Distributed graph database writes

1. A method, comprising:

executing a set of processes for processing queries of a graph database storing a graph, wherein the graph comprises a set of nodes, a set of edges between pairs of nodes in the set of nodes, and a set of predicates;

when a write query of the graph database is received, using one or more of the processes to process the write query by:

using a schema of the graph database to validate the write query;

transmitting the validated write query to a set of shards comprising the graph database; and

for each shard in the set of shards, processing the write query at the shard when a shard identifier representing the shard is matched to one or more attributes in the write query;

when an additional write query fails validation using the schema, retrieving a latest version of the schema from one or more of the shards; and

using the latest version of the schema to revalidate the additional write query.

2. The method of claim 1 , wherein using the schema to validate the write query comprises:

executing the write query in a branch of the graph database; and

validating the write query when the executed write query succeeds.

3. The method of claim 1 , wherein processing the write query at the shard when the shard identifier representing the shard is matched to the one or more attributes in the write query comprises:

generating one or more shard identifiers from the one or more attributes; and

when the one or more shard identifiers includes the shard identifier representing the shard, executing a portion of the write query associated with the shard identifier at the shard.

4. The method of claim 3 , wherein the one or more shard identifiers are generated from hashes of the one or more attributes.

5. The method of claim 3 , wherein the one or more shard identifiers are generated from a set of identity-giving nodes in a compound from the write query.

6. The method of claim 5 , wherein the one or more shard identifiers are further generated from a non-identity-giving attribute associated with the compound.

7. The method of claim 3 , wherein the one or more shard identifiers are generated from a subject and an object in an edge from the write query.

8. The method of claim 7 , wherein:

the one or more shard identifiers are further generated from an additional object in an existing edge comprising the subject and a predicate from the edge, wherein the predicate has a cardinality of 1; and

wherein processing the portion of the query associated with the shard identifier at the shard comprises deleting the existing edge at the shard.

9. The method of claim 1 , wherein the schema is used to validate at least one of:

a predicate; and

a compound.

10. The method of claim 1 , wherein each shard in the set of shards comprises:

a log-based representation of the graph database; and

an index of the graph database.

11. The method of claim 1 , further comprising:

when the write query includes a new edge with a subject, an object, and a predicate, applying an implicit delete functionality at the shard to delete an existing edge comprising at least the predicate and the subject.

12. An apparatus, comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the apparatus to:

execute a set of processes for processing queries of a graph database storing a graph, wherein the graph comprises a set of nodes, a set of edges between pairs of nodes in the set of nodes, and a set of predicates;

when a write query of the graph database is received, use one or more of the processes to process the write query by:

using a schema of the graph database to validate the write query;

transmitting the validated write query to a set of shards comprising the graph database; and

for each shard in the set of shards, processing the write query at the shard when a shard identifier representing the shard matches one or more attributes in the write query;

retrieve a latest version of the schema from one or more of the shards when an additional write query fails validation using the schema; and

use the latest version of the schema to revalidate the additional write query.

13. The system of claim 12 , wherein the memory further stores instructions that, when executed by the one or more processors, cause the apparatus to:

when the write query includes a new edge with a subject, an object, and a predicate, apply an implicit delete functionality at the shard to delete an existing edge comprising at least the predicate and the subject.

14. The apparatus of claim 12 , wherein using the schema to validate the write query comprises:

executing the write query in a branch of the graph database; and

validating the write query when the executed write query succeeds.

15. The apparatus of claim 12 , wherein processing the write query at the shard when the shard identifier representing the shard is matched to the one or more attributes in the write query comprises:

generating one or more shard identifiers from the one or more attributes; and

when the one or more shard identifiers includes the shard identifier representing the shard, executing a portion of the write query associated with the shard identifier at the shard.

16. The apparatus of claim 15 , wherein each portion in the one or more portions comprises at least one of:

an edge; and

a compound.

17. The apparatus of claim 15 , wherein the one or more shard identifiers are generated from hashes of the one or more attributes.

18. The apparatus of claim 15 , wherein the one or more shard identifiers are generated from at least one of:

a set of identity-giving nodes in a compound from the write query;

a non-identity-giving attribute associated with the compound;

a subject and an object in an edge from the write query; and

an additional object in an existing edge comprising the subject and a predicate from the edge, wherein the predicate has a cardinality of 1.

19. A system, comprising:

a graph database storing a graph, wherein the graph comprises a set of nodes, a set of edges between pairs of nodes in the set of nodes, and a set of predicates;

a broker comprising a non-transitory computer-readable medium comprising instructions that, when executed, cause the system to process a query of the graph database by:

using a schema of the graph database to validate the write query;

transmitting the validated write query to a set of shards comprising the graph database;

retrieving a latest version of the schema from one or more of the shards when an additional write query fails validation using the schema; and

using the latest version of the schema to revalidate the additional write query; and

the set of shards, wherein each shard in the set of shards comprises a non-transitory computer-readable medium comprising instructions that, when executed, cause the system to process the write query at the shard when a shard identifier representing the shard is matched to one or more attributes in the write query.

20. The system of claim 19 , wherein processing the write query at the shard when the shard identifier representing the shard is matched to the one or more attributes in the write query comprises:

generating one or more shard identifiers from the one or more attributes; and

when the one or more shard identifiers includes the shard identifier for the shard, executing a portion of the write query associated with the shard identifier at the shard.