Abstract: The present disclosure is directed to a scalable, extensible, fault-tolerant system for stateful joining of two or more streams that are not fully synchronized, event ordering is not guaranteed, and certain events arrive a bit late. The system can ensure to combine the events or link the data in near real-time with low latency to mitigate impacts on downstream applications, such as ML models for determining suspicious behavior. Apart from combining events, the system can ensure to propagate the needed entities to other product streams or help in entity resolution. If any of the needed data is yet to arrive, a user can configure a few parameters to achieve desired eventual and attribute consistency. The architecture is designed to be agnostic of stream processing framework and can work well with both streaming and batch paths.

Abstract: The present disclosure is directed to a scalable, extensible, fault-tolerant system for stateful joining of two or more streams that are not fully synchronized, event ordering is not guaranteed, and certain events arrive a bit late. The system can ensure to combine the events or link the data in near real-time with low latency to mitigate impacts on downstream applications, such as ML models for determining suspicious behavior. Apart from combining events, the system can ensure to propagate the needed entities to other product streams or help in entity resolution. If any of the needed data is yet to arrive, a user can configure a few parameters to achieve desired eventual and attribute consistency. The architecture is designed to be agnostic of stream processing framework and can work well with both streaming and batch paths.

Abstract: Techniques are provided for analyzing events incoming through a message broker and configuring a database schema for storing the events based on the analysis. The analysis is performed on all the attributes of the incoming events with reference to a primary identifier of an event source. The analysis determines the characteristics of the attributes, which facilitates development of the database schema with availability, accuracy, existence, and other factors of various attributes. Analysis is supported for various formats of events, such as AVRO, XML, complex JSON, etc. In some examples, the attributes of interest for database schema generation can be provided via a configuration for the respective databases including relational, time-series, analytical, graph, etc. Also, if a given database supports direct ingestion of data through the message broker, then the ingestion specification can be generated.

Abstract: Described herein are systems and methods for managing releases of an application in a controlled manner. A computing environment may include a first and second release of an application service, a message broker, a deployment management service and an application management service. The deployment management service may receive settings for the second release, configure the application management service in the computing environment with the settings, and transmit a request to the message broker to cause the application management service to receive messages published to a first category of the message broker. The application management service may receive a message from the message broker, identify a second category of the message broker corresponding to one of the first release or the second release, and publish the message to the second category of the message broker for consumption by one of the first release or the second release of the application service.

Abstract: Described embodiments provide systems and methods for executing one or more operations on a data source using an application programming interface (“API”) query language operations. The system can include a server that can receive a request specifying an API query language operation to execute on a data source having a structured query language (“SQL”) API. The server can identify a schema file for operating the API query language on the data source. The server can translate the API query language operation into an SQL operation using the API query language schema file configured for the data source. The server can transmit the SQL operation to the data source. The server can receive an SQL response to the SQL operation from the data source. The server can generate an API query language response based on the SQL response from the data source in response to the request from the client device.