Atscale, Inc.

Matthew Baird

Sarah Gerweck

David P Mariani

David Vigdor Schreibman

Josh Klahr

David A Ross

Rouzbeh Safaie

Joseph Barefoot

DaRen Drummond

Daniel J Dosch

Trystan R Leftwich

Gaurav Nitin Shetti

James Lai

A data query and retrieval system receives a query request and identifies a plurality of data warehouses, or stores, where data needed to satisfy the query is stored. Data items, or fields, called for by the query request are grouped into sets based on fields stored at the same data warehouse. Individual data connections are generated for each data warehouse accessed, and a database view defined for the fields available at each respective data warehouse. The individual views are invoked in a single query statement by a query processor for accessing the respective data warehouses via the corresponding database view. In this manner, a query operator or user need not define multiple query expressions or perform separate queries for accessing data fields or dimensions residing at different data warehouses.

Distributed queries through dynamic views

Techniques are presented for efficiently provisioning application-agnostic resource access to a variety of applications without modification to the native access control mechanisms of the applications and without transmission of a user's credentials over the network. A user of an application is authenticated by an authorization provider. An access token for the authenticated user is generated. A session password is generated based at least in part on the access token. The session password is applied by the user to the native access control mechanism of an application to facilitate access to resources (e.g., set of subject data) by the application. The resource access is achieved without modification to the native access control mechanism of the application and without transmission of the credentials (e.g., username, password, etc.) of the user over the network.

Application-agnostic resource access control

Techniques are presented for rule-based selection of alternate data sources for multidimensional data statements. A virtual multidimensional data model is implemented to represent datasets that are accessed at various data sources. Derivative cubes generated from the virtual multidimensional data model are structured to have a respective data source metadata layer that is populated at data statement execution time to identify a target data source. Data source selection rules are established to map the attributes of data statements to target data sources. The data source selection rules are evaluated subject to data statement attributes derived from detected data statements to dynamically select target data sources for the data statements. The derivative cubes codify, in their respective data source metadata layers, a unique set of data source attributes identifying the target data sources. The derivative cubes are accessed to facilitate execution of the data statements on datasets at the target data sources.

Rule-based data source selection

Techniques are presented for performing enhanced real-time data statement monitoring and control without requiring modifications to data analysis client applications or data storage environments. A data statement processing layer is implemented between the data analysis client applications and the data storage environments. A data statement management agent interacts with an event bus in the data statement processing layer. The data statement management agent accesses event objects on the event bus that are generated during the processing of data statements. Rules are applied to select particular event objects. The selected event objects are transformed into event messages that are delivered to the client applications to monitor the processing of the data statements. Control commands received from the client application are transformed into event objects that are issued to the event bus to control the processing of the data statement.

Data statement monitoring and control

Techniques are presented for rule-based selection of query engines for performing data operations on a variety of datasets stored in heterogeneous data storage environments. Data statements for a subject dataset are mapped to a logical data structure representation of the subject dataset to formulate a logical plan for the data statements. The data statements are analyzed to derive a set of statement attributes associated with the data statements. Query engine attributes describing the query engines available to query various instances of the subject dataset are accessed. A set of query engine selection rules are evaluated subject to the statement attributes and/or the query engine attributes to dynamically select a target query engine for the data statements. A physical plan structured for the target query engine is generated from the logical plan. The physical plan is executed on the target query engine to perform data operations on the subject dataset.

Query engine selection

The present disclosure provides a detailed description of techniques for data model design collaboration using semantically correct collaborative objects. Certain embodiments comprise detecting various interaction events from designers collaborating on designing a virtual multidimensional data model to generate unit of work objects characterizing certain design transactions invoked by the interaction events. The unit of work objects are used to determine a set of committed modifications to the virtual multidimensional data model design for broadcasting to the collaborating designers. Some embodiments comprise unit of work objects based on various multidimensional data model rules, syntactic rules, semantic rules, and/or other criteria to determine valid committed modifications. Other embodiments comprise unit of work objects from multiple design transactions that are merged to determine the committed modifications. Further embodiments comprise committed modifications that are derived from the differences between the design changes and the then current virtual multidimensional data model.

Growing community of inventors

Belmont, CA, United States of America

Matthew Baird