Graph database and relational database mapping
Examples for mapping a relational database to a graph database include a mapping engine to execute an arbitrary query on a relational database, identify a result column tag based on a tag of an underlying base table, process the result column into a typed, directed property graph based on the result column tag, and output the typed, directed property graph to a graph database. Examples for mapping a graph database to a relational database include processing a graph transaction by updating a mapping layer with a surrogate describing a change to a database object, determining, for an object in the mapping layer, if a database constraint defined on the object is satisfied, collecting database changes defined by the surrogate into a database change request, submitting the change request to a relational database as a transaction, and deleting the surrogate for the object in the mapping layer.
1. A method for mapping a graph database to a relational database, comprising:
processing, by a processor of a computing device, a graph transaction of a change in the graph database;
creating, by the processor, in a mapping layer, a surrogate describing a change to an object in the relational database corresponding to the change in the graph database;
validating, by the processor, the change to the object described in the surrogate, by determining, with the change to the object, whether a relational-database constraint defined on the object is satisfied; and
in response to a determination that the relational-database constraint defined on the object is satisfied, writing, by the processor, the change to the object described in the surrogate into a database change request, submitting the database change request to the relational database as a relational-database transaction, and deleting the surrogate in the mapping layer.
2. The method of claim 1 , further comprising:
in response to a determination that the relational-database constraint defined on the object is not satisfied, caching the change to the object in the mapping layer until the relational-database constraint defined on the object is satisfied.
3. The method of claim 1 , wherein the surrogate further describes a change to a class of the object.
4. The method of claim 1 , wherein the surrogate further describes a change to a relationship of the object.
5. An article comprising at least one non-transitory machine-readable storage medium comprising instructions executable by a processing resource of a graph database system to:
process a change in a graph database;
create, in a mapping layer, a surrogate describing a change to an object in a relational database corresponding to the change in the graph database;
validate the change to the object described in the surrogate, by determining, with the change to the object, whether a relational-database constraint defined on the object is satisfied; and
in response to a determination that the relational-database constraint defined on the object is not satisfied, cache the change to the object in the mapping layer until the relational-database constraint is satisfied.
6. The article of claim 5 , wherein the surrogate describes a change to a class of the object.
7. The article of claim 5 , wherein the surrogate describes a change to a relationship of the object.
8. The article of claim 5 , wherein the instructions are executable to cause the processing resource to:
create, in addition to the surrogate describing the change in the object, a plurality of surrogates for the object linked in order of time, wherein the plurality of surrogates describe additional changes in the object.
9. The article of claim 8 , wherein the plurality of surrogates describes the additional changes to classes and relationships of the object in the relational database.
10. The article of claim 8 , wherein the instructions are executable to cause the processing resource to:
validate the additional changes, by determining whether the additional changes violate the relational-database constraint defined on the object.
11. The article of claim 10 , wherein, to validate the additional changes, the instructions are executable to cause the processing resource to start the validation with an oldest surrogate in the order of time.