Why Apache Iceberg is essential for modern data lakehouses

Surging industry adoption

Originally developed and open-sourced by Netflix in 2018, Apache Iceberg has gained significant traction in recent years. Here are some examples:

• Google BigLake & BigQuery added support for Iceberg, embracing the open format.
• AWS now supports Iceberg across Athena, Redshift and Glue, making it the backbone of its analytics services.
• Snowflake launched Unified Iceberg Tables in 2023 and followed up with full integration in 2025, including AI-ready support and replication features.
• Apple deployed Iceberg across hundreds of data teams and contributed features such as copy-on-write and merge-on-read.
• Databricks, after acquiring Tabular -- the company founded by Iceberg creators -- now supports Iceberg alongside other open formats as part of its Unity Catalog and Lakebase.
• Cloudera was an early adopter, integrating Iceberg across its true hybrid data, AI and analytics platform.
• Qlik recently launched its Open Lakehouse within Qlik Talend Cloud, built atop Apache Iceberg.
• Dremio fully incorporated Apache Iceberg as part of its data lakehouse strategy, positioning as a unified lakehouse platform.
• Oracle adopted Apache Iceberg support in its Autonomous Database and Heatwave platforms.

This wave culminated in the ratification of the Iceberg v3 table specification in spring 2025, solidifying interoperability and vendor neutrality.

Why Apache Iceberg matters now

Several factors have converged to make Iceberg essential to modern data architectures:

True data lakehouse functionality

Iceberg brings features traditionally limited to data warehouses, such as ACID transactions, schema and partition evolution, time travel and support for concurrent reads/writes, into a data lake environment. Unlike other options, Iceberg separates metadata from file storage, enabling strong performance on cloud object storage.

Apache Iceberg isn't just buzz.

Open, vendor-agnostic format

Unlike proprietary formats, Iceberg is truly open. It's supported across engines including Spark, Flink, Trino/Presto, Hive and Athena. This open design removes vendor lock-in, providing organizations with flexibility.

Rapid innovation through community

A dedicated open source community with major contributions from Netflix, Apple, AWS, Snowflake, Dremio and Tabular accelerates feature development. The v3 spec is the product of a coordinated cross-industry effort.

AI and analytics alignment

The boom in generative AI and analytics infrastructure has propelled data openness. With Iceberg, companies can feed petabyte-scale data into AI pipelines, and "time travel" allows rollback and reproducibility, which are essential for auditing.

Hybrid and multi-cloud flexibility

Iceberg separates storage and compute, allowing organizations to adopt hybrid lakehouse architectures by keeping sensitive data on-premises while distributing compute across clouds. Iceberg solves for the metadata and consistency across these environments.

Final word

Apache Iceberg isn't just buzz. It's an important element for a unified, open and flexible data architecture. With its comprehensive feature set, vendor-neutral design, thriving open source ecosystem and AI-forward alignment, it's clear why so many leading technology vendors have adopted it. The result is a next-generation data lakehouse that finally blends flexibility with warehouse-grade reliability.

Stephen Catanzano is a senior analyst at Enterprise Strategy Group, now part of Omdia, where he covers data management and analytics.

Enterprise Strategy Group is part of Omdia. Its analysts have business relationships with technology vendors.