Triple Pattern Fragments: A low-cost knowledge graph interface for the Web

Abstract

Billions of Linked Data triples exist in thousands of RDF knowledge graphs on the Web, but few of those graphs can be queried live from Web applications. Only a limited number of knowledge graphs are available in a queryable interface, and existing interfaces can be expensive to host at high availability. To mitigate this shortage of live queryable Linked Data, we designed a low-cost Triple Pattern Fragments interface for servers, and a client-side algorithm that evaluates SPARQL queries against this interface. This article describes the Linked Data Fragments framework to analyze Web interfaces to Linked Data and uses this framework as a basis to define Triple Pattern Fragments. We describe client-side querying for single knowledge graphs and federations thereof. Our evaluation verifies that this technique reduces server load and increases caching effectiveness, which leads to lower costs to maintain high server availability. These benefits come at the expense of increased bandwidth and slower, but more stable query execution times. These results substantiate the claim that lightweight interfaces can lower the cost for knowledge publishers compared to more expressive endpoints, while enabling applications to query the publishers' data with the necessary reliability.

Self-Enforcing Access Control for Encrypted RDF

“…At this point we also want to stress that due to the nature of our approach, each result triple can be returned as soon as its decryption has finished. This is in line with the incremental nature of the Triple Pattern Fragment [31] approach, which paginates the query results (typically including 100 results per page), allowing users to ask for further pages if required. For example, decrypting Jamendo entirely took about 2256 s for VP and 2808 s for 3-Index, leading to respective triple decryption rates of 465 triples/s and 374 triples/s in a cold scenario, which already fulfils the performance requirements to feed several Triple Pattern Fragments per second.…”

“…In order to maintain simplicity and general applicability of the proposed store, both alternatives consider key-value backends, which are increasingly used to manage RDF data [8], especially in distributed scenarios. It is also worth mentioning that we focus on basic triple pattern queries as (i) they are the cornerstone that can be used to build more complex SPARQL queries, and (ii) they constitute all the functionality to support the Triple Pattern Fragments [31] interface. 3-Index Approach.…”

Live Storage and Querying of Versioned Datasets on the Web

“…In future work, we will extend the client [8] with support for this interface, so that it is able to consume data using the version query atoms for more complex queries, using an appropriate query language.…”

“…Querying on the Web is possible using the Triple Pattern Fragments ( ) framework [8], which was introduced as an alternative to endpoints [2] for publishing Linked Data at a low cost, while still enabling queryable access. The approach limits server interfaces to triple pattern queries and moves the effort of full query evaluation to the client.…”

HDTQ: Managing RDF Datasets in Compressed Space

“…HDT has been widely adopted by the community, (i) used as the main backend of Triple Pattern Fragments (TPF) [18] interface, which alleviates the traditional burden of LOD servers by moving part of the query processing onto clients, (ii) used as a storage backend for large-scale graph data [16], or (iii) as the store behind LOD Laundromat [3], serving a crawl of a very big subset of the LOD Cloud, to name but a few.…”