Abstract

There is an increasing interest in executing rich and complex analy-sis tasks over large-scale graphs, many of which require processing and reasoning about a large number of multi-hop neighborhoods or subgraphs in the graph. Examples of such tasks include ego net-work analysis, motif counting, finding social circles, personalized recommendations, link prediction, anomaly detection, analyzing in-fluence cascades, and so on. These tasks are not well served by the existing vertex-centric graph processing frameworks, whose com-putation and execution models limit the user program to directly access the state of a single vertex; this results in high communi-cation, scheduling, and memory overheads in executing such tasks using those frameworks. Further, most existing graph processing frameworks typically ignore the challenges in extracting the rele-vant portion of the graph that an analysis task needs, and loading it