud object storage, open table formats, ACID storage layers, metadata catalogs, and compute engines supporting SQL and ML workloads.
Category: Definition
How does Lakehouse Architecture compare to a Data Warehouse?
Lakehouses keep raw and curated data together with open storage and decoupled compute, while warehouses focus on curated structured data with tightly integrated engines.
Category: Alternatives
What is a common governance approach in Lakehouse Architecture?
Use catalogs, data contracts, row-level security, and lineage, often structured in medallion layers: bronze, silver, and gold.
Category: Management
What are typical risks of Lakehouse Architecture?
Poor table format choice, weak governance, and unmanaged schema evolution can lead to performance issues and data inconsistency.
Category: Risks
How does Lakehouse Architecture handle streaming data?
Streaming writers with transactional table formats (e.g., Delta, Hudi) support upserts, incremental processing, and time travel.
Category: Example
What performance features matter in Lakehouse Architecture?
Data skipping, compaction, Z-ordering, partitioning, caching, and vectorized execution improve query speed and cost.
Category: Requirements
What is the biggest misunderstanding about Lakehouse Architecture?
Some think it’s just rebranding; it materializes with open table formats, transactional lakes, and warehouse-grade governance.
Category: Misunderstanding
Where to start with Lakehouse Architecture?
Pilot a small domain, choose a table format, establish governance and modeling standards, then scale ingestion and consumption patterns.
Category: HowTo