Data Lake
The pattern of storing raw, heterogeneous data in object storage for later processing. Data arrives in its original form and is transformed downstream.
Summary
The pattern of storing raw, heterogeneous data in object storage for later processing. Data arrives in its original form and is transformed downstream.
Data lakes are the precursor to lakehouses. In the S3 world, a data lake is the simplest form — dump everything into S3 and figure out the schema later. Lakehouses add the structure that data lakes lack.
- "Schema-on-read" does not mean "no schema." Without any schema management, data lakes become data swamps — undiscoverable and untrusted.
- Data lakes and lakehouses are not mutually exclusive. Most lakehouses include raw data lake zones (e.g., Medallion Bronze layer).
is_aObject Storage — a data lake is a use of object storagescoped_toS3 — S3 is the dominant storage layer for data lakes- Apache Spark
scoped_toData Lake — the primary compute engine for lake workloads - Apache Flink
scoped_toData Lake — streaming ingestion into lakes - Write-Audit-Publish
scoped_toData Lake — quality gating pattern for lake data
Definition
The pattern of storing raw, heterogeneous data in object storage for later processing. Data arrives in its original form and is transformed downstream.
Organizations needed a central, low-cost repository for all data types (structured, semi-structured, unstructured) without requiring schema decisions at write time.
Recent developments
- Lakehouse pattern is the 2026 default — pure data-lake architectures are getting reframed. Per a February 2026 industry survey, over 50% of data teams are now implementing lakehouse patterns. The lakehouse is the data-lake + table-format (Iceberg/Delta/Hudi) combination — pure file-on-S3 deployments without a table-format overlay are increasingly rare in greenfield. Per Modern Data Lakehouse 2026 — Part 1.
- Apache Iceberg is the dominant open-table-format choice for new lakehouses. Iceberg adoption is accelerating as the open table format standard enabling multi-engine data access (Spark, Trino, Snowflake, Databricks, DuckDB all read the same tables). Per Modern Data Lakehouse 2026 — Part 2.
- Real-time emerged as the third lakehouse requirement. Streaming + Iceberg integration is the 2026 architectural extension: AI agents, operational dashboards, and event-driven apps need fresher data than batch ETL provides — driving the streaming-into-Iceberg pattern. Per Medium — Real-Time Data Lakehouse 2026.
- Governance is now a first-class lakehouse component. Access controls, audit logs, data lineage, and schema enforcement integrated into the lakehouse layer — not bolted on. Reflects the maturity from "dump data in S3" to "engineered data platform." Per Promethium — Lakehouse Architecture Complete Guide 2026.
- Data lake vs lakehouse vs data warehouse framing settled. The 2026 comparison guides converge on: data lake = raw + flexible + cheap; data warehouse = structured + governed + expensive; lakehouse = structured + governed + cheap (the open-format + object-storage combination). Per OvalEdge — Data Lake vs Lakehouse 2026.
Connections 15
Outbound 2
is_a1scoped_to1Inbound 13
scoped_to13Resources 3
AWS's official whitepaper on building data lakes defines architecture patterns, ingestion strategies, and governance frameworks for production data lakes on S3.
AWS's conceptual overview explains what a data lake is, how it differs from a data warehouse, and the key design principles.
Microsoft Azure's data lake overview provides an alternative cloud vendor's perspective, reinforcing the vendor-agnostic nature of the concept.