Schemaless to relational representation conversion

1. A system, comprising:

a memory; and

at least one processor coupled to the memory and configured to:

receive data structured in a schemaless data representation, wherein the data comprises an array value and a key associated with the array value;

generate, based on the key, a label path for an object of an ordered multi-map and an index path for the array value;

transform, based on the label path for the object, the index path for the array value, and a value type of the array value, the data to a relational representation; and

store the relational representation of the data in a key-value data store for a query-based retrieval.

2. The system of claim 1 , wherein the schemaless data representation comprises a schemaless JavaScript Object Notation (JSON) hierarchical data representation.

3. The system of claim 2 , wherein the data comprises an event data object, wherein the event data object comprises time series data from at least one of a plurality of sensors or a user device.

4. The system of claim 1 , wherein the at least one processor configured to transform the data to the relational representation is further configured to transform a nested array in the data to a nested ordered multi-map.

5. The system of claim 4 , wherein the at least one processor is further configured to transform the nested ordered multi-map to the relational representation.

6. The system of claim 1 , wherein the at least one processor is further configured to:

encode the relational representation to a columnar block of data; and

store the columnar block in the key-value data store.

7. The system of claim 1 , wherein the at least one processor is further configured to:

distribute the relational representation of the data across a plurality of partitions, and wherein metadata is stored in the plurality of partitions with the relational representation of the data.

8. The system of claim 1 , wherein the at least one processor is further configured to distinguish between an atomic value of the data and an array of a single value of the data.

9. The system of claim 1 , wherein the at least one processor is further configured to distinguish between an array map element of the data and a nested array element of the data.

10. The system of claim 1 , wherein the at least one processor is further configured to recognize a non-map root value of the data.

11. The system of claim 2 , wherein wherein each JSON object of the schemaless JSON hierarchical data representation of the data comprises a different structure.

12. The system of claim 1 , wherein the at least one processor is further configured to reconstruct original JSON formatted data without loss of information.

13. The system of claim 1 , wherein the at least one processor is further configured to reconstruct a subset of original JSON formatted data.

14. The system of claim 1 , wherein the at least one processor is further configured to encode the relational representation of the data into a plurality of columnar blocks of data.

15. The system of claim 1 , wherein the at least one processor is further configured to optimize the query-based retrieval based on push down projection of columns of the relational representation of the data.

16. The system of claim 1 , wherein the at least one processor is further configured to optimize the query-based retrieval based on push down predicate of time dimensions of the relational representation of the data.

17. A computer-implemented method, comprising:

receiving, by a computing device, a data structure in a schemaless data representation, wherein the data structure comprises a value and a key associated with the value;

generating, based on the key, a label path for an object of an ordered multi-map and an index path for the value;

transforming, based on the label path for the object, the index path for the value, and a value type of the value, the data structure to a relational representation of the data structure; and

storing the relational representation of the data structure in a key-value data store for a query-based retrieval.

18. A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, causes the at least one computing device to perform operations comprising:

receiving data structured in a schemaless data representation, wherein the data comprises an array value and a key associated with the array value;

generating, based on the key, a label path for an object of an ordered multi-map and an index path for the array value;

transforming, based on the label path for the object, the index path for the array, and a value type of the array value, the data into a relational representation; and

storing the relational representation of the data in a key-value data store for a query-based retrieval.

19. The computer-implemented method of claim 17 , wherein the data comprises an event data object, wherein the event data object comprises time series data from at least one of a plurality of sensors or a user device.

20. The non-transitory computer-readable medium of claim 18 , wherein the data comprises an event data object, wherein the event data object comprises time series data from at least one of a plurality of sensors or a user device.