Starburst Enterprise Platform offers a powerful SQL query engine which can be used to connect to different data sources. It promotes the fundamental paradigm shift from a single source of truth, that is, getting everything into one place, to a single point of access, that is, having access to everything from one location.
The Starburst Enterprise Platform comprises of the following key components:
- Federated Query Engine – Based on the Trino open-source software, which implements a distributed architecture with a coordinator and multiple workers. The coordinator accepts SQL queries from users, generates an optimized query plan for each query, and distributes the tasks of the query plan across available workers. Finally, it consolidates the results of the query execution from the workers and presents the output to the user.
- Connectors – Over 50 connectors to different types of data sources including relational databases, data lake, data warehouse, storage systems, cloud sources. Connectors are optimized to pushdown queries, add dynamic filtering, and deliver cost-optimal query responses.
- Catalog - Contains configuration and metadata to access data sources. To query a data source, configure a catalog for it and include it in the cluster. Catalog connects to a data source via a connector.
- Virtual Views – Create virtual, non-persistent views combining data across multiple data sources.
- Materialized Views – Create persistent views in your preferred storage systems for low latency production grade use cases. The views can be refreshed at regular intervals with the latest data from the sources.
- Data Products – Enable data producers and consumers to create, publish, discover, and manage curated datasets. This enables greater reuse of data assets across the enterprise.
- Fine grained access control – Column and row level permissions, role-based governance of data assets.
- Connectivity to consumption tools – Such as Tableau, PowerBI, Jupyter, R, Spark, and many more.
- Hive Metastore – Helps to map data in distributed file systems and object storage into tables and provides rich metadata.
- Stargate – Allows catalogs and data sources in one SEP cluster to be linked to those in another cluster, thus enabling a gateway to access data across geographies while complying with data residency requirements.