Snowflake Inc.

Emc Corporation

Google Inc.

Isilon Systems, Inc.

Tyler Arthur Akidau

Tyler Jones

Daniel E Sotolongo

Istvan Cseri

Daniel Mills

Dinesh Chandrakant Kulkarni

Zhuo Zhang

Aaron Passey

Benoit Dageville

Thierry Cruanes

Fabian Hueske

Leon Papke

Neal Thomas Fachan

Asif Arif Daud

Di Fei Zhang

Ilya Maykov

Prasanna Rajaperumal

Craig D Chambers

Frances J Perry

Reuven Lax

Robert Bradshaw

Yevgeniy Kogan

Ben Chambers

Dzmitry Pauliukevich

Subramanian Muralidhar

Peter John Godman

Ganeshan Ramachandran Iyer

Jonathan Mark Walton

Nate Dire

Zachary Merlynn Loafman

Brian Eng

A streaming ingest platform can improve latency and expense issues related to uploading data into a cloud data system. The streaming ingest platform can organize the data to be ingested into per-table chunks and per-account blobs. This data may be committed and may be made available for query processing before it is ingested into the target source tables. This significantly improves latency issues. The streaming ingest platform can also accommodate uploading data from various sources with different processing and communication capabilities, such as Internet of Things (IOT) devices.

Real-time streaming data ingestion into database tables

Techniques for continuous ingestion of files using custom file formats are described. A custom file format may include formats not natively supported by a data system. Unstructured files (e.g., images) may also be considered custom file formats. A custom file format may be set using a user defined table function and scanner options.

Continuous ingestion of custom file formats

Techniques for triggering pipeline execution based on data change (transaction commit) are described. The pipelines can be used for data ingestion or other specified tasks. These tasks can be operational across account, organization, cloud region, and cloud provider boundaries. The tasks can be triggered by commit post-processing. Gates in the tasks can be set up to reference change data capture information. If the gate is satisfied, tasks can be executed to set up data pipelines.

Cross-organization and cross-cloud automated data pipelines

Techniques described herein can accelerate change data capture determinations such as stream reads, which show changes made to a table between two points in time. Three distinct row bitests that mark deleted, updated, inserted, rows in micro-partitions can be added as metadata for the table. These bitsets can be generated during DML operations and then stored as metadata of the new partition generated by the DML operations. The bitsets can then be used to generate streams showing the changes in the table between two points in time (changes interval).

Accelerating change data capture determination using row bitsets preliminary

A system for a materialized table (MT) refresh using multiple processing pipelines includes at least one hardware processor coupled to memory storing instructions. The instructions cause the at least one hardware processor to perform operations including determining dependencies among a plurality of intermediate MTs generated from a source MT. The source MT uses a table definition with a query on one or more base tables and a lag duration value. A graph snapshot of dependencies among the plurality of intermediate MTs is generated. Processing pipelines are configured. Each of the processing pipelines corresponds to a subset of the plurality of intermediate MTs indicated by the graph snapshot. Responsive to detecting an instruction for a refresh operation on the source MT, refreshes on corresponding intermediate MTs of the plurality of intermediate MTs in each processing pipeline of the processing pipelines are performed to complete the refresh operation on the source MT.

Materialized table refresh using multiple processing pipelines

A method includes retrieving a plurality of materialized tables (MTs). Each of the plurality of MTs includes a lag duration and refers to a corresponding base table of a plurality of base tables. The lag duration indicates a maximum time period that a result of a prior refresh of a query on the corresponding base table can lag behind a current time instance. A plurality of time instances for the MT is determined based on the lag duration and a number of prior refreshes of the corresponding base table. A plurality of aligned time instances for the plurality of MTs is determined based on the plurality of time instances for each of the plurality of MTs. Refresh operations are scheduled for the plurality of MTs at one or more of the plurality of aligned time instances that are within the maximum time period.

Query execution using materialized tables

Hybrid tables can be used in different use-case scenarios. Hybrid tables provide a flexible mechanism to support files and data in different formats while providing access to the different types of data as part of one table. This flexibility can allow the use of hybrid tables in data lake or other similar environments.

Optimized processing of data in different formats

A system or persistent table may be generated storing changelog information of a primary base table. The system table may then be used to create streams of relevant information. In some examples, the streams may read from the system table for information past a retention period of the primary table while reading from the primary table information in the retention period.

Streams using persistent tables

Provided herein are systems and methods for a database object (e.g., materialized table) configuration including using the database object for incremental maintenance of query results. For example, a method includes creating a materialized table using a table definition. The table definition includes a query on one or more base tables and a lag duration value. The lag duration value indicates a maximum time period that a result of a prior refresh of the query can lag behind a current time instance. A delta set for the one or more base tables is determined. The delta set includes at least one change applied to the one or more base tables since the prior refresh of the query. The delta set is merged with the materialized table to perform an incremental update of the materialized table.

Incremental maintenance of query results

Provided herein are systems and methods for a database object (e.g., materialized table) configuration including scheduling refreshes of the materialized table. For example, a method includes determining a dependency graph for a first MT. The dependency graph comprises a second MT from which the first MT depends. The first MT includes a query on one or more base tables and a lag duration value. The lag duration value indicates a maximum time period that a result of a prior refresh of the query can lag behind a current time instance. A tick period is selected for a set of ticks based on the lag duration value. The set of ticks corresponds to a set of aligned time instances. Refresh operations are scheduled for the first and second MTs at corresponding time instances from the set of aligned time instances. The corresponding time instances are separated by the tick period.

Execution and consistency model for materialized tables

Growing community of inventors

Seattle, WA, United States of America

Tyler Arthur Akidau