Partition advisor for cloud-based database systems

A non-transitory computer-readable storage medium coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations for implementing partitions in cloud-based database systems, the operations comprising:

• for a historical query in a set of historical queries, determining a set of columns, the set of historical queries comprising queries that have been executed in a cloud-based database system;
• for at least a sub-set of columns of the set of columns, generating a first set of feature vectors, each feature vector in the first set of feature vectors being associated with a label indicating a partition type;
• providing a partition recommendation model by training using a plurality of sets of feature vectors that comprises the first set of feature vectors;
• receiving, from a slow query repository, a slow query that is executed in the cloud-based database system, the slow query comprising a query that is designated as slow based on previous execution of the query within the cloud-based database system;
• generating a second set of feature vectors representative of respective columns implicated by the slow query;
• and
• for a first feature vector in the second set of feature vectors:
• processing the first feature vector by the partition recommendation model to provide a first set of partition recommendations, and
• implementing a first partition recommendation of the first set of partition recommendations as a first partition in the cloud-based database system.

A system, comprising:

• a computing device;
• and
• a computer-readable storage device coupled to the computing device and having instructions stored thereon which, when executed by the computing device, cause the computing device to perform operations for implementing partitions in cloud-based database systems, the operations comprising:
• for a historical query in a set of historical queries, determining a set of columns, the set of historical queries comprising queries that have been executed in a cloud-based database system;
• for at least a sub-set of columns of the set of columns, generating a first set of feature vectors, each feature vector in the first set of feature vectors being associated with a label indicating a partition type;
• providing a partition recommendation model by training using a plurality of sets of feature vectors that comprises the first set of feature vectors;
• receiving, from a slow query repository, a slow query that is executed in the cloud-based database system, the slow query comprising a query that is designated as slow based on previous execution of the query within the cloud-based database system;
• generating a second set of feature vectors representative of respective columns implicated by the slow query;
• and
• for a first feature vector in the second set of feature vectors:
• processing the first feature vector by the partition recommendation model to provide a first set of partition recommendations, and
• implementing a first partition recommendation of the first set of partition recommendations as a first partition in the cloud-based database system.