MapReduce based Distributed Graph Grep using Edge Occurrence Index

1Fathimabi Shaik, Ebenezer Jangam


Graph query processing is a very important application for the graph data. The graph data set size increases day by day due to digitization of all types of data, in order to process the large amount of graph data using number of machines not by single machine. Graph query processing using distributed frameworks like Hadoop is a challenging task. Many users are giving graph queries to process in distributed environment, in an interactive way it has to process all the queries. It becomes hard to process graph queries from a big graph dataset. This paper mainly emphasis on processing of multiple graph queries over a large set of graphs, using MapReduce framework. We introduced edge occurrence index to process multiple queries using filter and verify technique in MapReduce. We are using structure based graph partitioning to distribute all the graphs to the machines in the cluster based on structure of the graphs. The proposed algorithm is called as MapReduce based Distributed Graph Grep using Edge Occurrence Index MRDGG. Extensive experimental result analysis on various real-world graph datasets proved that the proposed work improves the performance and reduces the time for multiple graph query processing for massive collections of graphs.


graph query, graph dataset, bigdata, parallel processing, MapReduce, Distributed graph query processing, Join technique.

Paper Details
IssueIssue 3