Granted Patent B2

US 12,271,361 · App. 17/407,993 · Granted Apr 8, 2025

Inventors: Igor Fedorenko (Toronto, CA); Helen Kwong (Danville, CA); Benjamin Busjaeger (San Carlos, CA)

Assignee: Salesforce, Inc.

G06F16/2255

Abstract

Claims (54)

1. A method, comprising:

accessing, by a computing system, a multi-tenant dataset that includes a relational database table in which multiple tenants share one or more columns and a first tenant implements at least one added custom column;

processing, by the computing system, the multi-tenant dataset to migrate the relational database table to non-relational underlying storage and divide the database table, including:

generating first logical table information that specifies a first logical table for a first portion of the dataset corresponding to the first tenant including the at least one added custom column;

generating second logical table information that specifies a second logical table for a second portion of the dataset corresponding to a second tenant; and

storing the first and second logical tables using non-relational key/value data structures in one or more storage elements, including:

storing entries from the first logical table for the first tenant across a first set of multiple key/value store buckets;

storing entries from the second logical table for the second tenant across a second set of multiple key/value store buckets; and

generating internal identifiers for key/value storage for names associated with the first and second logical tables according to a mapping function, wherein:

a given internal identifier uses a smaller number of bits than its corresponding name, thereby reducing storage size relative to utilizing the names for key/value storage; and

the mapping function preserves ordering among keys for a given bucket; and

retrieving, by the computing system, values from the first set of multiple key/value store buckets based on a query from the first tenant that specifies one or more keys.

2. The method of claim 1 , wherein entries in respective key/value store buckets are sortable.

3. The method of claim 1 , wherein the first and second sets of multiple key/value store buckets at least partially overlap.

4. The method of claim 1 , wherein the storing uses partition keys to spread entries from the first and second logical tables across multiple key/value store buckets.

5. The method of claim 4 , wherein the partition keys include a logical table identifier and a bucket number.

6. The method of claim 5 , further comprising generating bucket numbers using a stable hash function.

7. The method of claim 1 , wherein the mapping function provides consistency such that a first name value used during a first time interval and re-used during a second time interval is mapped to two different internal identifiers during the different first and second time intervals.

8. The method of claim 1 , further comprising:

wherein the mapping function encodes output values using a classification field that indicates a size and range of the output value.

9. A system, comprising:

one or more processors; and

one or more storage elements having program instructions stored thereon that are capable of being executed by the one or more processors to cause operations comprising:

accessing a multi-tenant dataset that includes a relational database table in which multiple tenants share one or more columns and a first tenant implements at least one added custom column;

processing the multi-tenant dataset to migrate the relational database table to non-relational underlying storage and divide the database table, including:

generating first logical table information that specifies a first logical table for a first portion of the dataset corresponding to the first tenant including the at least one added custom column;

generating second logical table information that specifies a second logical table for a second portion of the dataset corresponding to a second tenant; and

storing the first and second logical tables using non-relational key/value data structures in one or more storage elements, including:

storing entries from the first logical table for the first tenant across a first set of multiple key/value store buckets;

storing entries from the second logical table for the second tenant across a second set of multiple key/value store buckets; and

generating internal identifiers for key/value storage for names associated with the first and second logical tables according to a mapping function, wherein:

a given internal identifier uses a smaller number of bits than its corresponding name, thereby reducing storage size relative to utilizing the names for key/value storage; and

the mapping function preserves ordering among keys for a given bucket; and

retrieving values from the first set of multiple key/value store buckets based on a query from the first tenant that specifies one or more keys.

10. The system of claim 9 , wherein entries in respective key/value store buckets are sortable.

11. The system of claim 9 , wherein the storing uses partition keys to spread entries from the first and second logical tables across multiple key/value store buckets.

12. A non-transitory computer-readable medium having computer instructions stored thereon that are capable of being executed by a computer system to cause operations comprising:

accessing a multi-tenant dataset that includes a relational database table in which multiple tenants share one or more columns and a first tenant implements at least one added custom column;

processing the multi-tenant dataset to migrate the relational database table to non-relational underlying storage and divide the database table, including:

generating first logical table information that specifies a first logical table for a first portion of the dataset corresponding to the first tenant including the at least one added custom column;

generating second logical table information that specifies a second logical table for a second portion of the dataset corresponding to a second tenant; and

storing the first and second logical tables using non-relational key/value data structures in one or more storage elements, including:

storing entries from the first logical table for the first tenant across a first set of multiple key/value store buckets;

storing entries from the second logical table for the second tenant across a second set of multiple key/value store buckets; and

generating internal identifiers for key/value storage for names associated with the first and second logical tables according to a mapping function, wherein:

a given internal identifier uses a smaller number of bits than its corresponding name, thereby reducing storage size relative to utilizing the names for key/value storage; and

the mapping function preserves ordering among keys for a given bucket; and

retrieving, by the computing system, values from the first set of multiple key/value store buckets based on a query from the first tenant that specifies one or more keys.

13. The non-transitory computer-readable medium of claim 12 , wherein entries in respective key/value store buckets are sortable.

14. The non-transitory computer-readable medium of claim 12 , wherein the first and second sets of multiple key/value store buckets at least partially overlap.

15. The non-transitory computer-readable medium of claim 12 , wherein the storing uses partition keys to spread entries from the first and second logical tables across multiple key/value store buckets.

16. The non-transitory computer-readable medium of claim 15 , wherein the partition keys include a logical table identifier and a bucket number.

17. The non-transitory computer-readable medium of claim 16 , wherein the operations further comprise generating bucket numbers using a stable hash function.

18. The non-transitory computer-readable medium of claim 12 , wherein the mapping function provides consistency such that a first name value used during a first time interval and re-used during a second time interval is mapped to two different internal identifiers during the different first and second time intervals.

Assignments (2)

Continuity (1)

References Cited (9)