BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20131119T173000Z
DTEND:20131119T180000Z
LOCATION:205/207
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: As the size and variety of information networks continue to grow in many scientific domains, we witness a growing demand for efficient processing of large heterogeneous graphs using a cluster of compute nodes in the Cloud. One of the main open issues is how to effectively partition a large graph to process complex graph operations efficiently. In this paper, we present a distributed data partitioning framework for efficient processing of large-scale graphs in the Cloud. First, we introduce extended vertex blocks as graph partitioning building blocks. Second, we propose vertex block grouping algorithms which group those vertex blocks that have high correlation in graph structure to the same partition. Third, we propose a partition-guided query partitioning model which transforms graph queries into vertex block-based graph query patterns for parallel processing of graph queries. We conduct extensive experiments on several real-world graphs to show the effectiveness and scalability of our framework.
SUMMARY:Efficient Data Partitioning Model for Heterogeneous Graphs in the Cloud
PRIORITY:3
END:VEVENT
END:VCALENDAR