2014 | OriginalPaper | Buchkapitel
GoFFish: A Sub-graph Centric Framework for Large-Scale Graph Analytics
verfasst von : Yogesh Simmhan, Alok Kumbhare, Charith Wickramaarachchi, Soonil Nagarkar, Santosh Ravi, Cauligi Raghavendra, Viktor Prasanna
Erschienen in: Euro-Par 2014 Parallel Processing
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Vertex centric models for large scale graph processing are gaining traction due to their simple distributed programming abstraction. However, pure vertex centric algorithms under-perform due to large communication overheads and slow iterative convergence. We introduce
GoFFish
a scalable sub-graph centric framework co-designed with a distributed persistent graph storage for large scale graph analytics on commodity clusters, offering the added natural flexibility of shared memory sub-graph computation. We map Connected Components, SSSP and PageRank algorithms to this model and empirically analyze them for several real world graphs, demonstrating
orders of magnitude improvements
, in some cases, compared to Apache Giraph’s vertex centric framework.