Microsoft Technology Licensing, LLC

Edward John Triou, Jr

Jingren Zhou

Affan Arshad Dar

Hiren Shantilal Patel

Fei Xu

Osarumwemse U Agbonile

Sushil Kumar Chordia

Amol Janardan Kher

Andre Milbradt

Embodiments of the present invention allow multiple data streams to be analyzed as a single data set. The single data set may be described as a stream set herein. The multiple streams that are included in the stream set may be specified through a user script or query. For example, a query may be used to gather all streams created within a date range. The query could include one or more filters to gather certain information from the data streams or to exclude certain data streams that otherwise are in the query's range. A stream may be an unstructured byte stream of data. The stream may be created by append-only writing to the end of the stream. The stream could also be a structured stream that includes metadata that defines column structure and affinity/clustering information.

Analyzing multiple data streams as a single data object

Query testing is enabled across one or more query languages utilizing an abstract representation of a query. From this abstract representation, queries associated with one or more specific query languages are generated. Further, a verification scheme can be implemented for a specific store and/or query language to facilitate generation of an expected result for comparison with an actual result of test query evaluation. Further yet, data driving and/or modeling can be leveraged to further aid query testing.

Query builder for testing query languages

A model based testing framework is provided that can be incorporated into and used to supplement existing testing approaches. Via annotation of custom attributes and requirements, the framework provides the ability to author models in the same language as code (e.g., test code) that interacts with the model. Optionally, the model-based testing techniques are accessible and changeable at run-time so that tests and models that are executing can be dynamically altered. Optionally, weights (and other techniques) may be exerted on one or more portions of the model to influence the execution of the tests. The ability to execute the model-based testing techniques on some portions of test code, while running hand-crafted or custom test code on other portions, is also provided.

Model based testing language and framework

Test result data can be classified across multiple tests, test scenarios, and lab runs. Test result files can be parsed by extracting information from them. Extracted information can be compared to failure information in a database. If a match is found, the extracted information can be linked to the failure, creating a history for each failure. New failures can be identified when no match is found. Failure data can be cross-referenced to further aid in results analysis. For each failure, lists of useful information can be accessed. Analysis information can be associated with failures, for example whether the failure is new, occurred previously, is for a different reason than expected, or has been added to a baseline of expected failures. A Graphic User Interface ('GUI') is also provided to expose the analyzed results to the result analyzers.

Growing community of inventors

Duvall, WA, United States of America

Edward John Triou, Jr