TSO Triplestore FAQs

What is the TSO Triplestore?

TSO’s Triplestore is a platform for the storage and query of RDF triples to enable central government departments and local government to host and publish open linked data. TSO has brought together its unrivalled experience in managing public sector websites with the highly scalable, clustered 5Store database designed and developed by Garlik, leaders in semantic technology to provide a very large scale, robust platform.

How do you know TSO's Triplestore is capable of storing 1 trillion triples?

5Store was developed with scalability in mind. Its strengths are performance, scalability and stability in the domain of RDF storage and SPARQL queries. 5Store was designed from scratch for clusters of up to 1,000 machines and can scale to over 1 trillion (10¹²) triples given a large enough cluster. Due to the new indexing algorithms deployed, there is no effective limit on the number of triples that can be stored.

What software and hardware is used?

5Store is implemented in ANSI C and C++. It makes use of custom discovery and clustering technologies and proprietary, novel indexing algorithms. It is primarily intended to run on 64-bit Linux platforms with 32GB of RAM or more.

How do you query the TSO Triplestore?

TSO’s Triplestore implements the evolving SPARQL specification and continues to be updated as the specification advances. It passes the DAWG tests which test SPARQL compliance, with a very efficient query engine that is being continuously optimized for performance. It has very efficient streaming of query results, providing the ability to ask the server to return more results rather than start again which is very important for large results sets. It implements ‘delete triple’ which is needed for efficient implementation of SPARQL / Update, the next standard.

How fast can triples be imported into TSO's Triplestore?

TSO’s Triplestore is inherently very fast even on a single machine and small clusters, with the ability to handle over 200 million triples with import speeds of over 150 thousand triples per second on a single machine. However, unlike many RDF stores it can efficiently utilise large clusters, enabling it to maintain import and query speed on enormous databases, given enough hardware.

5Store allows for thousands of simultaneous connections. It has very efficient storage operating as low as 100 bytes per triple (depending on the type of data) and it supports dynamic configurable segmentation which is one of the design features that lets 5Store scale to such a large size.

What are the costs involved?

Hosting for the first three months is free. Beyond that period pricing is based on an annual hosting fee plus a variable fee based on the number of triples stored and bandwidth usage. Contact us for details of costs for your data.

Access to the TSO Triplestore is free for the developer community subject to the permission of data owners.

How do I get my data into TSO's Triplestore

A number of APIs are available including HTTP (PUT, DELETE, POST) .

Contact us for further details.