BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:2.0
BEGIN:VEVENT
DTSTART:20151117T231500Z
DTEND:20151118T010000Z
LOCATION:Level 4 - Lobby
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: RDF databases naturally map to labeled, directed graphs.=0ASPARQL is a query language for RDF databases that expresses queries as graph pattern matching operations. GEMS=0Ais a RDF database that, differently from other solutions, employs graph methods at all levels of its stack. Graph methods are inherently task parallel, but they exhibit an irregular=0Abehavior. In this poster we discuss an approach to accelerate=0AGEMS with reconfigurable devices. The proposed approach=0Aautomatically generates parallel hardware implementations=0Aof SPARQL queries using a customized High Level Synthesis (HLS) flow. The flow has been enhanced with solutions=0Ato address limitations of graph methods with conventional=0AHLS methods, enhancing TLP extraction and management=0Aof concurrent memory operations. We have validated our=0Aapproach by synthesizing and simulating seven queries from=0ALUBM. We show that that the proposed approach provides=0Aan average speed up or 2.1 with respect to the serial version=0Aof the hardware accelerators.
SUMMARY:High Level Synthesis of SPARQL Queries
PRIORITY:3
END:VEVENT
END:VCALENDAR