Enterprise-ready Triple Store

For too long triple stores have been separated from the data source itself, often causing meaning and context to be lost. Until now.

MarkLogic lets organizations operationalize their own Semantic Web Technologies. As a native RDF datastore, MarkLogic can handle billions and even a trillion triples that can be queried using W3C standard SPARQL. But what makes our triple store novel is that we address the hard, organizational problems that other triple stores can’t – in a robust, enterprise-ready way.

One of the reasons semantic web technologies have been just a glimmer in a researcher’s eyes is because the triple stores available were designed to do one thing – and one thing only. They don’t scale horizontally, they lose data when they lose power, and if you’re just a triple store you have real problems with things like provenance and reification.

Here’s how MarkLogic is different:



When we say “enterprise-ready” we mean:

  • High-availability and disaster recovery (HA/DR) protects against data loss and system downtime by replicating data for partial or complete site failures.
  • Government-grade security – users see only the triples they are allowed to see based on their roles.
  • Robust, trusted – many of the world’s biggest corporations run their business on MarkLogic.

Horizontally Scalable

Scale out by adding commodity hardware to your shared-nothing cluster. That means you can add terabytes of triples into our database – and we won’t break a sweat.

  • Keep on Clustering – Some triple stores let you build a cluster, but for parallel query only – that is, you can build a 3-node cluster as long as each node has the same data on it. With MarkLogic, if you have more triples than you can manage, you just add another node to your cluster.
  • No Memory Limitations – Some triple stores insist you store the whole triple store index in memory. The MarkLogic triple store is heavily cached, but doesn’t need to fit in memory, so you’re not limited by physical memory.

Reification & Provenance Made Easy

Add metadata to your triples by annotating them — and query across triples and annotations together. This allows you to query triples using SPARQL, and restrict the results to a source, a date range, an author – and report back on the “paper trail” of where those triples came from.