In one embodiment, a service divides one or more time series for a network key performance (KPI) into a plurality of time series chunks. The service clusters the plurality of time series chunks into a plurality of clusters. The service identifies a sketch that represents a particular one of the clusters. The service associates a label with the identified sketch. The service applies the label to a new KPI time series by matching the sketch to the new KPI time series.

Techniques described herein relate to a method for diagnosing and remediating service failures. The method includes identifying, by a diagnostic and remediation manager, a diagnostic event associated with a service of services; generating a dependency directed acyclic graph (DAG) associated with the service; generating health vectors associated with each node of the dependency DAG; updating the dependency DAG using the health vectors to generate an unhealthy subgraph; and remediating the service based on the unhealthy subgraph.

A method of abstracting network resources in a mobile communications network includes: determining a service coverage area for a class of service, the class of service defined by service parameters; determining a set of tracking areas that fall at least partly within the service coverage area; selecting available network resources for tracking areas of the set of tracking areas, for providing the class of service in the tracking areas; defining an abstraction view of the selected network resources for the class of service in the service coverage area, the abstraction view having deliverable values of the service parameters within the set of tracking areas; and outputting a communication signal having an indication of the abstraction view.

A method for task offloading based on power control and resource allocation in the Industrial Internet of Things includes establishing a computing model for computation tasks at different offloading locations, constructing communication power control, resource allocation and computation offloading problems as a mixed integer non-linear programming model, solving them using a deep reinforcement learning algorithm to obtain an optimal strategy for offloading of the computation tasks, thus achieving communication power optimization and cross-domain resource allocation.

A method for task offloading based on power control and resource allocation in the Industrial Internet of Things includes establishing a computing model for computation tasks at different offloading locations, constructing communication power control, resource allocation and computation offloading problems as a mixed integer non-linear programming model, solving them using a deep reinforcement learning algorithm to obtain an optimal strategy for offloading of the computation tasks, thus achieving communication power optimization and cross-domain resource allocation.

A method described herein involves various operations directed toward network security. The operations include accessing transaction data describing network traffic associated with a web server during an interval. Based on a count of new transactions involving an online entity during the interval according to the transaction data, a short-term trend is determined for the online entity. The operations further include applying exponential smoothing to a history of transactions of the online entity to compute a long-term trend for the online entity. Based on a comparison between the short-term trend and the long-term trend for the online entity, an anomaly is detected with respect to the online entity in the network traffic associated with the web server. Responsive to detecting the anomaly, an access control is implemented between the online entity and the web server.

A system and method to identify and prevent cybersecurity attacks on modern, highly-interconnected networks, to identify attacks before data loss occurs, using a combination of human level, device level, system level, and organizational level monitoring.

Computational methods and systems troubleshoot problems in a data center network. A dependency graph is constructed in response to an entity of the network exhibiting anomalous behavior. The dependency graph comprises nodes that correspond to metrics of entities that transmit data to and receive data from the entity over the network and edges that represent a connection between metrics. An anomaly score is determined for each metric of the dependency graph. Correlated metrics connected by the edges of the dependency graph are determined. Time-change events of the metrics of the dependency graph are also identified. Each metric of the dependency graph is rank ordered based on the anomaly scores, correlations with other metrics, and the time-change events. Higher ranked metrics are more likely associated with a problem in the network that corresponds to the anomalous behavior of the entity.

The present application describes a detect, alert and recovery system for various cloud-based and/or network-based services. The detect, alert and recovery system receives network performance data associated with a particular namespace from various network information sources. The network performance data may be aggregated based on various scopes. The aggregated data is then analyzed to determine whether an anomaly exists. If an anomaly exists, the detect, alert and recovery system may cause the performance of various actions in order to address the anomaly.

It is presented a method for determining packet loss in a fronthaul link of a radio access network. The method being performed in a packet loss determiner and comprising the steps of: obtaining a set of user equipments, UEs, that are all scheduled to communicate over a radio interface in a scheduling interval; creating a subset of UEs, comprising a number of UEs, from the set of UEs, that are the UEs in the set being most vulnerable to fronthaul packet loss; determining, for each UE in the subset of UEs, whether the communication in the scheduling interval was unsuccessful; and determining a packet loss in the fronthaul link depending on to what extent each one of the UEs in the subset of UEs is determined to have had unsuccessful use of the radio interface.