Accession Number : AD1038983


Title :   X-Graphs: Language and Algorithms for Heterogeneous Graph Streams


Descriptive Note : Technical Report,01 Sep 2012,01 Mar 2017


Corporate Author : The Leland Stanford Junior University Stanford United States


Personal Author(s) : Olukotun,Oyekunle


Full Text : http://www.dtic.mil/dtic/tr/fulltext/u2/1038983.pdf


Report Date : 01 Sep 2017


Pagination or Media Count : 14


Abstract : The overall goal of the X-Graphs project was to develop computational techniques and software tools for graph analytics. This report describes the two main components of this project. The first component focuses on support for interactive graph analytics applications on medium to large size graphs. The second component focuses on support for very high performance graph analytics on large to huge sized graphs. The first component of X-Graphs is SNAP. SNAP provides interactive analytics on graphs with tens of billions of edges that still fit into a single multi-CPU sever memory. The second component of X-Graphs is the Delite framework for building compilers for high-performance Domain Specific Languages (DSLs) that can be used to target heterogeneous architectures (multicore, GPU, cluster, FPGA). These tools are widely used by academia and industry.


Descriptors :   HIGH PERFORMANCE COMPUTING , computing system architectures , information processing , algorithms , artificial neural networks , field programmable gate arrays , automatic programming , monte carlo method , compilers , GRAPHS , information systems , machine learning


Subject Categories : Computer Programming and Software
      Computer Systems


Distribution Statement : APPROVED FOR PUBLIC RELEASE