Project Roadmap¶
For the latest status on implementation and planned features, please refer to the ROADMAP.md in the GitHub repository.
Future Phases¶
Multi-Column Indexes¶
We are currently researching the implementation of Composite Scalar Indexes (pre-computed roaring bitmaps for common multi-column filters) and Multi-Vector Search (simultaneous ranking across multiple embedding columns).
Distributed Query Engine¶
Future versions of HyperStreamDB will include a distributed query planner capable of coordinating multiple serverless execution nodes for ultra-large datasets.
Native Spark and Trino Connectors¶
Native integration for distributed SQL engines to leverage HyperStreamDB sidecar indexing.