Abstract: A server may be provided to obtain a load balancing artificial intelligence (AI) model for a plurality of base stations in a communication system. The server may obtain teacher models based on traffic data sets collected from the base stations, respectively; perform a policy rehearsal process including obtaining student models based on knowledge distillation from the teacher models, obtaining an ensemble student model by ensembling the student models, and obtaining a policy model by interacting with the ensemble student mode; provide the policy model to each of the base stations for a policy evaluation of the policy model; and based on a training continue signal being received from at least one of the base stations as a result of the policy evaluation, update the ensemble student model and the policy model by performing the policy rehearsal process on the student models.

Abstract: The techniques disclosed herein enable systems to enhance network packet processing functionality through an accelerated packet classifier algorithm. Network packet classifiers serve to evaluate characteristics of network packets against various rules to determine how to process the network packet (e.g., routing, rewriting data). In contrast to existing systems, the present techniques pack a plurality of rules into a rule set which can then be applied to a network packet as a cohesive unit. By leveraging characteristics of binary arithmetic, the network packet classifier can evaluate the network packet against multiple rules in parallel. For example, the network packet classifier can concurrently compare values encoded by the network packet against a plurality of target ranges defined by the rules. This enables the network packet classifier to complete a full rule set evaluation within a single pass of the network packet thereby improving computing efficiency.

Abstract: A method used by an egress router is disclosed. The egress router obtains a capacity index of an application server attached to the egress router. The egress router further obtains a load index describing a load measurement between the egress router and the application server during a certain time period. The egress router encodes the capacity index and the load index into a packet. The egress router transmits the packet to one or more routers in an Internet protocol (IP) network.

Abstract: Message splitting and aggregation in a multi-stage electrical interconnection network are disclosed. A method of operating an electronic device comprised of computing devices, includes splitting, into segments, a message to be transmitted from a first of the computing devices, transmitting the segments to a second of the computing devices through a multi-channel that is based on an electrical connection between the first computing device and a plurality of switches, wherein the multi-channel includes channels respectively including electrical connections, the electrical connections connecting the first computing device with the second computing device, and reconstructing the message by aggregating the segments in the second computing device, wherein a bandwidth of the multi-channel transmitting the segments is greater than a maximum bandwidth of a single electrical connection of the electrical connections.

Abstract: A fast-path node of a packet processing service receives a packet of a forward-direction flow. The node obtains an exception-path routing rule from an exception-path routing rule source. The node sends a query to an exception-path cell of the service based on the routing rule, and receives a packet rewriting rule in response to the query. The rewriting rule is used to send a rewritten version of the packet to a destination.

Abstract: In a system with a plurality of servers serving one or more clients, situations may arise in which a client has a connection to a first server which is providing a service to the client, and in which it would be advantageous to transfer the connection to a second server, in a manner that does not require the re-establishing of the connection or the making of a renewed request, by the client, for the service. As such, a system and method for transferring a client connection from one server to another server are provided.

Abstract: A system includes a memory, an interface, and a processor. The memory stores a set of messages associated with a first user. The processor receives an indication of a conversation between the first user and at least a second user. In response, the processor selects a first message from the set of messages associated with the first user and uses the interface to present the first message to the second user as a suggestion to transmit to the first user. The processor determines that the second user transmitted the first message to the first user. In response to determining that the second user transmitted the first message to the first user, the processor updates the set of messages associated with the first user.

Abstract: A cloud tracking system may store configuration data separately for each account of multiple accounts that an entity has with a cloud computing provider and may aggregate the configuration data into a data structure that stores the configuration data together for the multiple accounts. The cloud tracking system may receive a request that includes a user identifier of a user associated with the entity and may search for the user identifier in the data structure. The cloud tracking system may identify a set of deployed cloud computing services associated with the user identifier based on searching for the user identifier in the data structure, wherein the set of deployed cloud computing services is associated with more than one account of the multiple accounts. The cloud tracking system may identify one or more deployed cloud computing services, of the set of deployed cloud computing services, that are outdated.

Abstract: Systems, methods, and related technologies for device classification are described. Methods include determining device information associated with a device coupled to a network, the device information including information obtained from one or more sources, classifying the device using the device information as input to a classifier, and applying a policy to the device based on the classification of the device.

Abstract: Systems and methods provide a new type of time-length-value data stream (TLV) which can be encapsulated in an OAM ping packet. This “EVI TLV” encapsulating OAM ping packet can be sent to a network device (e.g., a border router) that a liaison VRF is provisioned on. Upon receipt, the encapsulated EVI TLV instructs the network device to determine whether the configuration of the liaison VRF is mapped to an EVI value specified by the EVI TLV. If the configuration of the liaison VRF is mapped to the specified EVI value, the network device returns an echo response indicating that the configuration of the liaison VRF is mapped onto the specified EVI value. Such a response may indicate that the liaison VRF is reachable for the EVPN.

Abstract: A method includes: sending a pause frame followed by a migrate frame when a value of one of per-flow buffer usage (FBU) counters (referred to as a counter value) exceeds a pause threshold; stopping dequeuing one of pause egress queues (PEQs) when receiving a pause frame; enqueuing a packet into one of the PEQs when receiving a migrate frame; sending a resume frame followed by a migrate-back frame when a previous counter value exceeds the pause threshold and a current counter value is smaller than a resume threshold; when receiving a resume frame, resuming dequeuing one of the PEQs until empty, and then dequeuing a default egress queue (DEQ); and enqueuing a packet into the DEQ when receiving a migrate-back frame.

Abstract: Various approaches are provided for enabling Edge Dual Deployment (EDD). EDD includes Multi-access edge computing (MEC) Dual Deployments (MDD) initialization performed as a preliminary step before the actual registration, as a mechanism to allow the two platforms (EES and MEC platform) to communicate directly. MDD establishment is a dual registration mechanism of the edge application in the EDD environment. MDD update and closing are analogous registration update and de-registration processes, with both sub-cases.

Abstract: Systems, computer program products, and methods are described herein for large scale data extraction and processing. The method includes receiving a transmission request. The transmission request indicates two or more accounts for which one or more account statements are to be retrieved. The method also includes selecting a first retrieval set of the two or more accounts. A first retrieval set size of the first retrieval set is based on a first expected network capacity. The method further includes causing a retrieval of one or more account statements associated with first retrieval set. The method still further includes selecting a second retrieval set of the two or more accounts. A second retrieval set size of the second retrieval set is based on a second expected network capacity. The method includes causing a retrieval of one or more account statements associated with the second retrieval set.

Abstract: Examples described herein relate to one or more processors, when operational, to execute instructions stored in memory device, to cause performance of: execute a driver that is to: negotiate capabilities of hardware with a control plane for a virtualized execution environment and limit capabilities of the hardware available to the virtualized execution environment based on a service level agreement (SLA) associated with the virtualized execution environment. In some examples, the driver is to advertise hardware capabilities requested by the virtualized execution environment. In some examples, the control plane is to set capabilities of a hardware available to the virtualized execution environment based on the SLA.

Abstract: A push notification delivery system includes a server system including a processor, a network interface, and memory storing program instructions having code segments for receiving a received push notification, code segments for determining at least one of a favorable push time and a favorable message format based upon a database of received push information developed from a plurality of prior sent push notifications, and code segments for pushing the message to the destination in accordance with the at least one of a favorable push time and a favorable message format. A method for delivering push notifications includes receiving a received push notification including a message and a destination, sending a sent push notification derived from the received push notification to the destination in accordance with at least one favorable condition, receiving received push information related to the sent push notification, and storing the received push information in a database.

Abstract: A method of loading a web page that includes: receiving, by a client computing device over a communications network, hint information that identifies a first plurality of child objects associated with a web page; prefetching, by the client computing device over the communications network, one or more of the first plurality of child objects identified in the hint information; and receiving, by the client computing device over the communications network, the web page, wherein the web page identifies a second plurality of child objects. The method further includes: loading, by the client computing device, the web page, the loading including: loading one or more of the second plurality of child objects that are prefetched child objects of the first plurality of child objects. The metho further includes fetching, over the communications network, and loading one or more of the second plurality of child objects that are non-prefetched child objects.

Abstract: An embodiment establishes an interaction database based at least in part on interaction data representative of user engagement with a set of features of an application. The embodiment measures user engagement with each feature of the set of features. The embodiment calculates a familiarity score for each feature of the set of features based at least in part on measured user engagement. The embodiment constructs a subset of features of the set of features, wherein the subset of features is characterized by low familiarity score. The embodiment generates at least one recommendation to utilize at least one feature of the subset of features. The embodiment assesses a focal state of a user based on a focal state indicator metric. The embodiment displays the recommendation on a user device at an optimal time, wherein the optimal time is based at least in part on the focal state of the user.

Abstract: Devices, systems and processes for providing local alerts to users via non-local channels are described. A process may include receiving, by a content distributor system, an alert message. The alert message includes a first alert condition and a second alert condition. The first alert condition and the second alert condition are identified in the alert message, and an alert notice is generated, based on the identified alert conditions, and transmitted to a user device coupled to the content distributor system. The first alert condition specifies an alert type. The second alert condition specifies a time period. The alert message may include a third alert condition specifying an action for a user of the user device to take or refrain from taking during the time period and the alert notice may identify the action. The alert message may include a target condition and the alert notice identifies actions to be performed.

Abstract: Techniques and mechanisms are disclosed that enable collection of various types of data from cloud computing services and the generation of various dashboards and visualizations to view information about collections of cloud computing resources. A user can configure collection of data from one or more cloud computing services and view visualizations using an application platform referred to herein as a cloud computing management application. A cloud computing management application further may be configured to generate and cause display of interactive topology map representations of cloud computing resources based on the collected data, where an interactive topology map enables users to view an intuitive visualization of a collection of computing resources, efficiently cause performance of actions with respect to various resources displayed in the topology map, and analyze the collection of resources in ways that are not possible using conventional cloud computing service management consoles.

Abstract: Aspects of the subject disclosure may include, for example, determining a personalized cloud service requirement based on a user identity and proximity to a first location. A first edge network is configured based on the personalized cloud service requirement and the first location, and the user is provided with access to a personalized cloud service via the configured first edge network according to their proximity to the first location. A second, different location is determined, and a second edge network is configured based on the personalized cloud service requirement and according to the second location. The user is provided with access to the personalized cloud service via the configured second edge network according to their proximity to the second location. In this manner, the edge network may be reconfigured to effectively follow the user, thereby extending remote access to a consistent, personalized cloud service. Other embodiments are disclosed.