Techniques for dynamic content delivery network (CDN) selection using the domain name service (DNS) protocol are described. A DNS resolver utilizes a network identifier provided within a DNS query seeking to resolve a domain to select between different CDNs. The selection can be based on an analysis of network metric summary data corresponding to the CDNs from the perspective of an approximate location of the requesting client, as determined via the network identifier as a proxy. The selection process and involved network metric types can be configured by the user associated with the domain via a selection policy. Network metrics can be provided by the user or collected based on reported data generated by remote clients through provided metric-generating code, and thereafter transformed into network metric summary data that is used for resolution.

A method includes causing, by a processor, a graphical user interface (GUI) to be output by a display. The GUI includes a first user input field identifying a target key performance indicator (KPI) associated with a network. A second user input field identifying a KPI peak-usage frequency. A third user input field identifying a peak-usage relationship. The method also includes creating a first monitoring profile based on the target KPI for the KPI peak-usage frequency for the peak-usage relationship. Storing the first monitoring profile. Monitoring the target KPI for the KPI peak-usage frequency based on the first monitoring profile. Collecting target KPI data over a period defined by the KPI peak-usage frequency. Determining a peak-usage during the period defined by the KPI peak-usage frequency based on the collected target KPI data. Periodically reporting peak-usage data based on the first monitoring profile.

A method includes causing, by a processor, a graphical user interface (GUI) to be output by a display. The GUI includes a first user input field identifying a target key performance indicator (KPI) associated with a network. A second user input field identifying a KPI peak-usage frequency. A third user input field identifying a peak-usage relationship. The method also includes creating a first monitoring profile based on the target KPI for the KPI peak-usage frequency for the peak-usage relationship. Storing the first monitoring profile. Monitoring the target KPI for the KPI peak-usage frequency based on the first monitoring profile. Collecting target KPI data over a period defined by the KPI peak-usage frequency. Determining a peak-usage during the period defined by the KPI peak-usage frequency based on the collected target KPI data. Periodically reporting peak-usage data based on the first monitoring profile.

Automatically detecting whether sessions are routed through proxy servers is provided. The system identifies a log with session information generated by a device for a session established between a client and a server traversing the device. The system compares a source internet protocol (“IP”) address for the session identified from the log with IP addresses of proxy servers. The system updates, responsive to a match based on the comparison, the log with an indication that the session was routed through a proxy server.

Techniques are described for monitoring application performance in a computer network. For example, a network management system (NMS) includes a memory storing path data received from a plurality of network devices, the path data reported by each network device of the plurality of network devices for one or more logical paths of a physical interface from the given network device over a wide area network (WAN). Additionally, the NMS may include processing circuitry in communication with the memory and configured to: determine, based on the path data, one or more application health assessments for one or more applications, wherein the one or more application health assessments are associated with one or more application time periods for a site, and in response to determining at least one failure state, output a notification including identification of a root cause of the at least one failure state.

Aspects described herein relate to a centralized computing system that interacts with a plurality of data centers, each having an edge server. Each edge server obtains sensor information from a plurality of sensors and processes the sensor information to detect an imminent shutdown and sends emergency data to a centralized processing entity when detected. In order to make a decision, the edge server processes the sensor data based on dynamic sensor thresholds and dynamic prioritizer data by syncing with the centralized computing system. Because of the short time duration to report emergency data before an imminent complete shutdown, an edge server may utilize a quantum data pipeline and quantum data storage as a key medium for all data transfer in a normal condition and at the time of emergency for internally transporting processed sensor data and providing the emergency data to the centralized processing entity.

Aspects described herein relate to a centralized computing system that interacts with a plurality of data centers, each having an edge server. Each edge server obtains sensor information from a plurality of sensors and processes the sensor information to detect an imminent shutdown and sends emergency data to a centralized processing entity when detected. In order to make a decision, the edge server processes the sensor data based on dynamic sensor thresholds and dynamic prioritizer data by syncing with the centralized computing system. Because of the short time duration to report emergency data before an imminent complete shutdown, an edge server may utilize a quantum data pipeline and quantum data storage as a key medium for all data transfer in a normal condition and at the time of emergency for internally transporting processed sensor data and providing the emergency data to the centralized processing entity.

A machine learning method executed by a computer, the method includes distributing a first learning model learned on the basis of a plurality of logs collected from a plurality of electronic devices to each of the plurality of electronic devices, the first learning model outputting operation content for operating an electronic device; when an operation different from an output result of the first learning model is performed by a user relative to a first electronic device among the plurality of electronic devices, estimating a similar log corresponding to a state of the learning model in which the different operation is performed from the plurality of logs; generating a second learning model on the basis of a log obtained by excluding a log of a second electronic device associated with the similar log from among the plurality of logs; and distributing the second learning model to the first electronic device.

A machine learning method executed by a computer, the method includes distributing a first learning model learned on the basis of a plurality of logs collected from a plurality of electronic devices to each of the plurality of electronic devices, the first learning model outputting operation content for operating an electronic device; when an operation different from an output result of the first learning model is performed by a user relative to a first electronic device among the plurality of electronic devices, estimating a similar log corresponding to a state of the learning model in which the different operation is performed from the plurality of logs; generating a second learning model on the basis of a log obtained by excluding a log of a second electronic device associated with the similar log from among the plurality of logs; and distributing the second learning model to the first electronic device.

Examples described herein include systems and methods for providing dynamic serviceability for a software-defined data center (“SDDC”). An example method can include collecting data-center metrics from a management service that monitors the SDDC, filtering the data-center information based on a predetermined list of metrics provided by a partner entity, and translating the filtered data-center information into a partner-specific format requested by the partner entity. The example method can also include generating metadata associated with the translated data-center information and transmitting the metadata and translated data-center information to a partner site associated with the partner entity. If the partner site is not available, the method can include transmitting the information to a partner-accessible storage location and, when the partner site becomes available, identifying the storage location and failed attempt to deliver the information.