Abstract: A computer-implemented method includes: connecting, by a computing device, to a database using an outbound connection, wherein the computing device is an information technology (IT) product in a private network and the database is outside the private network; receiving, by the computing device, a response from the database, the response including a command; executing, by the computing device, the command; and sending, by the computing device, result data to the database, wherein the result data is data that results from executing the command on the computing device.

Abstract: One or more processors receive resource type and capability information and activity information of workloads of a domain. A first model is generated and trained to map the resource information to the activity information of domain workloads. The activity information is decomposed into a set of activity core elements (ACEs). The one or more processors generate a second model, wherein the second model is trained to predict a set of resource types and resource capabilities of the respective resource types, based on an input of the first set of ACEs decomposed from the activity information of the workloads of the domain. The one or more processors receive a second set of ACEs that are decomposed from activities associated with an unprecedented workload, and the one or more processors generate a predicted set of resources to perform the second set of ACEs.

Abstract: Intelligent management of stub files in hierarchical storage is provided by: in response to identifying a file to migrate from a file system to offline storage, providing metadata for the file to a machine learning engine; receiving a stub profile for the file from the machine learning engine that indicates an offset from a beginning of the file and a length from the offset for previewing the file; and migrating the portion of the file from the file system to an offline storage based on the stub profile. In some embodiments this further comprises: monitoring file system operations; in response to detecting a read operation of the portion of the file: determining a file type; providing file data to the machine learning engine; and performing a supervised learning operation based on the file type and the file data to update the machine learning engine.

Abstract: A method that includes joining, by a first node, a network connecting a set of nodes, where each node has an agent and is associated with a sensor, and where the sensor generates sensor data, and where each agent sends and receives broadcast messages, and generating, by the sensor, sensor data associated with a second node. The method also includes analyzing, by the agent, the sensor data, where the analyzing causes a detection of a first fault condition of the second node, and where the first fault condition is an indication of a problem with the second node, generating, by the agent, a broadcast message, where the broadcast message includes the first fault condition, and sending, by the first node, the broadcast message to at least one member of the set of nodes via the network.

Abstract: A computer-implemented method includes: connecting, by a computing device, to a database using an outbound connection, wherein the computing device is an information technology (IT) product in a private network and the database is outside the private network; receiving, by the computing device, a response from the database, the response including a command; executing, by the computing device, the command; and sending, by the computing device, result data to the database, wherein the result data is data that results from executing the command on the computing device.

Abstract: Geographic location of a virtual server is predicted by determining a baseline geographic location of a virtual server of a computing environment, where the determining uses triangulation processing and known locations of multiple reference servers of the computing environment. Further, network communication-related data for communications between the multiple reference servers across a network is obtained, and a machine learning model is generated to predict an actual geographic location of the virtual server using, at least in part, the baseline geographic location of the virtual server and the obtained network communication-related data. The machine learning model is used to predict a current geographic location of the virtual server based, at least in part, on current network communication-related data for communications between the virtual server and one or more of the reference servers.

Abstract: Generating a data protection and recovery data backup option by identifying static and runtime metadata for a computing application, determining application criticality of the computing application according to the static metadata using a first machine learning model, determining a data backup option for the computing application according to application criticality and the runtime metadata, using a second machine learning model.

Abstract: Aspects of the present invention disclose a method for deploying an application in a computing environment receiving an application module, determining values of a first set of metadata for the received application module, determining a classification of the received application module based at least in part on the values of the first set of metadata, and determining whether the received application is deployable in at least an off-premise system based at least in part on the classification.

Abstract: One or more computer processors install a trigger on a primary database. The one or more computer processors, responsive to the trigger activating and a data modification associated with the primary database, format the data modification into a universal format for a plurality of backup databases. The one or more computer processors rotate an active backup database from the plurality of backup databases based on a determined alternating backup period. The one or more computer processors synchronize in real-time the primary database with the active backup database.

Abstract: A computer system synchronizes data in an edge computing network. A leader node is elected from a plurality of nodes, wherein the plurality of nodes includes a plurality of follower nodes that each cast a single vote for a candidate node, and wherein the candidate node is elected as the leader node in response to the candidate node receiving votes from a majority of the nodes. The leader node receives a request from a follower node comprising data to be replicated across the nodes, and transmits the data to the other nodes. When a majority of nodes receive the data, the leader node transmits instructions to the nodes to cause each node to commit the data to a data log maintained by each node. Embodiments of the present invention further include a method and program product for synchronizing data in substantially the same manner described above.

Abstract: Aspects of the present invention disclose a method for deploying an application in a computing environment receiving an application module, determining values of a first set of metadata for the received application module, determining a classification of the received application module based at least in part on the values of the first set of metadata, and determining whether the received application is deployable in at least an off-premise system based at least in part on the classification.

Abstract: Generating a data protection and recovery data backup option by identifying static and runtime metadata for a computing application, determining application criticality of the computing application according to the static metadata using a first machine learning model, determining a data backup option for the computing application according to application criticality and the runtime metadata, using a second machine learning model.

Abstract: One or more processors receive resource type and capability information and activity information of workloads of a domain. A first model is generated and trained to map the resource information to the activity information of domain workloads. The activity information is decomposed into a set of activity core elements (ACEs). The one or more processors generate a second model, wherein the second model is trained to predict a set of resource types and resource capabilities of the respective resource types, based on an input of the first set of ACEs decomposed from the activity information of the workloads of the domain. The one or more processors receive a second set of ACEs that are decomposed from activities associated with an unprecedented workload, and the one or more processors generate a predicted set of resources to perform the second set of ACEs.

Abstract: A question and answer pair is received from an external knowledge base. From the question, a set of intents is extracted. Whether the set of intents exceeds a match threshold with a subset of a plurality of intents within an internal knowledge base is determined. In response to determining a match threshold success, associating the question with the subset of intents within the plurality. A virtual assistant is trained to answer the question using the subset of intents.

Abstract: Intelligent management of stub files in hierarchical storage is provided by: in response to identifying a file to migrate from a file system to offline storage, providing metadata for the file to a machine learning engine; receiving a stub profile for the file from the machine learning engine that indicates an offset from a beginning of the file and a length from the offset for previewing the file; and migrating the portion of the file from the file system to an offline storage based on the stub profile. In some embodiments this further comprises: monitoring file system operations; in response to detecting a read operation of the portion of the file: determining a file type; providing file data to the machine learning engine; and performing a supervised learning operation based on the file type and the file data to update the machine learning engine.

Abstract: The present disclosure relates to a method of coordinating personal assistants. The method comprises: sending a request to a set of personal assistants based on a received query. Assistant responses may be received in response to the request from respective personal assistants of the set of personal assistants. Each assistant response of the assistant responses comprises a confidence score of the respective assistant response. The confidence scores may be adjusted to a common scale. The adjusted confidence scores may be weighted using domain-specific weights of the personal assistants in accordance with the domain of the query. One or more assistant responses of the received assistant responses may be selected based on the weighted scores and the selected assistant responses may be provided to the user.

Abstract: One or more computer processors install a trigger on a primary database. The one or more computer processors, responsive to the trigger activating and a data modification associated with the primary database, format the data modification into a universal format for a plurality of backup databases. The one or more computer processors rotate an active backup database from the plurality of backup databases based on a determined alternating backup period. The one or more computer processors synchronize in real-time the primary database with the active backup database.

Abstract: Embodiments for managing natural language user input are provided. Natural language input is received from a user utilizing a computing node. The natural language input is analyzed. The analyzing of the natural language input includes attempting to identify at least one of an intent associated with the natural language input, an entity associated with the natural language input, and context data associated with the natural language input. The natural language input is evaluated against a pre-trained model based on the analyzing of the natural language input. A quality score for the natural language input is calculated based on the evaluating of the natural language input. An action is caused to be performed utilizing the computing node based on the calculated quality score.

Abstract: The present disclosure relates to a method of coordinating personal assistants. The method comprises: sending a request to a set of personal assistants based on a received query. Assistant responses may be received in response to the request from respective personal assistants of the set of personal assistants. Each assistant response of the assistant responses comprises a confidence score of the respective assistant response. The confidence scores may be adjusted to a common scale. The adjusted confidence scores may be weighted using domain-specific weights of the personal assistants in accordance with the domain of the query. One or more assistant responses of the received assistant responses may be selected based on the weighted scores and the selected assistant responses may be provided to the user.

Abstract: User bioelectrical authentication is provided. A bioelectrical configuration signature of a user in physical contact with a hardware token is received, via a token reader, during a low range radio frequency user authentication process to permit a secure transaction by the user. The bioelectrical configuration signature of the user is utilized as part of the low range radio frequency user authentication process to authenticate the user to perform the secure transaction.