RDF Fusion
RDF Fusion is an embeddable SPARQL engine based on Apache DataFusion. Our main goal is to provide a performant SPARQL engine that integrates with modern data ecosystems. We also aim to provide support for non-graph data models (e.g., time series) in the future. Note that RDF Fusion is still in an early stage of development.
Open Source Ecosystem
Built on top of projects like Arrow and DataFusion we can enhance the interoperability between projects.
Embeddable
Integrate RDF Fusion directly into your Rust applications. No server infrastructure required.
Disaggregated Storage
We are investigating the use of technologies pioneered for analytical "Big Data" workloads in the context of RDF and SPARQL. RDF Fusion can store data in Parquet files and Delta Lake tables. Both can be queried while residing in an object store.
Customizable
Extend functionality via custom functions and optimization rules. We try to provide a framework for customization, inspired by the extensibility of DataFusion.
Performance
We are building on the shoulders of giants: DataFusion is among the best performing single-node query engines out there. There are certain areas where we can improve performance (e.g., query planning) but you should be able to get solid performance out of the box.