Abstract: Techniques for orchestrating a machine learning (ML) system on a distributed network. Determined performance levels for a ML system, determined from performance data received from the distributed network, are compared to performance requirements from the ML system. An orchestration module for the ML system then determines adjustments for the ML system that will improve the performance of the ML system and executes the adjustments for the ML system.

Abstract: In one embodiment, a device including a processor, and a memory to store data used by the processor, wherein the processor is operative to run a manufacturer usage description (MUD) controller operative to obtain a MUD profile of an Internet of Things (IoT) device from a MUD server, the MUD profile of the IoT device including: access rights of the IoT device, and any one or more of the following a default device username and/or a default device password of the IoT device, a recommended/required device password complexity of the IoT device, at least one service that should be enabled/disabled on the IoT device, and/or allowed security protocols and/or ciphers for communication to and/or from the IoT device, enforce security of the IoT device according to the MUD profile of the IoT device. Related apparatus and methods are also described.

Abstract: In one embodiment, a device including a processor, and a memory to store data used by the processor, wherein the processor is operative to run a manufacturer usage description (MUD) controller operative to obtain a MUD profile of an Internet of Things (IoT) device from a MUD server, the MUD profile of the IoT device including: access rights of the IoT device, and any one or more of the following a default device username and/or a default device password of the IoT device, a recommended/required device password complexity of the IoT device, at least one service that should be enabled/disabled on the IoT device, and/or allowed security protocols and/or ciphers for communication to and/or from the IoT device, enforce security of the IoT device according to the MUD profile of the IoT device. Related apparatus and methods are also described.

Abstract: Systems, methods, and computer-readable mediums for distributing machine learning model training to network edge devices, while centrally monitoring training of the models and controlling deployment of the models. A machine learning model architecture can be generated at a machine learning structure controller. The machine learning model architecture can be deployed to network edge devices in a network environment to instantiate and train a machine learning model at the network edge devices. Performance reports indicating performance of the machine learning model at the network edge devices can be received by the machine learning structure controller from the network edge devices.

Abstract: In one embodiment, a supervisory service for a wireless network obtains frequency-time Doppler profile information for an endpoint node attached to a first access point in the wireless network. The supervisory service uses the frequency-time Doppler profile information for the endpoint node as input to a machine learning model. The machine learning model is trained to output an action for the endpoint node with respect to the wireless network. The supervisory service causes the action for the endpoint node with respect to the wireless network to be performed.

Abstract: Systems and methods provide for Selective Tracking of Acknowledgments (STACKing) to improve buffer utilization and traffic shaping for one or more network devices. A network device can identify a first flow that corresponds to a predetermined traffic class and a predetermined congestion state. The device can determine a current window size and congestion threshold of the first flow. In response to a determination to selectively track a portion of acknowledgments of the first flow, the device can track, in main memory, information of a first portion of acknowledgments of the first flow. The device can exclude, from one or more buffers, a second portion of acknowledgments of the first flow. The device can re-generate and transmit segments corresponding to the second portion of acknowledgments at a target transmission rate based on traffic shaping policies for the predetermined traffic class and congestion state.

Abstract: Systems and methods provide for generating traffic class-specific congestion signatures and other machine learning models for improving network performance. In some embodiments, a network controller can receive historical traffic data captured by a plurality of network devices within a first period of time that the network devices apply one or more traffic shaping policies for a predetermined traffic class and a predetermined congestion state. The controller can generate training data sets including flows of the historical traffic data labeled as corresponding to the predetermined traffic class and predetermined congestion state. The controller can generate, based on the training data sets, traffic class-specific congestion signatures that receive input traffic data determined to correspond to the predetermined traffic class and output an indication whether the input traffic data corresponds to the predetermined congestion state.

Abstract: Techniques described herein relate to automatically generating standard network device configurations. In one example, one or more groups of network device configuration blocks may be obtained. An analysis of the one or more groups of network device configuration blocks may be performed, including identifying respective frequencies associated with respective network device configuration blocks of the one or more groups of network device configuration blocks. Based on the respective frequencies, one or more network device configuration blocks of the one or more groups of network device configuration blocks may be automatically aggregated into a standard network device configuration.

Abstract: In one example, a logical representation of a first graph is generated. The first graph indicates a configuration of a network device in a network at a first time. The first graph includes a first node representative of a first configuration block of the network device, a second node representative of a second configuration block of the network device, and a first link that indicates, by connecting the first node and the second node, that the first configuration block is associated with the second configuration block. The logical representation of the first graph is compared to a logical representation of a second graph that indicates an actual or planned configuration of the network device at a second time subsequent to the first time. In response, one or more changes in the configuration of the network device from the first time to the second time are identified.

Abstract: Techniques described herein relate to automatically generating standard network device configurations. In one example, one or more groups of network device configuration blocks may be obtained. An analysis of the one or more groups of network device configuration blocks may be performed, including identifying respective frequencies associated with respective network device configuration blocks of the one or more groups of network device configuration blocks. Based on the respective frequencies, one or more network device configuration blocks of the one or more groups of network device configuration blocks may be automatically aggregated into a standard network device configuration.

Abstract: A method includes obtaining performance characterization values from endpoints managed by a first fog node at a first hierarchical level in a hierarchy of fog nodes. The method includes changing a first operating characteristic of the wireless network based on the performance characterization values. The first operating characteristic affects the operation of one or more of the endpoints. The method includes transmitting a portion of the performance characterization values to a second fog node at a second hierarchical level in the hierarchy of fog nodes. The method includes changing a second operating characteristic of the wireless network based on an instruction from the second fog node. The second operating characteristic affects the operation of the first fog node and/or other fog nodes at the first hierarchical level. Changing one or more of the first operating characteristic and the second operating characteristic satisfies an operating threshold for the wireless network.

Abstract: In one example, a logical representation of a first graph is generated. The first graph indicates a configuration of a network device in a network at a first time. The first graph includes a first node representative of a first configuration block of the network device, a second node representative of a second configuration block of the network device, and a first link that indicates, by connecting the first node and the second node, that the first configuration block is associated with the second configuration block. The logical representation of the first graph is compared to a logical representation of a second graph that indicates an actual or planned configuration of the network device at a second time subsequent to the first time. In response, one or more changes in the configuration of the network device from the first time to the second time are identified.

Abstract: Techniques described herein relate to automatically generating standard network device configurations. In one example, one or more groups of network device configuration blocks may be obtained. An analysis of the one or more groups of network device configuration blocks may be performed, including identifying respective frequencies associated with respective network device configuration blocks of the one or more groups of network device configuration blocks. Based on the respective frequencies, one or more network device configuration blocks of the one or more groups of network device configuration blocks may be automatically aggregated into a standard network device configuration.

Abstract: A packet capture operation is configured via a first computing device. The packet capture operation is configured to capture packets provided by a second computing device. The first computing device obtains an indication that a user is within a predetermined location proximity to the second computing device. The packet capture operation is initiated in response to obtaining the indication at the first computing device.

Abstract: Presented herein are methodologies to on-board and monitor Internet of Things (IoT) devices on a network. The methodology includes receiving at a server, from a plurality of IoT devices communicating over a network, data representative of external environmental factors being experienced by individual ones of the plurality of IoT devices at a predetermined location; generating, using machine learning, an aggregated model of the external environmental factors at the predetermined location; receiving, at the server, a communication indicative that a new IoT device seeks to join the network at the predetermined location; receiving, from the new IoT device, data representative of external environmental factors being experienced by the new IoT device; determining whether there is a discrepancy between the external environmental factors of the new IoT device and the aggregated model; and when there is such a discrepancy, prohibiting the new IoT device from joining the network.

Abstract: In one embodiment, a device obtains content data provided by a social media platform to a user of the social media platform. The social media platform selects the content data for the user based on a behavioral model of the user. The device maintains an artificial intelligence-based model that models associations between the content data and interaction data indicative of interactions between the user and the social media platform. The device selects, using the artificial intelligence-based model, an obfuscation action to lower an accuracy of the behavioral model of the user, based on one or more configuration parameters set by the user. The device initiates performance of the obfuscation action.

Abstract: Embodiments herein describe association rules (e.g., affinity and anti-affinity rules) that a wireless device can use to optimize its performance in a Wi-Fi network. While BSS coloring is typically used to eliminate color collisions, the embodiments herein use BSS coloring to define what BSS colors should be on the same channel and which should not. For example, an affinity rule can indicate that a wireless device assigned a first BSS color (e.g., red) can share the same channel with wireless devices (or BSSs) assigned a second BSS color (e.g., green). In contrast, an anti-affinity rule can indicate that a wireless device in the red BSS color cannot share a channel with a wireless device assigned to a third BSS color (e.g., blue). The embodiments herein permit the wireless devices to be grouped with, or separated from, wireless devices having different BSS colors.

Abstract: In one embodiment, a supervisory service for a wireless network obtains frequency-time Doppler profile information for an endpoint node attached to a first access point in the wireless network. The supervisory service uses the frequency-time Doppler profile information for the endpoint node as input to a machine learning model. The machine learning model is trained to output an action for the endpoint node with respect to the wireless network. The supervisory service causes the action for the endpoint node with respect to the wireless network to be performed.

Abstract: Embodiments herein receive a request to reserve a fog computing resource for an end device, where the request includes a specified future time at which the fog computing resource will be used by the end device. It is determined that sufficient fog computing resources are available at the specified future time on a first fog node of a plurality of fog nodes. The fog computing resource of the first fog node is reserved for the specified future time, and an address corresponding to the first fog node is transmitted.

Abstract: In one embodiment, an apparatus includes a multi-socket motherboard, a processor connected to a first socket on the multi-socket motherboard, and an RDMA (Remote Direct Memory Access) interface module connected to a second socket on the multi-socket motherboard and in communication with the processor over a coherency interface. The RDMA interface module provides an inter-server interface between servers in an RDMA domain. A method for transferring data between servers with RDMA interface modules is also disclosed herein.