Efficient denormalization of data instances

An apparatus comprising:

• a processor;
• and
• a computer storage medium having computer executable instructions stored thereon which, when executed by the processor, cause the processor to:
• register an annotated schema that comprises an annotation for denormalizing data instances of the annotated schema,
• process a data instance of the annotated schema according to the annotation to generate a plurality of sub per object indexes (“sub POIs”), the plurality of sub POIs comprising at least a target sub POI to store data of the data instance and a source sub POI to store data of a source data instance that the data instance depends on, wherein the data instance is linked to the source data instance through a normalized data relationship,
• process the source data instance to obtain data values for the source sub POI,
• populate the data values from the source data instance into the source sub POI, and cause the plurality of sub POIs to be stored in a POI storage, the POI storage exposing the plurality of sub POIs to an indexing engine to enable generation of an index based on the plurality of sub POIs so that a search on the data instance can be performed utilizing the index, wherein the plurality of sub POIs represent a denormalized form of the data values stored in the normalized data relationship among the data instance and the source data instance, and wherein the search is performed on the denormalized form of the data values utilizing the index.

A system for denormalizing data instances, the system comprising:

• one or more computing devices configured to:
• receive an annotated schema comprising an annotation for denormalizing data instances of the annotated schema, the annotation comprising an indication that data instances of the annotated schema are dependent on data instances of another schema;
• receive a first data instance of the annotated schema;
• process the first data instance according to the annotation to generate two or more sub per object indexes (“sub POIs”) for the first data instance;
• receive a second data instance of the another schema that the first data instance depends on, wherein the second data instance is linked to the first data instance through a normalized data relationship;
• and
• process the second data instance to update at least one of the two or more sub POIs that is related to the second data instance, wherein the plurality of sub POIs represent a denormalized form of data values stored in the normalized data relationship among the second data instance and the first data instance;
• and
• a POI storage configured to store a plurality of sub POIs of a plurality of data instances comprising the two or more sub POIs;
• and
• an indexing engine configured to:
• access the plurality of sub POIs in the POI storage, and
• generate an index for the plurality of data instances to facilitate a search on the denormalized form of data values utilizing the index.