Show HN: Tonbo – an embedded database for serverless and edge runtimes
4 months ago
- #edge-computing
- #serverless
- #embedded-database
- Tonbo is an embedded database for serverless and edge runtimes, storing data as Parquet on S3 with stateless compute.
- Key features include async-first design, no server management, Arrow-native data types, and compatibility with various runtimes like Tokio, WASM, and edge environments.
- Tonbo supports open formats like Parquet, enabling data to be read by any tool without vendor lock-in.
- Use cases include building serverless/edge apps needing durable state, storing append-heavy data in S3, and embedding a lightweight MVCC + Parquet storage engine.
- Provides Rust API for schema definition, data insertion, and querying with predicates, supporting local development and S3 storage.
- Implements a merge-tree optimized for object storage with MVCC for snapshot isolation and manifest-based coordination.
- Includes features like projection pushdown, composite keys, streaming, nested types, and time-travel queries.
- Currently in alpha, with ongoing development for storage, schema & query, backends, runtime, and integrations.
- Documentation includes user guides, API references, and RFCs for contributors.
- Licensed under Apache License 2.0.