RegattaDB is a distributed Relational Database that runs transactional, analytical, and vector-search workloads against the same data. It is designed for applications that need online transaction processing (OLTP), online analytical processing (OLAP), and semantic retrieval without replicating operational data into separate systems.

RegattaDB uses SQL and implements the PostgreSQL dialect where compatible with its architecture. Its distributed concurrency model supports cross-node transactions and serializable isolation, allowing applications to choose strong transactional consistency for operations that require it.

Key Capabilities

Unified transactional and analytical processing: RegattaDB executes OLTP writes and analytical queries on the same distributed dataset. This supports applications that need current operational state and analytical context without a separate extract, transform, and load (ETL) pipeline.
Vector search: The platform includes Vector Database capabilities for storing, indexing, and querying vector embeddings alongside relational data. This allows AI applications to combine semantic retrieval with transactional and analytical queries.
Distributed SQL: Tables and indexes are divided into data segments that can be placed across cluster nodes and storage devices. Administrators can use default distribution or specify placement for workload-specific tuning.
Transaction isolation: RegattaDB supports Serializable, Read Committed, and Repeatable Read isolation levels. Its documentation states that these guarantees apply across nodes, including transactions that access data stored on multiple nodes.
Cluster architecture: A Regatta cluster consists of database modules for relational-data processing, catalog management, logical sequencing, and distributed deadlock detection. Relational-data modules can be distributed across nodes for scale-out processing.
Deployment options: RegattaDB is available as the managed Regatta Cloud service or for self-hosted deployment. Self-hosted deployments use Linux-based x86-64 infrastructure.

Audience & Use Cases

Audience: Database, platform, data-infrastructure, and application-engineering teams building high-concurrency services or AI-enabled applications.
Use cases: Operational systems that require transactional consistency at scale; real-time analytics on live application data; AI-agent systems that need transactional state and vector search; and applications consolidating separate transactional, analytical, and vector data stores.

Technical Specifications

Query language: SQL, using the PostgreSQL dialect where compatible.
Processing model: Distributed relational data and index segments across cluster nodes and storage devices.
Transaction model: Supports serializable, Read Committed, and Repeatable Read isolation levels.
Self-hosted baseline: Regatta documents support for Red Hat Enterprise Linux 8–based distributions on x86-64 systems.

RegattaDB

What is RegattaDB?

Categories & Use Cases