Abstract: In one embodiment, a device obtains temperature and utilization measurements for a set of network interface transceivers in a network. The device computes, for each of the transceivers, a correlation coefficient between its temperature and utilization measurements. The device applies a k-nearest neighbor classifier to the correlation coefficients, to predict a correlation coefficient. The device uses the predicted correlation coefficient to determine an impact of temperature on utilization of a particular network interface transceiver. The device initiates a mitigation action, when the determined impact of temperature on the utilization of the particular network interface transceiver exceeds a predefined threshold.

Abstract: This disclosure describes techniques and mechanisms for using a domain-specific language (DSL) to express and compile serverless network functions, and optimizing the deployment location for the serverless network functions on network devices. In some examples, the serverless network functions may be expressed entirely in the DSL (e.g., via a text-based editor, a graphics-based editor, etc.), where the DSL is a computer language specialized to a particular domain, such as a network function domain. In additional examples, the serverless network functions may be expressed and compiled using a DSL in combination with a general-purpose language (GSL). Once the serverless network function have been expressed and/or compiled, the techniques of this disclosure further include determining an optimized network component on which the serverless network function is to execute, and deploying the serverless function to the optimized network component.

Abstract: In one embodiment, a router includes one or more processors and one or more computer-readable non-transitory storage media coupled to the one or more processors. The one or more computer-readable non-transitory storage media include instructions that, when executed by the one or more processors, cause the router to perform operations including receiving software-defined networking in a wide area network (SD-WAN) policies from a component of an SD-WAN network. The operations also include establishing a session with a mobile device and receiving information associated with the mobile device in response to establishing the session with the mobile device. The operations further include filtering the SD-WAN policies based on the information associated with the mobile device to generate SD-WAN device-specific policies and communicating the SD-WAN device-specific policies to the mobile device.

Abstract: A system and method for providing network and port address translation is provided. A global IP address and a block (chunk) of ports are allocated for each mobile subscriber (MS) on first data connection. Subsequent data connections from the same MS are assigned the same IP address and a new port from this block. The mapping information is communicated, processed, and stored once for the complete block, instead of for every new data connection. This process reduces processing, communication, and storage requirements.

Abstract: In one embodiment, a technique comprises monitoring data transfer over a radio frequency (RF) link between a first device and a second device in a mesh network where the second device is a descendent node and the first device is a parent node. The technique further transfers the data over a power link communication (PLC) when the RF link is inactive. The method also includes broadcasting, by the second device, RF link availability to at least a third device in the mesh network when the RF link with the first device is inactive where the third device has an active link with the second device and the third device is a descendent node of the second device. The method then includes communicating, between the second device and the third device, through the active RF link.

Abstract: Coordinated radio fine time measurement is provided via sending, from a client device, a ranging request to a first radio; receiving a first response sent at a first time from the first radio over a first channel; receiving a second response sent at the first time from a second radio over a second channel; and calculating, based on times of flight for the first response and the second response, a location of the client device relative to the first radio and to the second radio. Coordinated radio fine time measurement is also proved via in response to receiving, at an Access Point (AP), a ranging request from a client device and determining to respond using multiple channels: sending, both at a first time, a first response from a first radio over a first channel a second response from a second radio over a different channel.

Abstract: In one embodiment, a device classification service receives data indicative of network traffic policies assigned to a plurality of device types. The device classification service associates measures of policy restrictiveness with the device types, based on the received data indicative of the network traffic policies assigned to the plurality of device types. The device classification service determines misclassification costs associated with a machine learning-based device type classifier of the service misclassifying an endpoint device of one of the plurality device types with another of the plurality of device types, based on their associated measures of policy restrictiveness. The device classification service adjusts the machine learning-based device type classifier to account for the determined misclassification costs.

Abstract: The disclosed technology relates a system is configured to generate a protected configuration for a network device based on network connectivity data for a plurality of devices in a managed network associated with a cloud management system. The system is further configured to receive a configuration change for the managed network, determine that the configuration change is incompatible with the protected configuration, and generate a notification that the configuration change is incompatible with the protected configuration.

Abstract: In one embodiment, a device health assessment service extracts device health status indicators from health assessment data that the service uses to determine a device health status of a networking device. The service forms, using the extracted set of device health status indicators, a health status signature for a particular device health status. The service trains a machine learning-based model to classify whether a networking device has the particular device health status, based in part on the health status signature. The service deploys the machine learning-based model to a target network for local device health assessment of one or more networking devices in the target network.

Abstract: In one embodiment, a device receives data regarding a wireless client attempting to form an association with a particular wireless access point in a network. The device queries a blockchain ledger in a blockchain database for wireless roaming data regarding the wireless client. The device uses, when available from the blockchain ledger, the roaming data regarding the wireless client to complete the association between the wireless client and the particular wireless access point. The device updates the blockchain ledger with data regarding the association between the wireless client and the particular wireless access point.

Abstract: In one embodiment, load balancing criteria and an indication of a plurality of network nodes is received. A plurality of forwarding entries are created based on the load balancing criteria and the indication of the plurality of nodes. A content addressable memory of a network element is programmed with the plurality of forwarding entries. The network element selectively load balances network traffic by applying the plurality of forwarding entries to the network traffic, wherein network traffic meeting the load balancing criteria is load balanced among the plurality of network nodes.

Abstract: In one embodiment, a method for FPGA accelerated serverless computing comprises receiving, from a user, a definition of a serverless computing task comprising one or more functions to be executed. A task scheduler performs an initial placement of the serverless computing task to a first host determined to be a first optimal host for executing the serverless computing task. The task scheduler determines a supplemental placement of a first function to a second host determined to be a second optimal host for accelerating execution of the first function, wherein the first function is not able to accelerated by one or more FPGAs in the first host. The serverless computing task is executed on the first host and the second host according to the initial placement and the supplemental placement.

Abstract: A data model can be customized by a user and executed in real-time at a network device. The user provides definitions for the customized data model based on a data model locally stored on the network device. The user provided definitions are used to generate a mapping contract which is processed by a mapping package generator to generate a mapping package. The mapping package can then be processed by a translation engine to dynamically execute a customized data model in real-time.

Abstract: The present technology is directed to a system and method for implementing passive power harvesting from ambient electromagnetic emissions with a smart rectenna that incorporates automatic frequency response tuning features. The disclosed system incorporates a tunable High Pass Filter and voltage multiplier rectifier with a front-end ultra wide band antenna unit. The frequency response of tunable components can be actively adjusted to match the frequency band containing most of the energy in the incident electromagnetic emission. A look up table is used for determining the appropriate biasing levels of the tunable components for each frequency in a frequency band of interest. By tuning a frequency response of impedance matching, filtering and rectifying components to correspond to a frequency region of maximum power spectral density in the incident energy signal, the system facilitates the scavenging of ambient electromagnetic energy from the spectral region with the highest power spectral density.

Abstract: A method is provided in one example embodiment and may include maintaining, by a Diameter Routing Agent (DRA), an availability status for a plurality of network elements; receiving a request associated with a user equipment (UE) session, wherein a first network element of the plurality of network elements is serving the UE session; determining that the first network element serving the UE session is unavailable; and re-establishing the UE session at a second network element of the plurality of network elements that is available, wherein the re-establishing is performed without terminating the UE session.

Abstract: A hybrid cooling server assembly can have a printed circuit board (PCB) with a processor socket disposed thereon and a hybrid cooling plate can be operably coupled with the processor socket. A radiator can having a working fluid received therein and be in fluidic communication with the radiator and the hybrid cooling plate by one or more tubular members. One or more cooling fans can be proximal to the radiator. The working fluid can be operable to receive heat from the cooling plate and reject heat at the radiator and the one or more cooling fans can be operable to produce an airflow across the hybrid cooling plate, thereby allowing the hybrid cooling plate transfer thermal energy to the airflow.

Abstract: A power management application running in a Chassis Management Controller reads utilization values of each server node dynamically in real time and assigns a respective priority to each server node based on its utilization value. The range of the utilization values is divided into terciles and the corresponding priorities assigned to the terciles are as HIGH, MEDIUM and LOW. The priorities are uses as guidelines for allocating power from a manageable power budget to each server node. A chassis power budget specified by an administrator includes the manageable power and unmanageable power used, for example, to power utilities, such as fans. Care is taken that a HIGH priority server node always receives its maximum power consumption rate, with the LOW priority server node receiving no less than its maximum power consumption rate. The MEDIUM priority server node receives at least the mean between its maximum and minimum power consumption rate.

Abstract: Systems, methods, and computer-readable media for updating configurations in sensors deployed in multi-layer virtualized environments. In some examples, a system can track information of sensors and collectors in the network. In response to determining that a specific collector becomes unavailable (e.g., the specific collector is down, offline or becomes unsupported), the system can determine affected sensors corresponding to the specific collector, determine a new collector among active collectors of the network for each of the affected sensors, and dynamically update configuration and settings of the affected sensors to maintain proper collector-to-sensor mappings and other settings on the affected sensors.

Abstract: Systems and methods for providing access prioritization to a private Long Term Evolution (LTE) network operating in a Citizens Broadband Radio Service (CBRS) spectrum include receiving, at a network device of the private Long Term Evolution (LTE) network, a functional group associated with a user equipment (UE) and a traffic load of a cell of the private LTE network associated with the UE. An access priority associated with the functional group and the traffic load is determined and provided to a CBRS access point (AP) which controls access to the cell. The access priority indicates to the CBRS AP a priority with which the UE is allowed to access the cell and can include a high priority, a low priority, or no access indication.

Abstract: Aspects of the technology provide solutions for determining a time period (“epoch”) required to monitor or analyze a tenant network. Some implementations of the technology include a process for making automatic epoch determinations, which includes steps for identifying one or more network parameters for a tenant network, analyzing the tenant network using the network parameters to discover one or more configuration settings of the tenant network, and determining a first epoch for the tenant network, the first epoch corresponding with a period of time to complete analysis of the tenant network using the network parameters. In some aspects, the process can further include steps for generating a tenant profile for the tenant network, the tenant profile based on the network parameters, the first epoch, and the one or more configuration settings of the tenant network. Systems and machine-readable media are also provided.