PROJECT TITLE :

High-Level Programming Abstractions for Distributed Graph Processing - 2018

ABSTRACT:

Efficient processing of large-scale graphs in distributed environments has been an increasingly fashionable topic of research lately. Inter-connected data that may be modeled as graphs seem in application domains like machine learning, recommendation, net search, and social network analysis. Writing distributed graph applications is inherently onerous and requires programming models which will cowl a diverse set of problems, including iterative refinement algorithms, graph transformations, graph aggregations, pattern matching, ego-network analysis, and graph traversals. Many high-level programming abstractions are proposed and adopted by distributed graph processing systems and massive knowledge platforms. Even though significant work has been done to experimentally compare distributed graph processing frameworks, no qualitative study and comparison of graph programming abstractions has been conducted nevertheless. During this survey, we review and analyze the foremost prevalent high-level programming models for distributed graph processing, in terms of their semantics and applicability. We have a tendency to review 34 distributed graph processing systems with respect to the graph processing models they implement and we tend to survey applications that seem in recent distributed graph systems papers. Finally, we tend to discuss trends and open analysis queries in the world of distributed graph processing.