Abstract: Aspects of the present invention disclose a method for dynamically adjusting a clock speed of a core of a multicore processor of user equipment. The method includes one or more processors generating a dedicated logical channel from a user device to a multidomain service orchestration layer of a fifth generation (5G) telecom network. The method further includes collecting information of the user device through the dedicated logical channel. The method further includes identifying situational insight of the user device based at least in part on the information of the user device. The method further includes identifying determining a workload forecast of one or more cores of the user device based at least in part on the situational insight. The method further includes identifying determining a recommended central processing unit (CPU) clock speed for a CPU core of the user device based at least in part on the workload forecast.

Abstract: In an approach for in-band selective data snapshot using fifth generation (5G) radio link control (RLC) channels for edged cloud application backups, a processor creates one or more RLC channels to transfer network data packets in a telecommunication network using a network slicing technology. A processor allocates the one or more RLC channels to one or more applications in a user device for dedicated packet routing requirements. A processor maps the corresponding RLC channels and applications with corresponding identifiers, respectively. A processor monitors the one or more applications for backups. A processor, in response to receiving an application's backup request, creates a copy of a datastore associated with the application based on changes monitored in the one or more applications.

Abstract: A computer-implemented method includes: receiving, by a computing device, a query from a user device; generating, by the computing device and using a machine learning model, an answer to the query; determining, by the computing device, an urgency level of the query; generating, by the computing device, an explanation of the answer based on the determined urgency level; and presenting, by the computing device, the answer and the explanation to a user via the user device.

Abstract: Emergency monitoring application mobility management is provided. A portion of battery power on a mobile device is reserved for an emergency monitoring application between defined boundaries of a certain area based on analysis of a profile identifying battery consumption of each respective application running on the mobile device. A dedicated logical communication channel is established with the mobile device that the emergency monitoring application connects to in a detected emergency situation between the defined boundaries of the certain area. The dedicated logical communication channel is utilized to shut down a number of applications running on the mobile device during the detected emergency situation so that the reserved portion of battery power is retained to run the emergency monitoring application on the mobile device between the defined boundaries of the certain area during the detected emergency situation.

Abstract: An approach is provided in which the approach stores, in a cached recycle bin, a set of sensor data that is sent from a sensor at a transmit frequency. The approach samples at least a portion of the set of sensor data from the cached recycle bin based on a sampling frequency and training a machine learning model using the sampled data. In response to detecting that a performance of the machine learning model falls below a threshold during the training, the approach adjusts the sampling frequency and re-sampling at least a portion of the sensor data based on the adjusted sampling frequency. The approach instructs the sensor to adjust the transmit frequency in response to determining that the performance of the machine learning model reaches the threshold using the re-sampled data.

Abstract: An embodiment generates a candidate retraining dataset in response to a retraining request. The embodiment generates first metadata comprising attribute information for data points in the candidate retraining dataset and second metadata comprising a range of attributes from training data previously used to train the model. The embodiment determines whether the candidate retraining dataset includes new information for the machine learning model based on a comparison of the first metadata to the second metadata. In a first case in which the candidate retraining dataset is determined to include new information, the embodiment initiates retraining of the machine learning model using the candidate retraining dataset. In a second case in which the candidate retraining dataset is determined to not include new information, the embodiment generates a response to the retraining request indicating that retraining the machine learning model using the candidate retraining dataset is not recommended.

Abstract: Provided is a system and method which prioritizes data replication packets between a private cloud and a public cloud which provides a backup for the private cloud. In one example, the method may include receiving a request from a software application to write data to a storage location of a private cloud that hosts the software application, identifying storage attributes of the storage location of the private cloud, generating a replication request for replicating the data over a network to a public cloud, embedding a priority tag into the replication request based on the identified storage attributes of the storage location of the private cloud, and transmitting the tagged replication request over the network from the private cloud to the public cloud based on a bandwidth assigned to the embedded priority tag.

Abstract: A processor may receive a media frame collection. The media frame collection may include two or more frames. The processor may identify, from the media frame collection, a location of an entity. The processor may determine whether the location of the entity is within a threshold location in regard to the other two or more frames in the media frame collection. The processor may display the one or more frames that exceed the threshold location to a user.

Abstract: A method, computer program product, and computer system for using agents at a data lake to assist training, testing, and deploying a model for predicting one or more outputs from specified input to the model, using agents at the data lake. The agents identify at the data lake: hardware resources, a first portion of model input data for training and testing the model, a training language, and a first software resource. A first execution of the first software resource is triggered to split the first portion of the model input data into training input data and testing input data. A first portion of the hardware resources is identified for training and testing the model. A second software resource stored at the data lake is identified for optimizing the model in accordance with an optimization algorithm. Training the model at the data lake is triggered.

Abstract: A method includes collecting, by one or more processors of a computer system, digestive capabilities for Input Output Queues (IOQs) of infrastructure components in a hybrid cloud infrastructure, and allocating, by the one or more processors of the computer system, nonvolatile memory express (NVMe) storage cloud resources for the hybrid cloud infrastructure based on the collected digestive capabilities for IOQs of the infrastructure components.

Abstract: A first version of a model specified by a model execution request is executed, producing a first execution result. A second version of the model is selected according to an input data attribute specified by the model execution request. The second version of the model is executed, producing a first execution result. Using a natural language processing engine, responsive to the first execution result and the second execution result differing by more than a threshold amount, a natural language explanation of a difference between the first execution result and the second execution result is constructed.

Abstract: A system, method, and computer program product for implementing dynamic telecom network agent filtering is provided. The method includes retrieving peripheral data associated with user equipment (UE) enabled with respect to a telecom network associated with a telecom network hardware device. The peripheral data is mapped with a pre-defined filter selection model and a filtering agent model is selected. Likewise, filtering agents associated with the UE are selected and environmental characteristics associated with the telecom network are retrieved. A location of the UE is tagged and the filtering agents are pushed to the UE. Network commands associated with execution of the filtering agents are generated and the filtering agents are executed with respect to the UE. In response, noise and distortion occurring during operation of the UE with respect to the telecom network is reduced.

Abstract: A system, method, and computer program product for implementing data replication generation is provided. The method includes utilizing hardware and software resources within a hybrid cloud environment. A non-volatile memory host system and an associated target system are enabled for operational functionality and the non-volatile memory host system is connected to an I/O queueing component. In response, a plurality of queue structures is generated with respect to a host driver component and a connection between the non-volatile memory host system and the associated target system is detected. In response, a special purpose cache structure is generated and the plurality of queue structures and the special purpose cache structure are enabled such that remote data mirroring functionality is enabled.

Abstract: A method, computer program product, and a system where a processor(s) ingests content from a source(s) with an attribute(s) comprising a verified temporal context(s) of the source(s). The processor(s) cognitively analyzes the content, by applying an entity recognition algorithm(s) to identify and extract entities in the source(s). The processor(s) classifies each extracted entity into a given grouping from a plurality of groupings based on at least one attribute comprising the verified temporal context of the source from which the extracted entity was extracted. The processor(s) generates a corpus comprising the groupings; each grouping comprises extracted entities with verified temporal contexts consistent with a defined time period.

Abstract: A computer-implemented method, system and computer program product for managing queries to non-relational databases. A query to retrieve an object from a storage system of a non-relational database system is received from an application. A disk volume of the storage system is then identified from a record of the requested object. Upon identifying the disk volume, the logical and/or physical input/output connections to the identified disk volume that are tagged to dissimilar CPU cores of the storage system are identified. One of the identified logical and/or physical input/output connections is then selected based on the input/output access characteristics of the identified logical and/or physical input/output connections, the CPU core speeds for the CPU cores and a determined level of urgency for the retrieval of the object. The request is then sent to the storage system over the selected input/output connection.

Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: obtaining service request data by a service application; generating query data for query of one or more machine learning model in dependence on the service request data; examining model data of a plurality of candidate machine learning models; selecting at least one model from the candidate machine learning models in dependence on the examining model data of the plurality of candidate machine learning models, wherein the at least one model defines a selected at least one model; and sending the query data to the selected at least one model for return of responsive prediction data.

Abstract: A computer implemented method for forecasting sensor states in a sensor network is provided. A number of processor units identifies weights and changes in the weights over time in a temporal graph of the sensor network using a temporal graph neural network trained using the temporal graph. The number of processor units determines patterns of the weights based on the weights and the changes in the weights over time. The number of processor units trains a regression machine learning model using the patterns of the weights to forecast sensor states for sensors in the sensor network.

Abstract: A computer-implemented method for altering an appearance of an electrochromic coating of a vehicle is provided. The method includes obtaining vehicle information related to at least one vehicle attribute. The method also includes applying weightage factors to the at least one vehicle attribute. The method also includes determining that a vehicle is a boundary area of a testing zone based on a determined location of the vehicle. The method also includes changing a base appearance of the electrochromic coating of the vehicle to an altered appearance based on the determination that the vehicle is in the boundary area and based on the weightage factors of the vehicle components.

Abstract: A method for power-savings of UE operating within a 5G network. Transmission requirement data of logical channels for application-level data are collected from SDAP, including logical channel ID, application ID, and QCI indexing information. A list of logical channels with QCI indexing for permissible delay of packets is created. The logical channels of the list are mapped to related RLC channels. Responsive to receipt of a packet by a RLC multiplexing layer, the RLC channel ID from a SARQ packet is mapped to the list of permissible transmission delay logical channels of packets. responsive to a match between the RLC channel ID and the logical channels of the list, the packet is saved to an RLC data structure in allocated memory, and responsive to receipt of a time-critical packet, the time-critical packet, and the saved packet are submitted to a MAC carrier controller for transmission processing.

Abstract: Imposing and/or readjusting the CPU clocks for services of containers from various cloud-based cognitive systems in a multi-cloud cognitive computing environment for performing a particular job. The particular job having an emergency performance requirement. During processing of the job, a need to tune the execution of instructions is identified through a pre-existing emergency identification process that includes collecting clock data for each container supporting the performance of the particular job.