Modern business applications request support for easy to use interfaces to modify, store, and query data graphs with in the database management system. Graph abstraction layer in the SAP HANA database gives an imperative approach to co-operate with graph data stored in the database by showing graph concepts, like vertices and edges, straight to the application developer. Its programming border is called as a Graph API, can be used by the application layer through remote procedure calls.

The graph abstraction layer allows creating a new graph by identifying minimal database schema information, like as an edge store name, a vertex in- store name, and a vertex identifier in description.

An existing graph can be unlocked by specifying the edge store and also the vertex store of the graph. All that are missing information, such as like the edge description, are routinely collected from the store metadata.

Vertices and edges are signified by light weight objects that perform as an abstract representative of the object that has been stored in the database. The objects in the graph abstraction layer which only point to the actual data of the object and make them to hold the internal state during processing of the object. If the graph abstraction layer performs a function call that requests data from the object, which gets loaded on demand.

Vertices can be assigned with multiple properties. These properties can be used for the usage of filter vertices. Example: Graph traversal operations.

In the same way, properties on limits can be leveraged to choose some of the possible paths to track in graph traversal.

Well-organized support for extensible and configurable graph traversals on big graphs is a core asset for business applications to be able to implement modified graph algorithms on top of the Graph API. The SAP HANA database delivers native and extensive support for traversals on big graphs on the basis of a graph traversal operator implemented directly in the database kernel.