Snowflake Inc.

Oracle International Corporation

Snowflake Computing Inc.

Allison Waingold Lee

Thierry Cruanes

Benoit Dageville

Jiaqi Yan

Marcin Zukowski

Philipp Thomas Unterbrunner

Robert Muglia

Shrainik Jain

Mohamed Zait

Martin Hentschel

Peter Povinec

Prasanna Rajaperumal

Subramanian Muralidhar

Justin Langseth

Matthew J Glickman

Christian Kleinerman

David Schultz

Igor Demura

Qiuye Jin

Stratis Viglas

Daniel Freundel

Yongsik Yoon

Boyung Lee

Florian Andreas Funke

Artin Avanes

Monica J Holboke

Bowei Chen

Rafi Ahmed

William H Waddington

Khalid Zaman Bijon

Simon Holm Jensen

Sangyong Hwang

Nicola Dan Onose

Li Yan

Xuelai Cui

Daniel N Meredith

Zixi Zhang

Sunil Parrokkaren Chakkappen

Hong Su

Ali Cakmak

Hari Sankar Sivarama Subramaniyan

Andrew Witkowski

Unmesh Jagtap

Dinesh Das

Yali Zhu

Phillip Thomas Unterbrunner

A data platform creates an application in a data-provider account, where the application includes one or more application programming interfaces (APIs) corresponding to one or more underlying code blocks. The data platform shares homomorphically encrypted provider data with the application in the data-provider account. The data platform installs, in a data-consumer account, an application instance of the application. The data platform shares homomorphically encrypted consumer data with the application instance in the data-consumer account. The data platform invokes one or more of the APIs of the application instance to execute respective associated underlying code blocks, which are not visible to the data-consumer account, and which operate on the shared homomorphically encrypted provider data and the shared homomorphically encrypted consumer data. The data platform saves homomorphically encrypted output of the one or more respective associated underlying code blocks locally within the data-consumer account.

Data clean rooms using defined access with homomorphic encryption

A database system facilitates secure data sharing by implementing projection constraints within a query processing framework. Upon receiving a query directed to a shared dataset, the system, utilizing hardware processors, identifies a subset of data within the dataset that is subject to a projection constraint policy. The applicability of the projection constraint is determined based on the context of the query, which is derived from a data sharing agreement. The system processes the query by selectively restricting the projection of data values from constrained columns, while allowing specific operations to be performed on the data. The output generated in response to the query is compliant with the projection constraint policy, providing derived data based on the allowed operations without revealing the actual data values. This ensures the confidentiality of sensitive information while enabling collaborative data analysis and sharing among various users of the database system.

Projection constraint policies in a database system

Systems, methods, and devices for sharing materialized views in multiple tenant database systems. A method includes defining a materialized view over a source table that is associated with a first account of a multiple tenant database. The method includes defining cross-account access rights to the materialized view to a second account such that that second account can read the materialized view without copying the materialized view. The method includes modifying the source table for the materialized view. The method includes identifying whether the materialized view is stale with respect to the source table by merging the materialized view and the source table.

Sharing materialized views in multiple tenant database systems

Systems, methods, and devices for implementing secure user-defined function (UDF) in a multi-tenant database system are disclosed. A method includes receiving a grant to access a share object comprising usage functionality associated with a secure UDF to underlying data. The method includes accessing the share object using the grant. The method includes causing a share component to implement the secure view and the usage functionality associated with the secure UDF.

Implementing secure user-defined functions in a multi-tenant database system

In an embodiment, an application is created on a data-provider platform. The application includes one or more application programming interfaces (APIs) corresponding to one or more underlying code blocks. Provider data is shared with the application on the data-provider platform. An application instance of the application is installed in a trusted execution environment (TEE). The application instance includes one or more APIs corresponding to the one or more APIs in the application on the data-provider platform. Consumer data is shared with the application instance from a data-consumer platform. One or more of the APIs of the application instance are invoked to execute, on the TEE, respective associated underlying code blocks that are not visible on the TEE. The output of the one or more respective associated underlying code blocks is saved to the data-consumer platform.

Data clean rooms using defined access in trusted execution environment

A system to identify optimal cloud resources for executing workloads. The system deduplicates historical client queries based on a workload selection configuration to determine a grouping of historical client queries. The system generates a workload based on at least a portion of the grouping of historical client queries. The system repeatedly executes a test run of the workload using resources of a cloud environment to determine whether there is a performance difference in the test run. The system, in response to determining that there is no performance difference, identifies one or more sets of decreased resources of the cloud environment. The system re-executes the test run using the one or more sets of decreased resources of the cloud environment to determine whether there is a performance difference in the test run that is attributed to the one or more sets of decreased resources of the cloud environment.

Optimized identification of performance regression

In an embodiment, a data platform creates an application in a data-provider account. The application includes one or more APIs corresponding to one or more underlying code blocks. The data platform shares provider data with the application in the data-provider account, and also installs, in a data-consumer account, an application instance of the application. The application instance includes one or more APIs corresponding to the one or more APIs in the application in the data-provider account. The data platform shares consumer data with the application instance in the data-consumer account, and invokes one or more of the APIs of the application instance to execute respective associated underlying code blocks, which are not visible to the data-consumer account. The data platform also saves output of the one or more respective associated underlying code blocks locally within the data-consumer account.

Data clean rooms using defined access

A method, apparatus, and system for join operations of a plurality of relations that are distributed over a plurality of storage locations over a network of computing components.

Adaptive distribution method for hash operations

A query directed to database data stored across a set of files is received. The query includes predicates applied against metadata of the set of files. One or more files are removed from the set of files whose metadata does not satisfy a predicate of the plurality of predicates. Rows of the one or more remaining files of the set of files are filtered based on the one or more predicates.

Utilizing metadata to prune a data set

A system and method for managing data storage and data access with querying data in a distributed system without buffering the results of intermediate operations in persistent storage. The method includes pushing, by a first operator, as a result of executing a query plan, an intermediate result to a plurality of secondary operators. Processing of the intermediate result by the plurality of secondary operators is delayed to coordinate generation of a plurality of second results based on beginning a final consumption operation of the plurality of second results.

Query plans for analytic SQL constructs

The subject technology receives a query, the query including a set of statements for performing the query. The subject technology populates a compilation context based at least in part the query. The subject technology invokes a compiler to perform a compilation process based on the compilation context. The subject technology performs a lookup operation on a stored plan cache for an exact match based on information from the compilation context. The subject technology, in response to determining an exact match, determines whether the particular query plan requires re-compilation based on a data dependent optimization. The subject technology determines whether a plan cache entry corresponding to the particular query plan includes a data property constraint. The subject technology determines whether the data property constraint still holds based on a set of data properties.

Query plan cache in database systems

Queries executed against a materialized view can execute up to orders of magnitude faster than equivalent queries on a source (or base) table. However, although a query can reference a materialized view directly, a user (e.g., query author) may not know about a relevant materialized view. Moreover, if a source table has multiple materialized views generated, the user may not know which materialized view to reference in the query. Thus, embodiments of the present disclosure provide techniques for automatically rewriting queries directed to a source table to utilize existing materialized views.

Transient materialized view rewrite

A system for enforcing projection constraints on data values stored in specified variables of a shared dataset of a cloud data platform. A request is received from a first account of the cloud data platform that identifies a first operation to be performed on the shared dataset. A first set of data, including data accessed from a first variable, is accessed from the shared dataset to use in performing the first operation. A projection constraint policy attached to the first variable of the shared dataset is determined, and the projection constraint policy is further determined to be enforced based on the request. Based on the first set of data and the first operation, an output to the first request is generated.

Projection constraints enforced in a database system

The subject technology receives a query plan, the query plan comprising a set of query operations, the set of query operations including at least one aggregation and at least one join operation. The subject technology analyzes the query plan to identify an aggregation that is redundant. The subject technology removes the aggregation based at least in part on the analyzing. The subject technology determines at least one aggregation property corresponding to at least one query operation of the query plan. The subject technology inserts at least one adaptive aggregation operator in the query plan based at least in part on the at least one aggregation property, the at least one aggregation property comprising a set of aggregation properties. The subject technology provides a modified query plan based at least in part on the inserted at least one adaptive aggregation operator in the query plan.

Placement of adaptive aggregation operators and properties in a query plan

A constraint system enforces projection constraints on data values stored in specified columns of a shared dataset when queries are received by a database system. A projection constraint identifies that the data in a column may be restricted from being projected (e.g., presented, read, outputted) in an output to a received query, while allowing specified operations to be performed on the data and a corresponding output to be provided. For example, the projection constraint may indicate a context for a query that triggers the constraint, such as based on the user that submitted the query. Enforcing projection constraints on queries received at the database system allows for data to be shared and used anonymously by entities to perform various operations without the need to tokenize the data.

Projection constraints in a query processing system

Growing community of inventors

San Mateo, CA, United States of America

Allison Waingold Lee