Discover and explore top open-source AI tools and projects—updated daily.
ankur-anandA log-native, streaming multimodal database for Edge AI
Top 78.7% on SourcePulse
UnisonDB addresses the challenge of unifying data persistence and real-time streaming for Edge AI and Edge Computing. It targets developers building reactive, local-first distributed systems, offering a log-native database that enables near-instant replication and low-latency responsiveness across numerous edge nodes without complex external pipelines.
How It Works
UnisonDB is a log-native, multi-model database that fuses storage and streaming. It utilizes a B+Tree engine (BoltDB/LMDB) for durable, queryable data and Write-Ahead Logging (WAL) for high-throughput ingest and efficient streaming replication. Every write becomes a durable stream, enabling sub-second fan-out to hundreds of nodes and simplifying distributed application architectures.
Quick Start & Requirements
go build -o unisondb ./cmd/unisondb, and run ./unisondb --config config.toml replicator. An example HTTP API PUT request is provided.mkcert for local TLS certificate generation.https://github.com/ankur-anand/unisondb. Documentation includes Getting Started, Configuration, Architecture, API Reference, and Deployment guides.Highlighted Details
Maintenance & Community
No specific details regarding maintainers, community channels (e.g., Discord, Slack), or roadmap content were present in the provided README excerpt.
Licensing & Compatibility
Limitations & Caveats
The README focuses on core functionality and benefits. Specific limitations such as alpha/beta status, known bugs, unsupported platforms, or detailed performance degradation factors under various conditions are not explicitly detailed. The development section implies TLS setup is required for secure communication.
4 days ago
Inactive
Lightning-AI
Eventual-Inc
deepseek-ai