TDB is an open source RDF database developed and maintained by the Apache Jena project
TDB is an open source RDF database developed and maintained by the Apache Jena project. It is a Java based embedded database that may be exposed over HTTP using the Fuseki Server also from the Apache Jena project.
Features
It's features includes the following:
- Persisting RDF Triples or Quads
- Full SPARQL execution
- Write ahead logging to provide serializable transactions and fault tolerance
Scalability and Performance
TDB typically scales up to datasets of a few hundred million triples or quads.
Compared to commercial alternatives TDB is often less scalable compared primarily because while it is persisted to disk it is fundamentally designed as an in-memory database. It relays heavily on RAM caches and memory mapped files so scalability tends to be limited by machine RAM.
Also TDB does not have any option of a clustered mode so cannot be scaled horizontally without additional technologies (e.g. manually created replicas and load balancers)
TDB is typically included in the Berlin SPARQL Benchmark Results for those interested in comparative performance data.
