Automatic data optimization
Arctic automates all the tedious bits of data management for the lakehouse, including compaction, repartitioning, and indexing, so data teams no longer need to worry about how data is physically organized in files.
Automate your lakehouse
and simplify data workflows
-
-
Isolation and consistency
Use git-like branching to transform data across tables and schemas in isolation without impacting production workloads, and merge your data changes once they’re tested and ready.
-
Reproduce models and dashboards
Turn back the clock on your entire lakehouse environment (not just a single table) and reproduce ML models or dashboards based on a specific point in time. Track the evolution of lakehouse data over time, with visibility into who made changes to data.
Key Arctic capabilities
Metastore
A metastore service based on Nessie that enables a git-like experience for the lakehouse across any engine, including Sonar, Flink, Presto, and Spark.
Data optimization
A data optimization service that automates data management tasks in your lakehouse, including compaction, repartitioning, and indexing, so any compute engine running on that data can operate more efficiently.
What an intelligent metastore brings to the table(s)
| Traditional Metastore (Hive Metastore) | Intelligent Metastore (Dremio Arctic) | |
|---|---|---|
| Table catalog |  | |
| Data access privileges |  Partial | |
| Cross-table transactions |  | |
| Automatic data optimization | | |
| Git-like experience | | |
