Communications of the ACM
A Shiny, New Graph Query System
Credit: Indiana University Bloomington
Pacific Northwest National Laboratory (PNNL) researchers have developed the Graph Engine for Multithreaded Systems (GEMS) program, a multilayer software system for semantic graph databases. The researchers say GEMS offers a data-scalable, graph-oriented query system that can answer science-based queries over large-scale, real-world, curated scientific metadata.
The researchers examined how GEMS answered queries on scientific metadata and compared its scaling performance against generated benchmark data sets. They found GEMS can answer queries over scientific metadata in seconds and scaled well for larger quantities of data. The researchers also demonstrated how GEMS outperformed a custom-hardware solution.
GEMS works with the Resource Description Framework data model to query large data sets. The researchers say the GEMS compiler translates SPARQL queries into C++ code, which runs the query to quickly and naturally support parallel graph walking. GEMS "is designed to scale out on clusters, allowing us to effectively increase global memory by adding nodes to the cluster," says PNNL researcher Jesse Weaver.
In the future, the researchers will pursue design changes that will enable queries on data sets of more than 100 billion triples.
From Pacific Northwest National Laboratory
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found