Microsoft Technology Licensing, LLC

Avrilia Floratou

Ashvin Agrawal

Srikanth Kandula

Ke Wang

Carlo Aldo Curino

Alekh Jindal

Ishai Menache

Sriram S Rao

Joseph Naor

Christopher William Douglas

Karthik Saligrama Ramachandra

Virajith Jalaparti

Joyce Yu Cahoon

Kameswara Venkatesh Emani

Daniel E Musgrave

Mainak Ghosh

Vu Minh Le

Fang Liu

Vivek Gupta

Subramaniam Venkatraman Krishnan

Irene Rogan Shaffer

Anja Gruenheid

Ehimwenma Nosakhare

Jordan Joseph Henkel

Dalitso Hansini Banda

Fotios Psallidas

Wee Hyong Tok

Jason Wang

Andreas Christian Mueller

Ling Zhang

Shaily Jignesh Fozdar

Richard Anarfi

Daniel Muñoz Huerta

David Joseph Donahue

Noelle Yanhui Li

Gust Verbruggen

Abhishek Narain

Shaleen Deep

Yan Qiu

Neha Godwal

MohammadHossein Namaki

Yinghui Wu

Regular expression ('regex') decomposition and evaluation is disclosed. In an aspect of the disclosure, a literal component and a pattern component in a regex are determined. A plurality of evaluation processes is determined based on the regex, including a first evaluation process configured to identify text that matches the literal component and the pattern component. The evaluation processes are executed with respect to a sample of data to determine performance characteristics including a performance characteristic respective to each evaluation process. An evaluation process of the evaluation processes is selected based on the determined performance characteristics. The selected evaluation process is executed with respect to further data to retrieve results that satisfy the selected evaluation process. In another aspect, the pattern component is decomposed into subcomponents. In another aspect, the evaluation processes include an evaluation process that specifies a process to provide the regex to a regex engine for execution thereof.

Regular expression decomposition and evaluation

A computer-implemented method for performing natural language-based data integration includes causing execution of a data integration application on a remote device via a network and causing surfacing of a GUI corresponding to the data integration application on a display of the remote device. The method includes receiving, via the GUI, a natural language input representing a data integration task, generating, via an LLM, a set of ordered activities corresponding to the data integration task represented by the natural language input, and selecting, via the LLM, one or more APIs for performing each activity within the set of ordered activities. The method also includes generating a data pipeline based on the set of ordered activities and the API(s) for performing each activity, as well as back-translating the data pipeline to a desired data format for execution by the data integration application.

Natural language-based data integration

Systems, methods, and devices are described for performing scalable data processing operations. A queue that includes a translatable portion comprising indications of data processing operations translatable to data queries and a non-translatable portion comprising indications of non-translatable data processing operations is maintained. A determination that a first data processing operation of a first code block statement is translatable to a database query is made. An indication of the first data processing operation is included in the translatable portion of the queue. Responsive to a determination that a second data processing operation of a second code block statement is undeferrable, the translatable portion of the queue is compiled into a database query. An execution of the database query to be executed by a database engine to generate a query result is caused. A result dataset corresponding to the query result is transmitted to an application configured to analyze the result dataset.

System and method for scalable data processing operations

Systems, methods, and devices are described for enabling a user to import a library into a computer program under development. The library includes a data storage interface, one or more semantic objects, and one or more data manipulation or data analysis operations. A user is able to reference code of the library within the computer program under development to generate a dataset from data obtained via the data storage interface and associate the one or more semantic objects with the dataset to generate a semantically-annotated dataset. Systems, methods, and devices enable, based on the importing: the user to invoke a semantic-guided operation of the library that utilizes the semantically-annotated dataset to infer an aspect of a data manipulation or data analysis operation to be performed on the semantically-annotated dataset; or the suggestion of a data manipulation or data analysis operation to the user based on the semantically-annotated dataset.

System and method for semantic aware data science

A system is described herein for mitigating slow process instances in a streaming application. The system includes a slow process instance candidate identifier configured to identify, based on a relative watermark latency, a set of slow process instance candidates from among a plurality of process instances that comprise the streaming application. The system further includes a set of filters configured to remove false positives from the set of slow process instance candidates. The filters account for window operations performed by the process instances as well as stabilization time needed for downstream process instances to stabilize after a slow upstream process instance is mitigated by a mitigation implementer, which may also be included in the system.

Mitigating slow instances in large-scale streaming pipelines

A system enables tracking machine learning ('ML') model data provenance. In particular, a computing device is configured to accept ML model code that, when executed, instantiates and trains an ML model, to parse the ML model code into a workflow intermediate representation (WIR), to semantically annotate the WIR to provide an annotated WIR, and to identify, based on the annotated WIR and ML API corresponding to the ML model code, data from at least one data source that is relied upon by the ML model code when training the ML model. A WIR may be generated from an abstract syntax tree (AST) based on the ML model code, generating provenance relationships (PRs) based at least in part on relationships between nodes of the AST, wherein a PR comprises one or more input variables, an operation, a caller, and one or more output variables.

Tracking provenance in data science scripts

Methods, systems, apparatuses, and computer program products are provided for prefetching data. A workload analyzer may identify job characteristics for a plurality of previously executed jobs in a workload executing on a cluster of one or more compute resources. For each job, identified job characteristics may include identification of an input dataset and an input bandwidth characteristic for the input dataset. A future workload predictor may identify future jobs expected to execute on the cluster based at least on the identified job characteristics. A cache assignment determiner may determine a cache assignment that identifies a prefetch dataset for at least one of the future jobs. A network bandwidth allocator may determine a network bandwidth assignment for the prefetch dataset. A plan instructor may instruct a compute resource of the cluster to load data to a cache local to the cluster according to the cache assignment and the network bandwidth assignment.

Cache and I/O management for analytics over disaggregated stores

Growing community of inventors

Mountain View, CA, United States of America

Avrilia Floratou