Hadoop summit 2017 enterprise graph analytics

Graph approaches to structuring, analyzing data have been a significant area of interest, Graphs are well-suited to expressing complex interconnections and clusters of highly related entities. Large-scale graph analytics research is growing fast in recent years, to leverage Hadoop2 ecosystem for graph is a good approach, enterprise graph computer requires to store large graph and do fast computing against graph. One for the OLTP database systems which allow the user to query the graph in real-time, Hbase as the distributed NOSql database can be the backend storage to persistent large graph, the property graph stored its vertices and edges in key-value pairs in Hbase, it also provide highly reliable, scalable and fault tolerant to the data, Solr as the distributed indexing will make the query more efficient. Titan itself will handle cache, transaction; And another for the OLAP analytics systems, use TinkerPop hadoop gremlin SparkGraphComputer to processed a large graph, every vertex and edge is analyzed, a cluster-computing platform will help for the processing of large distributed in memory graph datasets. Graph DB base on Hbase/Solr and graph computing analysis base on spark is powerful for discovering valuable information about relationships in complex and large data, representing significant business opportunity in enterprise. It will help graph data analytics in a wide range of domains such as social networking, recommendation engines, advertisement optimization, knowledge representation, health care, education, and security.