Granted Patent B2

US 12,314,254 · App. 17/652,504 · Granted May 27, 2025

Inventors: Jiansheng Huang (San Mateo, CA); Jiaxing Liang (Bellevue, WA); Scott Ziegler (San Mateo, CA); Haowei Yu (Newark, CA); Benoit Dageville (San Mateo, CA); Varun Ganesh (San Bruno, CA)

Assignee: Snowflake Inc.

G06F16/2386G06F16/2255G06F16/2282G06F16/254

Abstract

Claims (66)

1. A method comprising:

polling, by at least one hardware processor, a first file queue to determine whether any new files have been committed to the first file queue since a last time the first file queue was polled;

accessing a data storage device to detect whether a new file has been added to the data storage device;

determining that the new file comprising new file data has been added to a database source comprising the data storage device;

processing information associated with the new file to retrieve a table identifier for an individual target table corresponding to the new file automatically in response to determining that the new file has been added to the database source comprising the data storage device;

in response to determining that the individual target table corresponding to the table identifier represents a first target table, assigning a pipe to the new file for the pipe to retrieve and direct the new file data to the first target table;

without requiring a specific command from a user of a first account and in response to determining that the new file has been added:

dividing the new file into a plurality of partitions; and

incrementally inserting each of the plurality of partitions into a respective micro-partition of a plurality of micro-partitions of the first target table partition-by-partition; and

providing a second file queue that corresponds to a second target table, the second target table different than the first target table associated with the first file queue, the first file queue corresponding to the first account and the second file queue corresponding to a second account, the first target table corresponding to the first account being isolated from the second target table corresponding to the second account.

2. The method of claim 1 , further comprising:

receiving notifications via a notification channel associated with a database source, the database source receiving files comprising file data to be ingested into the database source;

associating one or more pipes with the notification channel, each of the one or more pipes being assigned one or more target tables of the database and directing the file data from the database source to a first target table of the one or more assigned target tables; and

deploying a resource manager assigned to the notification channel.

3. The method of claim 1 , wherein the new file data comprises hashing indicating a table identification of the first target table, further comprising:

generating and maintaining metadata indicating data and user files that have been successfully committed to the database source, the metadata indicating which micro-partition and in which table the data was inserted, whether the data was successfully inserted, and when the data was inserted.

4. The method of claim 3 , wherein the metadata is registered after the data has been fully committed, and wherein the metadata is stored in a metadata micro-partition of the database source.

5. The method of claim 3 , wherein each of a plurality of pipes is assigned based on a different portion of the hashing indicating a different range of table identifications, the method further comprising consolidating metadata data micro-partitions in a background process.

6. The method of claim 5 , wherein a notification channel is assigned based on a portion of the hashing that indicates each of the different ranges of table identifications assigned to each of the plurality of pipes associated with the notification channel.

7. The method of claim 1 , further comprising assigning a notification channel to the database source in response to determining that no notification channel had previously been assigned to the database source.

8. The method of claim 1 , further comprising:

polling a notification channel to query whether the new file has been received via the notification channel;

receiving an indication via the notification channel that the new file has been received via the notification channel; and

receiving an indication from the database source that the new file has been added to the database source.

9. The method of claim 1 , further comprising temporarily storing files in an account queue associated with the database source, wherein the account queue is associated with a single account.

10. The method of claim 1 , further comprising generating a new micro-partition of the first target table, the new micro-partition comprising at least a portion of the new file data, wherein the first target table comprises a plurality of micro-partitions.

11. The method of claim 1 , wherein the database source comprises a data bucket associated with a customer account of the database, the data bucket comprising data storage containing a plurality of files.

12. A non-transitory computer readable storage media containing instructions executable by at least one processor for causing the at least one processor to perform operations comprising:

polling a first file queue to determine whether any new files have been committed to the first file queue since a last time the first file queue was polled;

accessing a data storage device to detect whether a new file has been added to the data storage device;

determining that the new file comprising new file data has been added to a database source comprising the data storage device;

processing information associated with the new file to retrieve a table identifier for an individual target table corresponding to the new file automatically in response to determining that the new file has been added to the database source comprising the data storage device;

in response to determining that the individual target table corresponding to the table identifier represents a first target table, assigning a pipe to the new file for the pipe to retrieve and direct the new file data to the first target table; and

without requiring a specific command from a user of a first account and in response to determining that the new file has been added:

dividing the new file into a plurality of partitions; and

incrementally inserting each of the plurality of partitions into a respective micro-partition of a plurality of micro-partitions of the first target table partition-by-partition; and

providing a second file queue that corresponds to a second target table, the second target table different than the first target table associated with the first file queue, the first file queue corresponding to the first account and the second file queue corresponding to a second account, the first target table corresponding to the first account being isolated from the second target table corresponding to the second account.

13. The non-transitory computer readable storage media of claim 12 , the operations further comprising:

receiving notifications via a notification channel associated with a database source, the database source receiving files comprising file data to be ingested into the database source;

associating one or more pipes with the notification channel, each of the one or more pipes being assigned one or more target tables of the database and directing the file data from the database source to a first target table of the one or more assigned target tables; and

deploying a resource manager assigned to the notification channel.

14. The non-transitory computer readable storage media of claim 12 , wherein the new file data comprises hashing indicating a table identification of the first target table.

15. The non-transitory computer readable storage media of claim 14 , the operations further comprising:

identifying the hashing in the new file data; and

identifying the first target table for the new file data based on the table identification indicated by the hashing.

16. The non-transitory computer readable storage media of claim 15 , wherein each of a plurality of pipes is assigned based on a different portion of the hashing indicating a different range of table identifications.

17. A system comprising:

at least one processor; and

one or more non-transitory computer readable storage media containing instructions executable by the at least one processor for causing the at least one processor to perform operations comprising:

polling a first file queue to determine whether any new files have been committed to the first file queue since a last time the first file queue was polled;

accessing a data storage device to detect whether a new file has been added to the data storage device;

determining that the new file comprising new file data has been added to a database source comprising the data storage device;

processing information associated with the new file to retrieve a table identifier for an individual target table corresponding to the new file automatically in response to determining that the new file has been added to the database source comprising the data storage device;

in response to determining that the individual target table corresponding to the table identifier represents a first target table, assigning a pipe to the new file for the pipe to retrieve and direct the new file data to the first target table; and

without requiring a specific command from a user of a first account and in response to determining that the new file has been added:

dividing the new file into a plurality of partitions; and

incrementally inserting each of the plurality of partitions into a respective micro-partition of a plurality of micro-partitions of the first target table partition-by-partition; and

providing a second file queue that corresponds to a second target table, the second target table different than the first target table associated with the first file queue, the first file queue corresponding to the first account and the second file queue corresponding to a second account, the first target table corresponding to the first account being isolated from the second target table corresponding to the second account.

18. The system of claim 17 , the operations further comprising:

receiving notifications via a notification channel associated with a database source, the database source receiving files comprising file data to be ingested into the database source;

associating one or more pipes with the notification channel, each of the one or more pipes being assigned one or more target tables of the database and directing the file data from the database source to a first target table of the one or more assigned target tables; and

deploying a resource manager assigned to the notification channel.

19. The system of claim 17 , wherein the new file data comprises hashing indicating a table identification of the first target table.

20. The system of claim 19 , the operations further comprising:

identifying the hashing in the new file data; and

identifying the first target table for the new file data based on the table identification indicated by the hashing.

Assignments (2)

Continuity (4)

References Cited (117)