Dynamic database indexes for entity attribute value stores

1. A database for storing entity-attribute-value (“EAV”) data, the database comprising:

an identifier for each entity of the EAV data, wherein an entity includes a plurality of attributes that each have a value and the identifiers are stored in a first column;

serialized data that includes both the attributes of the EAV data and the values of the EAV data, wherein each of the attributes is assigned to one of the values and corresponds to each of the identifiers, further wherein the serialized data is arranged in series, wherein the serialized data is stored in a second column; and

one or more dynamic attributes from the EAV data that are indexed separately from the serialized data, wherein the one or more dynamic attributes comprise attributes that are removed from the serialized data and the one or more dynamic attributes are stored in different columns from the serialized data, wherein the different columns are dynamically added based on an analysis of statistics of queries for the dynamic attributes such that frequent queries for a particular attribute result in that particular attribute being identified as a dynamic attribute.

2. The database of claim 1 wherein the serialized data comprises one or more attributes and respective values stored together for each entity.

3. The database of claim 1 wherein the one or more dynamic attributes are copied from the serialized data column.

4. The database of claim 1 wherein the one or more dynamic attributes are moved from the serialized data column such that the one or more dynamic attributes are not present with the serialized data.

5. The database of claim 1 wherein the different columns are dynamically added based on seasonal or temporal changes to queries for the dynamic attributes.

6. A computerized method with a database that optimizes entity-attribute-value (“EAV”) data, the method comprising: monitoring queries to a database that includes the EAV data in a serialized data format in a single column that includes both attributes of the EAV data and values of the EAV data, wherein each of the attributes is associated with one of the values and the serialized data format comprises serial data;

identifying attributes from those queries that are queried frequently;

indexing the identified attributes in a dynamic column that is added to the EAV data and is different from the single column and that includes the identified attribute, wherein the attributes that are not indexed with the dynamic column are maintained in the serialized data format and are not in the dynamic column, further wherein the identified attributes indexed in the dynamic column comprise dynamic attributes stored in the dynamic column that are dynamically added based on an analysis of statistics of gueries for the dynamic attributes such that frequent queries for a particular attribute result in that particular attribute being identified as a dynamic attribute; and iteratively monitoring and indexing the identified attributes based on the queries to the database.

7. The method of claim 6 wherein the attributes that are not indexed are stored in a serialized data format in the single column.

8. The method of claim 6 wherein the identification of attributes that are queried frequently is determined when a number of queries of the identified attribute exceeds a threshold number of queries.

9. The method of claim 6 wherein the identification of attributes is based on seasonal or temporal predictions for changes in the queries to the database.

10. A computer system comprising:

a database storing entity-attribute-value (“EAV”) type data, the database comprising:

a first column storing identifiers for each entity from the EAV type data;

a second column storing one or more attributes and respective values for those attributes from the EAV type data, wherein the second column utilizes a serialized format; and

a dynamic column, different from the first column and the second column, that is added for an identified attribute and that includes respective values for the identified attribute; and

a processor coupled with the database for dynamically identifying attributes to be stored in the dynamic column, wherein the identified attributes to be stored in the dynamic column are dynamic attributes that are dynamically added based on an analysis of statistics of queries for the dynamic attributes such that frequent queries for a particular attribute result in that particular attribute being identified as a dynamic attribute, further wherein the dynamic attributes are indexed separately from the EAV data in the serialized format and the dynamic attributes comprise attributes that are removed from the second column.

11. The computer system of claim 10 wherein database comprises additional dynamic columns that correspond to additional identified attributes.

12. The computer system of claim 11 wherein each of the identified attributes that is stored in a dynamic column is indexed for a quicker response to queries for values corresponding to the identified attributes.

13. The computer system of claim 12 wherein each of the identified attributes is stored in a separate dynamic column and multiple dynamic columns are indexed together.

14. The computer system of claim 11 wherein the dynamic identifying of attributes is based on monitoring of queries to the database, such that the identified attribute is dynamically identified when it is queried more than a threshold number of queries.

15. The computer system of claim 10 wherein the identified attribute is dynamically identified based on seasonal or temporal predictions for changes in queries to the database.

16. The computer system of claim 10 wherein the identified attribute is indexed differently when in the dynamic column than when stored in the serialized format in the second column.