Project Nessie review
A self-hosted catalog for branching, history, and transactions across Iceberg tables.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
Project Nessie is an open-source transactional catalog and data version-control service for data lakes. It is designed for teams managing Apache Iceberg tables and views that need Git-like commits, branches, tags, hashes, and historical access. Nessie tracks table metadata and associated file lists without copying the underlying data, supporting separate development, staging, and production changes. Its deployment options include Docker, a standalone Java application, and Kubernetes, with web, API, and self-hosted access models.
The strongest fit is the Iceberg ecosystem. Nessie integrates with Apache Iceberg and connects with Apache Spark, Apache Flink, Apache Hive, and Trino through Iceberg; the listed ecosystem also includes Presto. Its transaction model supports atomic changes spanning multiple tables, while point-in-time access can use tags, timestamps, or commit hashes. Garbage collection for unused data files adds a lifecycle control, and users can work through a REST API, CLI, or browser UI. These capabilities make Nessie a focused choice for table-level branching and historical workflows rather than a general-purpose system for every data asset.
Nessie is open source and self-hosted, so teams choose and operate their deployment environment and version-store backend. Supported pluggable backends include DynamoDB, Bigtable, Cassandra, PostgreSQL, MariaDB, MySQL, and RocksDB. That flexibility suits organizations with an existing infrastructure standard, but it also places deployment and operational decisions with the adopter. Choose Project Nessie when Apache Iceberg metadata versioning, multi-table transactions, and branch-based workflows are central requirements. Consider another category fit when the primary need extends beyond table metadata or when a managed hosted service is preferred.
Project Nessie pros and cons
- Where it wins
- Git-like commits, branches, tags, hashes, and historical access
- Atomic transactions across multiple tables
- Integrates with Iceberg, Spark, Flink, Hive, and Trino
- Where it doesn't
- Focused on table metadata rather than broad data-asset versioning
- Self-hosted deployment requires infrastructure management
- Requires a bring-your-own version-store backend
Project Nessie fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update