A method, including receiving, from multiple sources, respective sets of incidents, and respective suspiciousness labels for the incidents. A set of rules are applied so as to assign training labels to respective incidents in a subset of the incidents in the received sets. For each given incident in the subset, the respective training label is compared to the respective suspiciousness label so as to compute a respective quality score for each given source. Any sources having respective label quality scores meeting a predefined criterion are identified, and a model for computing predicted labels is fit to the incidents received from the identified sources and the respective suspiciousness labels of the incidents. The model is applied to an additional incident received from one of the sources to compute a predicted label for the additional incident, and a notification of the additional incident is prioritized in response to the predicted label.

Example methods and systems for logical network health check. One example may comprise obtaining network configuration information and network realization information associated with a logical network; processing the network configuration information and the network realization information to determine the following: (a) network configuration health information specifying a network configuration issue and a first remediation action; and (b) network realization health information specifying a network realization issue and a second remediation action; and providing, to a user device, multiple user interfaces (UIs) specifying the first health information and the second health information along with a visualization of the logical network. In response to detecting an instruction initiated by the user device using at least one of the multiple UIs, the first remediation action or the second remediation action may be performed.

The disclosed technology is directed towards automatically detecting failure states and the cause of the failure. For a network, the technology collects status messages from equipment and customers into batches as they occur. The technology groups and aggregates messages, then transforms the aggregations to the frequency domain. Anomalies induce detectable changes in the particle distribution of a trained particle filter, from which an anomalous spectrogram is generated. The status messages of each device are iteratively removed from the larger set of messages, resulting in reduced subsets that are each aggregated, transformed into a modified spectrogram and compared against the anomalous spectrogram to obtain a distance score. The distance score for each device is used to rank the devices with respect to being the cause of the failure.

Methods, systems, and computer-readable media are disclosed herein that monitor and improve network performance and reliability of a plurality of devices and nodes. In aspects, alert types are categorized based on the role, model, and operating system of a device or node within the network for which the alert was generated. A command set that is responsive to the alert and that is specially configured for the role, model, and operating system of the device or node is automatically selected to address the alert. The command set can be executed against the device or node (or neighboring device/node) in order to investigate the cause or source of the alert. Based on the results returned by the command set's execution, remediation actions can be selected and implemented to improve the technological performance (e.g., memory, CPU, connectivity) of the device or node in the network.

An agent interfaces with a sending computing device and a receiving computing device to automatically adjust one-way or two-way real-time audio and real-time video transmission parameters responsive to changing network conditions and/or application requirements. The agent incorporates a reinforcement learning model that adjusts transmission parameters to maximize an expected value of a sum of future rewards; the expected value of the sum of future rewards is based on a current state of the sending computing, a current action (e.g. a current set of transmission parameters) at the sending computing device and a reward provided by the receiving computing device. The reward is representative of a user-perceived quality of experience at the receiving computing device.

During operation, an electronic device receives, from a second electronic device in a network, a request for testing. In response, the electronic device set ups a video call with a video-call service. Then, the electronic device provides, to the second electronic device, an invitation for the video call. When the electronic device receives a notification (e.g., from the video-call service) that the video call has started, the electronic device provides content via the video-call service for the second electronic device. Next, the electronic device obtains communication-performance metrics associated with communication via the network during the video call and video-service performance metrics associated with the video call. Furthermore, the electronic device diagnoses a type of problem experienced at the second electronic device during the video call based at least in part on the communication-performance metrics, the video-service performance metrics and a pretrained machine-learning model.

During operation, an electronic device receives, from a second electronic device in a network, a request for testing. In response, the electronic device set ups a video call with a video-call service. Then, the electronic device provides, to the second electronic device, an invitation for the video call. When the electronic device receives a notification (e.g., from the video-call service) that the video call has started, the electronic device provides content via the video-call service for the second electronic device. Next, the electronic device obtains communication-performance metrics associated with communication via the network during the video call and video-service performance metrics associated with the video call. Furthermore, the electronic device diagnoses a type of problem experienced at the second electronic device during the video call based at least in part on the communication-performance metrics, the video-service performance metrics and a pretrained machine-learning model.

Systems and methods are provided for analyzing one or more root causes of service degradation events in a network or other environment. A method, according to one implementation, includes a step of monitoring a plurality of overlying services offered in an underlying infrastructure having a plurality of resources arranged with a specific topology. In response to detecting a negative impact on the overlying services during a predetermined time window and based on an understanding of the specific topology, the method further includes the step of identifying suspect components from the plurality of resources in the underlying infrastructure. The method also includes the step of obtaining status information with respect to the suspect components to determine a root cause of the negative impact on the overlying services.

Systems and methods are provided for analyzing one or more root causes of service degradation events in a network or other environment. A method, according to one implementation, includes a step of monitoring a plurality of overlying services offered in an underlying infrastructure having a plurality of resources arranged with a specific topology. In response to detecting a negative impact on the overlying services during a predetermined time window and based on an understanding of the specific topology, the method further includes the step of identifying suspect components from the plurality of resources in the underlying infrastructure. The method also includes the step of obtaining status information with respect to the suspect components to determine a root cause of the negative impact on the overlying services.

Techniques for enabling secure access to data using data blocks is described. Computing device(s) can provide instruction(s) to a component associated with an entity, wherein the instruction(s) are associated with an identifier corresponding to a data block of a plurality of data blocks. The computing device(s) can receive, from the component, data associated with the component, wherein the data is associated with the identifier and is indicative of a state of the component. The computing device(s) can store the data in the data block and monitor, using rule(s), changes to the state of the component based at least partly on the data in the data block. As a result, techniques described herein enable near real-time—and in some examples, automatic—reporting and/or remediation for correcting changes to the state of the component using data that is securely accessed by use of data blocks.