Abstract: Systems and methods include detection of deletion of a record of a database table due to a database transaction, creation of a log entry including one or more key field values of the record of the database table, an identifier of a user associated with the database transaction, and an indicator of a delete operation, determination of whether the identifier identifies an archiving user, and, in a case that the identifier identifies an archiving user, change the indicator to an indicator of an archive operation to create a changed log entry.

Abstract: In one embodiment, a method receives a search query including terms formed in a natural language expression. The terms are included in entities that are categorized in types of entities. The method analyzes the entities for the terms to generate a set of branch queries that include different combinations of entities for the terms. The set of branch queries are ranked based on the types of entities associated with entities of the respective branch queries and a branch query is selected based on the ranking for the branch query. The method uses the entities of the branch query to perform a search of a structured database of structured objects.

Abstract: Data recovery in a database includes performing a backward scan of a transaction log that records transactions made on the database. The backward scan includes traversing log records starting from a latest-in-time log record toward the beginning of the transaction log. Each log record can be replayed to recover data for a database row associated with the log record. An earlier-in-time log record is skipped if the associate database row has been recovered by later-in-time log record.

Abstract: The present disclosure describes methods, systems, and computer program products for providing additional stack trace information for time-based sampling (TBS) in asynchronous execution environments. One computer-implemented method includes determining whether time-based sampling is activated to capture a time-based sampling data during execution of a JavaScript function; in response to determining that the time-based sampling is activated to capture the time-based sampling data, determining whether a callback stack trace is active; in response to determining that the callback stack trace is active, loading the callback stack trace; retrieving a current stack trace of the JavaScript function; and saving the loaded callback stack trace and the current stack trace of the JavaScript function as the time-based sampling data.

Abstract: A method, a system and a computer program product for performing software configuration alignment. An indication of a source configuration change of one or more source computing processes of a source computing system is received from the source computing system. A determination is made whether the source configuration change is applicable to one or more target computing processes of a target computing system. The target computing system is communicatively coupled to the source computing system. Using the source configuration change, a target configuration change for implementation in the target computing processes of the target computing system is generated. The generated target configuration change is transmitted to the target computing system. The generated target configuration change is implemented in the target computing processes of the target computing system.

Abstract: Disclosed herein are system, method, and computer program product embodiments for utilizing look-ahead-staging (LAS) to guarantee the ability to rollback and reconstruct a package while minimizing locking duration and enabling multiple packages to be processed in a data pipeline simultaneously. An embodiment operates by receiving a package from a source system for processing through a data pipeline. The embodiment stores the package in a persistent storage together with a respective package status. The embodiment transmits the package to the data pipeline in response to the storing. The embodiment receives a commit notification for the package from a target system in response to the transmitting. The embodiment then removes the package from the persistent storage in response to receiving the commit notification for the package.

Abstract: Techniques for processing combinations of timeseries data and time-dependent semantic data are provided. The timeseries data can be data from one or more Internet of things (IOT) devices having one or more hardware sensors. The semantic data can be master data. Disclosed techniques allow for time dependent semantic data to be used with the timeseries data, so that semantic data appropriate for a time period associated with the timeseries data can be used. Changes to semantic data are tracked and recorded, where the changes can represent a new value to be used going forward in time or an update to a value for a prior time period. Timeseries data and semantic data can be stored with identifiers that facilitate their combination, such as date ranges, identifiers of analog world objects, or identifiers for discrete sets of semantic data values.

Abstract: The present disclosure pertains to computer readable documents including structured data content for visualizing a structure with a plurality of structure elements and a plurality of relationships between the plurality of structure elements. The document includes the structured data content, a data type indicator, and a structure type indicator. The data type indicator is used to determine a parsing algorithm, The parsing algorithm and the structure type indicator are used to parse the structured data content to determine the structure elements and the relationships between the structure elements. The lexical structure of the structured data content reflects the relationships between the structure elements. A visualization of the structure can be automatically generated. The visualization can be linked to or embedded within visualizations of other structures.

Abstract: Implementations of the present disclosure include receiving a query, the query including an expression macro (EM), processing the query to provide a raw parse tree, the raw parse tree including an initial node representative of the EM, retrieving metadata corresponding to the EM, the metadata including a definition string, replacing the initial node with a node based on the definition string to provide a consumable parse tree, and executing the query within the database system using the consumable parse tree to provide a query result.

Abstract: A method for machine-learning based user interface generation is provided. The method may include training a machine-learning model by processing, with the machine-learning model, a training data that includes one or more sequences of software applications accessed by a user at a client. The machine-learning model may be applied to identify a software application based at least on a probability of the user at the client accessing the software application. A user interface may be generated for display at the client. The user interface may provide, to the user at the client, access to the software application. Related systems and articles of manufacture, including computer program products, are also provided.

Abstract: A computer-implemented method for failure diagnosis using fault tree can include: receiving a fault tree comprising a node representing a top event, a plurality of nodes representing respective basic events, and one or more logic gates connecting the plurality of nodes representing the respective basic events to the node representing the top event; obtaining reliability parameters corresponding to the basic events; calculating fault tree importance measures corresponding to the basic events; calculating failure impact factors of the top event corresponding to the basic events, wherein the failure impact factors of the top event are products of the corresponding reliability parameters and the corresponding fault tree importance measures; ranking the basic events based on the failure impact factors of the top event; and identifying a most significant contributor to the top event, wherein the most significant contributor is a basic event having the highest failure cause probability of the top event.

Abstract: Methods, systems, and computer-readable storage media for providing a binary classifier include receiving a biased dataset, the biased data set including a plurality of records, each record being assigned to a class of a plurality of classes, one class including a majority class, performing data engineering on at least a portion of the biased dataset to provide a revised dataset, providing a trained deep autoencoder (DAE) by training a DAE using only records assigned to the majority class from the revised dataset, the trained DAE including a binary classifier that classifies records into one of the majority class and a minority class, validating the trained DAE using validation data that is based on at least a portion of the biased dataset, and providing the trained DAE for production use within a production system.

Abstract: Systems, methods, and computer media are described for memory-efficient unarchiving. Data items can be archived and removed from primary data store over time. An archive block containing both a data item of interest and additional data items can be located. The data item of interest, as well as the additional data, can be loaded into a secondary data store. The data item of interest can then be extracted from the secondary data store and reloaded into the primary data store. The secondary data store can then be cleared. Memory-efficient unarchiving more efficiently uses computing resources and unarchives only the data that is needed.

Abstract: Recovery of a database system by taking the database system offline is initiated. Thereafter, recovery operations specified by a redo log of the database system are replayed. A cleanup log is generated that identifies cleanup operations occurring during the replay of the recovery operations for garbage collection. Concurrent with the startup of the database, garbage collection of the cleanup operations as specified in the database savepoint is initiated. In addition, concurrent with the replay of the recovery operations, garbage collection of the cleanup operations specified by the cleanup log is initiated. The database system is later brought online after all of the recovery operations are replayed.

Abstract: Described herein is a framework for classifying social media inputs. In accordance with one aspect of the framework, one or more social media inputs is acquired from one or more social media platforms. The social media inputs are cleaned to remove redundant elements. One or more features are extracted from the cleaned social media inputs. The social media inputs are classified by a trained classifier into predefined categories using the extracted one or more features.

Abstract: Methods, systems, and computer-readable storage media for determining a sub-set of services, each service in the sub-set of services being in an overload status and having a SSF associated therewith, providing statistics associated with each service in the sub-set of services and relating service call statistics to each API group of a set of API groups, determining a set of sessions based on the service call statistics, each session in the set of sessions associated with at least one service in the sub-set of services having a service call statistic that exceeds a threshold service call statistic, defining an execution group and a waiting group, determining that a session is in the waiting group, and in response, blocking a request of the session from being processed by a service in the sub-set of services, and providing a wait indication to a user associated with the session.

Abstract: A system and method including the reception of an input of a set of textual terms including a subject matter parameter value and an indeterminate term parameter value; automatically determining, by a machine learning process, the subject matter parameter value is subsumed within a specified data model including the indeterminate term parameter value; automatically processing the indeterminate term parameter value to execute a combination of literal, systemic, historical perspective, and teleological interpretations thereof to generate an overall assessment that includes a non-arbitrary interpretation of the indeterminate term parameter value; and presenting a representation of the set of textual terms expanded to include the non-arbitrary interpretation of the indeterminate term parameter value.

Abstract: A computer-implemented method, a computer system and a computer program for controlling electric vehicle charging infrastructure that is used by a plurality of electric vehicles, wherein the infrastructure comprises a plurality of charging points. The method comprises receiving schedule information for the electric vehicles, the schedule information including, for each of the electric vehicles, at least one vehicle power characteristic. The method further comprises determining at least one power constraint for each of the electric vehicles based on the vehicle power characteristic. The method further comprises receiving an arrival indication for a respective one of the electric vehicles. The method further comprises assigning the respective electric vehicle to charge at one of the charging points for at least one time interval according to the respective power constraint.

Abstract: Techniques are described for predicting test scenarios and generating test case documents and/or automation scripts using machine-learning algorithms. For example, input may be received representing a web page, and an HTML Document Object Model (DOM) of the web page may be analyzed. From the DOM, a plurality of HTML elements may be extracted and processed by a machine-learning algorithm. Based on the processed plurality of HTML elements, a plurality of predictions for test case scenarios may be generated, and converted into a set of human-readable instructions, such as a test case document, and/or a set of machine-readable instructions, such as an automation script. In some instances, a user selection of at least one predicted test scenario from a displayed list of one or more predicted test scenarios is received and corresponding human-readable instructions and/or machine-readable instructions are generated for the selected scenario(s).

Abstract: System computing resources are dynamically managed by a resource management layer together with resource server pool. A group monitor collects data of server operation (e.g., CPU usage, memory consumption, queue size) in a particular group (e.g., customer-facing servers, job servers). The group monitor: calculates a weighted score for each server, sorts the scores of each server, maintains a sequence of scores, and periodically reports the score list to a resource manager. When an incoming request is directed to a particular server group, the group monitor redirects the request to a lowest-scoring server, and recalculates/sorts the score list. The resource manager monitors and throttles the servers in each group. Where servers in a group are experiencing high workload, the manager adds to the group server(s) from the resource pool. Conversely, where server group members are experiencing a low workload, the manager removes the lowest scoring server(s) to the resource pool.