Mechanisms for deploying database clusters

    公开(公告)号:US11914580B2

    公开(公告)日:2024-02-27

    申请号:US17491102

    申请日:2021-09-30

    Abstract: Techniques are disclosed that pertain to deploying immutable instances of a system. A computer system may maintain an active generation value that indicates an immutable instance of a database system that is permitted to write data to a database. The computer system may deploy a first immutable instance of the database system and update the active generation value to permit the first immutable instance to write data to the database. The computer system may receive a request to deploy a second immutable instance of the database system that includes an update not found in the first immutable instance. The computer system may deploy the second immutable instance and update the active generation value to cause the first immutable instance to cease writing data to the database and to permit the second immutable instance to write data to the database.

    Systems and methods for workflow visualization in a multi-tenant management platform

    公开(公告)号:US11907880B2

    公开(公告)日:2024-02-20

    申请号:US17589444

    申请日:2022-01-31

    Inventor: Sönke Rohde

    CPC classification number: G06Q10/0633 G06Q10/063114

    Abstract: Embodiments described herein provide a workflow visualization tracking tool based on text mining on a conversation channel. Specifically, a communication interface receives a plurality of user activity items that are related to a shared work item among the plurality of tenants. A first sequence of status indicators representing a workflow progress of the shared work item over a plurality of time instances may be generated from the plurality of user activity items. A first subset of the plurality of user activity items are determined, which are related to at least one sub-item reflected in the first sequence of status indicators. A second sequence of status indicators representing a second workflow progress of the at least one sub-item over the plurality of time instances may be generated from the first subset of the user activity items. The first sequence of status indicators and the second sequence of status indicators may be presented in a visualization structure via a user interface.

    Advanced searching of relational data via a mobile device

    公开(公告)号:US11907210B2

    公开(公告)日:2024-02-20

    申请号:US16529951

    申请日:2019-08-02

    CPC classification number: G06F16/2425

    Abstract: A system, method and apparatus for providing for searching of relational data via a mobile communications device are provided. Information identifying searchable tables containing fields is received using a metadata application program interface (API). The metadata API provides information about a schema of the fields and a relationship of at least one of the fields to its corresponding object. Information for generating a view of relational data is provided to the mobile communications device. The view is based on at least one of the fields of at least one of the searchable tables, and the relationship between the at least one of the fields and its corresponding object. A set of user inputs is received based on the relationship presented in the view. A query definition corresponding to the set of user inputs for searching the relational data in accordance with the set of user inputs is created and stored.

    SYSTEMS AND METHODS FOR A DISTRIBUTED TRAINING FRAMEWORK USING UNIFORM CLASS PROTOTYPES

    公开(公告)号:US20240054350A1

    公开(公告)日:2024-02-15

    申请号:US18064122

    申请日:2022-12-09

    CPC classification number: G06N3/098

    Abstract: Embodiments described herein provide systems and methods for federated learning. A central system may store a neural network model which has a body of a number of layers, and a classification layer comprising class prototypes which classifies the latent representations output by the body of the model. The central system may initialize the class prototypes so that they are uniformly distributed in the representation space. The model and class prototypes may be broadcast to a number of client systems, which update the body of the model locally while keeping the class prototypes fixed. The clients may return information to the central system including updated local model parameters, and a local representation of the classes based on the latent representation of items in the local training data. Based on the information from the clients, the neural network model may be updated. This process may be repeated iteratively.

    FEDERATED DISTRIBUTED MUTATION
    158.
    发明公开

    公开(公告)号:US20240054026A1

    公开(公告)日:2024-02-15

    申请号:US17818826

    申请日:2022-08-10

    CPC classification number: G06F9/541 G06F9/547

    Abstract: Methods, systems, and devices for data processing are described. The method may include generating a federated application programming interface (API) from a plurality of data sources. One or more of the plurality of data sources may include an annotated data schema that defines one or more API endpoints for performing a distributed mutation of the plurality of data sources according to a supported distributed transaction protocol. The method may include receiving, via a single call to a single field of the federated API, a request at the federated API to perform a distributed mutation on the plurality of data sources. The method may include orchestrating, via a federation management service and the one or more API endpoints, a plurality of mutation calls to the plurality of data sources according to the distributed transaction protocol.

Patent Agency Ranking