Abstract: In various embodiments, a device classification service obtains traffic telemetry data for a plurality of devices in a network. The service applies clustering to the traffic telemetry data, to form device clusters. The service generates a device classification rule based on a particular one of the device clusters. The service receives feedback from a user interface regarding the device classification rule. The service adjusts the device classification rule based on the received feedback.

Abstract: Chat-based interaction with an in-meeting virtual assistant may be provided. First, audio input associated with a meeting may be received. Next, an intent from the audio input may be detected. Text content associated with the audio input may then be generated in response to detecting the intent from the audio input. The text content may be displayed in a chat interface.

Abstract: In one embodiment, a device obtains routing forecasts for a software defined network. The device splits a particular routing policy for the software defined network into two or more routing policies, based on the routing forecasts. The device makes an evaluation as to whether the two or more routing policies should be reverted back into the particular routing policy. The device sends, to a user interface, data indicative of the particular routing policy that was split into the two or more routing policies and the evaluation as to whether the two or more routing policies should be reverted back into the particular routing policy.

Abstract: Service cognizant radio role assignments may be provided. A computing device may receive a beacon message associated with a tag. Then, based on information derived from the beacon message, an optimum radio in a network may be determined to monitor the tag. The optimum radio may be associated with an Access Point (AP) comprising one of a plurality of APs in the network. The optimum radio associated with the AP in the network may then be provisioned to monitor the tag.

Abstract: Adaptive guard interval calibration may be provided. A computing device may receive a first plurality of delay spread values. Each of the first plurality of delay spread values may respectively comprise an amount of time between when each of a respective first plurality Access Points (APs) receives a first tuning symbol from a first calibrating AP and when each of the respective first plurality APs receives a final multipath reflection of the first tuning symbol. Next, a first Guard Interval (GI) may be determined based on the first plurality of delay spread values. The first calibrating AP may then be provisioned with the first GI.

Abstract: In one embodiment, a device predicts a failure of a first tunnel in a software-defined wide area network (SD-WAN). The device determines that no backup tunnel for the first tunnel exists in the SD-WAN that can satisfy one or more service level agreements (SLAs) of traffic on the first tunnel, were the traffic rerouted from the first tunnel onto that tunnel. The device predicts, using a machine learning model, that a backup tunnel for the first tunnel exists in the SD-WAN that can satisfy an SLA of a subset of the traffic on the first tunnel, in response to determining that no backup tunnel exists in the SD-WAN that can satisfy the one or more SLAs of the traffic on the first tunnel. The device proactively reroutes the subset of the traffic on the first tunnel onto the backup tunnel, in advance of the predicted failure of the first tunnel.

Abstract: In one embodiment, a machine learning model evaluation system may define standardized, extensible class hierarchies for evaluating performance of a given machine learning model. The class hierarchies may include a plurality of target classes that formalize an expected output of the given machine learning model based on a given dataset, a plurality of output classes that formalize an actual output of the given machine learning model based on the given dataset, a plurality of metric classes that formalize a comparison of the expected output of the given machine learning model with the actual output of the given machine learning model, and a plurality of datasets. When a machine learning model is received for evaluation, the system may identify a target class, an output class, and a metric class that are applicable to the machine learning model. The system may also retrieve a dataset applicable to the machine learning model.

Abstract: Techniques for utilizing a Software-Defined-Networking (SDN) controller and/or a Data Center Network Manager (DCNM) and network border gateway switches associated with a multi-site cloud computing network to provide reachability data indicating physical links between the border gateways disposed in different sites of the multi-site network to establish secure connection tunnels utilizing the physical links and unique encryption keys. The SDN controller and/or DCNM may be configured to generate a physical underlay model representing the physical underlay, or network transport capabilities, and/or a logical overlay model representing a logical overlay, or overlay control-plane, of the multi-site network. The SDN controller may also generate an encryption key model representing the associations between the encryption keys and the physical links between the associated network border gateway switches.

Abstract: In one embodiment, a controller obtains data indicative of an application experience metric for an online application having application traffic conveyed via the network. The controller predicts the application experience metric that would result from a first edge router conveying its application traffic to the online application via a second edge router that is not currently connected to the first edge router via a tunnel, based on the obtained data. The controller makes a determination that the first edge router should route its application traffic to the online application via a tunnel between the first edge router and the second edge router, based on the predicted application experience metric. The controller causes a tunnel to be established in the network between the first edge router and the second edge router, whereby the first edge router routes its application traffic to the online application via the second edge router.

Abstract: In one embodiment, a method for classifying an encrypted flow includes receiving a plurality of packets associated with an encrypted flow traversing a network, collecting telemetry data from the flow without decrypting the flow, sending the telemetry data to a backend system for classification, using the telemetry data to classify the flow using a machine learning classifier, creating a classification response, and using the classification response to modify processing of the flow. In another embodiment, a method for classifying an encrypted flow includes receiving a plurality of packets associated with an encrypted flow traversing a network, collecting telemetry data from the first plurality of packets associated with the flow, sending the telemetry data to a backend system for classification, using the telemetry data to classify the flow using a machine learning classifier, and using the output of the classifier to modify processing of the flow.

Abstract: Failure prediction signaling and cognitive user migration may be provided. A client device may receive at least a portion of failure prediction data. The client device may then analyze the at least the portion of the failure prediction data. The client device may then roam from a first computing device to a second computing device in response to analyzing the at least the portion of the failure prediction data.

Abstract: An apparatus includes a connector for coupling a cable comprising at least one optical fiber and at least one electrical wire to an optical module at a network communications device, the connector comprising an electrical contact plate for engagement with an electrical contact on the optical module, and a ferrule for receiving the at least one optical fiber. The electrical contact plate is configured for electrically coupling the at least one electrical wire to the electrical contact on the optical module for delivery of power through the optical module.

Abstract: Techniques to facilitate the provision of server-management microservices for baseboard management controllers from a storehouse of such microservices are described herein. A technique described herein includes a server-management microservice storehouse obtaining a request for a server-management microservice for a baseboard management controller (BMC) of a server of a communications network. Further, the storehouse delivers the microservice to the BMC over the communication network. The server-management microservice is a modularized application that interacts with the operating system of the BMC while the microservice executes on the BMC, and the server-management microservice cause management of operations of the server or monitoring of the status of the server.

Abstract: This disclosure describes techniques for discovering network metrics relating to communications between a client device and a server device via a network. The client device may send one or more queries to a network device of the network. The network device may reply to the query of the client device with network metrics. The client device may forward the network metrics to the server device. For instance, an administrator at the server device may receive the network metrics. As such, the network metrics are discoverable by the client device and/or the administrator. Furthermore, the administrator may be able to use the discovered network metrics to monitor and/or troubleshoot a performance issue.

Abstract: Embodiments herein describe using a dual assess point (AP) to establish two access points that both are established by two individual radios (e.g., two 5 GHz radios). Generally, APs experience highly degraded performance when two co-located radios operate within the same band. In one embodiment, AP devices can deploy same band radios using a macro-micro cell approach. Thus, the AP may intelligently hand off client devices between the micro and macro cell in a way that optimizes the system for overall throughput and low packet latency while creating minimal oscillation of clients between cells. The embodiments in this disclosure disclose techniques that direct clients in a manner that optimizes these factors.

Abstract: Techniques for providing a non-blocking fabric in a network are described. A network controller determines the network requirement for various network traffic types on the network and determines the allocation of resources across the network needed to establish a midlay, including midlay components on the network. The network controller then establishes the midlay on the network according to the determined allocation. At least one of the midlay components is a virtually non-blocking fabric for high-priority traffic or fully non-blocking fabric for deterministic traffic.

Abstract: Aspects described herein include an optical apparatus comprising an input port configured to receive an optical signal comprising a plurality of wavelengths, and a plurality of output ports. Each output port is configured to output a respective wavelength of the plurality of wavelengths. The optical apparatus further comprises a first plurality of two-mode Bragg gratings in a cascaded arrangement. Each grating of the first plurality of two-mode Bragg gratings is configured to reflect a respective wavelength of the plurality of wavelengths toward a respective output port of the plurality of output ports, and transmit any remaining wavelengths of the plurality of wavelengths.

Abstract: Techniques for deploying, monitoring, and modifying network topologies operating across multi-domain environments using formal models and weighting factors assigned to computing elements in the network topologies. The weighting factors restrict or allow the movement of various computing elements and/or element groupings to prevent undesirable disruptions or outages in the network topologies. Generally, the weighting factors may be determined based on an amount of disruption experienced in the network topologies if the corresponding computing element or grouping was migrated. As the amount of disruption caused by modifying a particular computing element increases, the weighting factor represents a greater measure of resistivity for migrating the computing element. In this way, topology deployment systems may allow, or disallow, the modification of particular computing elements based on weighting factors.

Abstract: In one embodiment, a service receives signal data indicative of phases and gains associated with wireless signals received by one or more antennas located in a particular area. The service determines, from the received signal data, changes in the phases and gains associated with the wireless signals. The service estimates a direction of motion of one or more objects located in the particular area, based on the determined changes in the gains associated with the wireless signals. The service estimates a total mass of the one or more objects located in the particular area based on a ratio of the determined changes in the gains associated with the wireless signals over the determined changes in the phases associated with the wireless signals.

Abstract: A method includes receiving, at an access point, a buffer status report from a device wirelessly connected to the access point. The buffer status report indicates a queue depth of the device and a head of line delay for the device. The method further includes scheduling uplink resources at the access point for the device based on the queue depth and the head of line delay.