WCSE 2020 Summer
ISBN: 978-981-14-4787-7 DOI: 10.18178/wcse.2020.06.045

TriAG:Answering SPARQL Queries Accelerated by GPU

Jinhui Pang, Shujun Wang, Jie Jiao, Weikang Zhou, Fan Feng, Ding Zhang

Abstract— In this paper, we present a new RDF engine accelerated by GPU, named TriAG, to query the RDF graph efficiently. Firstly, to improve the processing efficiency of SPARQL on RDF, new storage models of RDF systems is proposed. Then we use query decomposition to further reduce the query response time; at the same time, a cost model based on machine learning is used to determine the granularity of query decomposition. After this, we develop a MapReduce-based algorithm to join solutions of SPARQL subqueries in a parallel way. Finally, we implement TriAG and evaluate it by comparing it with two popular SPARQL query engines, namely, gStore and RDF3X on the LUBM benchmark. The experiments demonstrate that TriAG is highly efficient and effective

Index Terms—SPARQL,RDF,GPU

Jinhui Pang, Shujun Wang, Jie Jiao, Weikang Zhou, Fan Feng, Ding Zhang
Beijing Institute of Technology, CHINA
Tianjin University, CHINA

[Download]


Cite:Jinhui Pang, Shujun Wang, Jie Jiao, Weikang Zhou, Fan Feng, Ding Zhang , "TriAG:Answering SPARQL Queries Accelerated by GPU " Proceedings of 2020 the 10th International Workshop on Computer Science and Engineering (WCSE 2020), pp. 300-306, Shanghai, China, 19-21 June, 2020.