Comparative Survey of Query Processing on Graph Databases

Graph Databases are rapidly increasing in popularity, size and application. Currently, graph query processing involves some form of isomorphism test, which results in very high response times. Indexing is the most popular way to optimize query processing times. In this paper, we compare some of the existing work on subgraph query processing including cIndex, gIndex and FG-Index. There is a precision performance trade-off involved in subgraph queries on graph databases. There is a need to distinguish efficient querying methods tailored to certain applications. By analyzing the state of the art and comparing the methods in use, we can identify the key aspects in each and build a new indexing mechanism that can be adjusted according to the application and boost performance. This new index will map graphs in the dataset onto a plane and borrow some properties of similarity search techniques to greatly reduce the size of the candidate set of graphs on which the isomorphism test is performed.