Abstract: A system includes a memory; and a processor configured to train a first machine learning model based on the first dataset labeling; provide the second dataset to the trained first machine learning model to generate an updated second dataset including an updated second dataset labeling, determine a first difference between the updated second dataset labeling and the second dataset labeling; train a second machine learning model based on the updated second dataset labeling if the first difference is greater than a first threshold value; provide the first dataset to the trained second machine learning model to generate an updated first dataset including an updated first dataset labeling, determine a second difference between the updated first dataset labeling and the first dataset labeling; and train the first machine learning model based on the updated first dataset labeling if the second difference is greater than a second threshold value.

Abstract: A system and method of serverless function scaling is described. The method may include receiving one or more types of serverless function invocations. The method also includes invoking one or more serverless functions and receiving metrics from the one or more serverless functions. The method also includes generating scaling heuristics for scaling of a number of serverless functions in view of the received metrics.

Abstract: An example method for discovering and grouping application endpoints in a network environment is provided and includes discovering endpoints communicating in a network environment, calculating affinity between the discovered endpoints, and grouping the endpoints into separate endpoint groups (EPGs) according to the calculated affinity, each EPG comprising a logical grouping of similar endpoints for applying common forwarding and policy logic according to logical application boundaries. In specific embodiments, the affinity includes a weighted average of network affinity, compute affinity and user specified affinity.

Abstract: System, methods, and devices provide prioritized scaling for persistent content sharing platforms. Systems include a processing device comprising one or more processors configured to generate a first display configured to represent a private computing environment associated with a user, and configured to generate a second display configured to represent a shared computing environment associated with the user. Systems further include an encoder configured to generate a packetized data stream representing the second display. Systems also include a multiview module comprising a decoder configured to generate a third display configured to represent a unified view of a plurality of active shared computing environments associated with a plurality of users, the decoder being further configured to decode at least one packetized data stream representing at least one of the plurality of active shared computing environments.

Abstract: An example method for discovering and grouping application endpoints in a network environment is provided and includes discovering endpoints communicating in a network environment, calculating affinity between the discovered endpoints, and grouping the endpoints into separate endpoint groups (EPGs) according to the calculated affinity, each EPG comprising a logical grouping of similar endpoints for applying common forwarding and policy logic according to logical application boundaries. In specific embodiments, the affinity includes a weighted average of network affinity, compute affinity and user specified affinity.

Abstract: Methods, computer readable media, devices, and systems provide for distributed machine learning. In one aspect, a method of training a model is disclosed. The method includes receiving, by a client device, from one or more servers, an intermediate model, training, by the client device, the intermediate model based on private data, and transmitting, by the client device, to the one or more servers, the trained intermediate model.

Abstract: A first server can generate user profiles and receive requests from user devices for enrollment in a first server-managed system that includes user groups. The first server can provide a unique key to a user device during an enrolment process based on a user group the user device is assigned to. The first server can include an enrollment notification for the user device in a first notification transmitted to a messaging service. The messaging service can transmit a second notification to the user device, and the user device can request a user profile from a second server based on second server access information included in the second notification. The second server can use the unique key to access user profile information which it transmits to the user device based on the request. The user device can access the user profile from the profile information using the unique key.

Abstract: A communication system is a communication system connected to an upper network and a lower network, and includes at least one intermediate communication device including a plurality of upper ports for inputting and outputting optical signals to and from the upper network, and a plurality of lower ports for inputting and outputting optical signals to and from the lower network, the intermediate communication device being configured to relay communication between the upper network and the lower network, a switching device that is connected to the upper and lower networks and the plurality of upper and lower ports, switches a connection between the upper network and the plurality of upper ports, and switches a connection between the lower network and a plurality of lower ports, and a control device that outputs, to the switching device, an instruction to switch the connection by the switching device according to a communication status between the upper network and the lower network.

Abstract: A communication network may be provided having a plurality of network nodes and which may be configured to enable instantiation of different network slices which represent virtual networks with different feature sets, e.g., providing different network functions and/or having different network characteristics. The communication network may be configured to allow a network slice to be used for data communication between i) a user equipment (UE) connected to the network and configured to execute an application, and ii) a content or application server (CAS) configured to provide an application service via the network. Network functions may be provided which allow influencing the UE's slice usage based on requirements of the application service, even if the CAS is located outside the connectivity provider's network, e.g., outside a mobile operator's domain.

Abstract: Methods, systems, and apparatuses are described for dynamic management of file system objects (e.g., a file or a directory). Such management includes syncing, routing, and storing capabilities. A file system object may be tagged with one or more tags based on at least one of file system object content or file system object location. Dynamic rules that control routing of file system objects to one or more locations based on the one or more tags are accessed and searched for an applicable rule. The file system object may be routed and/or stored based on the applicable rule. The rules may specify allowable locations as well as locations that are not allowed for a given file system object. The location may be a cloud-based location, a location that is on the premises of the enterprise, a location provided by and/or serviced by a competing entity, etc.

Abstract: In general, techniques are described for slice assurance within a mobile network. In some examples, a method includes obtaining, by a slice assurance function (SAF) executed by a device, key performance indicator (KPI) values for a first slice of a plurality of slices implemented by a plurality of base stations serving a tracking area of a mobile network; determining, by the SAF, based in part on the KPI values for the first slice, a service level agreement (SLA) for the first slice has not been met; re-allocating, by the SAF in response to the determining, slice resources associated with any of the plurality of slices to compute a new slice configuration parameter for the first slice; and reconfiguring, by the SAF, at least one of the plurality of base stations to implement the new slice configuration parameter for the first slice.

Abstract: Disclosed are a method for realizing an nGraph framework supporting an FPGA backend device, and a related apparatus. The method includes: integrating an OpenCL standard API library into an nGraph framework; creating, in the nGraph framework, an FPGA backend device creation module for registering an FPGA rear-end device, initializing an OpenCL environment and acquiring the FPGA backend device; creating, in the nGraph framework, an FPGA buffer space processing module for opening up an FPGA buffer space and for reading and writing an FPGA cache; creating, in the nGraph framework, an OP kernel implementation module for creating an OP kernel and compiling the OP kernel; and creating, in the nGraph framework, an FPGA compiling execution module for registering, scheduling and executing the OP kernel.

Abstract: A communication device includes a parameter setter and a communication controller for cyclic communication to transmit data including a first data piece and a second data piece different from the first data piece and having lower priority for real-time transmission than the first data piece to slave stations. The parameter setter presets a number of divisions n to divide the second data piece before the cyclic communication is started. The communication controller transmits the first data piece to a first slave station in one cycle, divides the second data piece by the number of divisions n, and transmits the divided second data piece to a second slave station in n cycles, where n is the number of divisions. The communication controller transmits, before transmitting the second data piece, data indicating the number of divisions n set by the parameter setter to the second slave station.

Abstract: A switching fabric role assignment system includes a plurality of switch devices coupled together in a switching fabric. A first switch device included in the plurality of switch devices receives an endhost device identification communication from an endhost device when the endhost device is connected to the first switch device. The first switch device uses the endhost device identification communication to identify an endhost device type of the endhost device and determines, based on the endhost device type, a first switch device role for the first switch device. The first switch device then transmits a first switch device role communication that identifies the first switch device role to a second switch device included in the plurality of switch devices and connected to the first switch device. The second switch device may then determine, based on the first switch device role, a second switch device role for the second switch device.

Abstract: The disclosure generally describes methods, software, and systems for providing solution descriptions. A problem description of a problem is received, from a client, at a cloud-based reasoning service. A solution description for a solution to the problem is received. Case metadata for a case defining the problem and solution are generated by the cloud-based reasoning service. The case metadata, including the problem description and solution description, are stored by the cloud-based reasoning service in a cases repository associating solutions with problems. A new problem is received at the cloud-based reasoning service. An automated analysis of the new problem is performed, and a comparison is made of the new problem with existing solutions in the cases repository to identify solutions matching the new problem. A new solution description is provided that is based on a match between the new problem description and the problem description and using the problem solution.

Abstract: A system, method, and machine-readable storage medium for providing a recommendation to a client to modify minimum IOPS settings are provided. In some embodiments, a client may assign a minimum input/output operations per second (IOPS) setting to each volume of a plurality of volumes. The plurality of volumes may reside in a common cluster and include a first number of volumes. A set of volumes of the plurality of volumes may be determined, where each volume of the set of volumes has a relative disparity that satisfies a relative threshold. The set of volumes includes a second number of volumes. A minimum IOPS recommendation may be transmitted to the client to modify (e.g., increase or decrease) the minimum IOPS settings of the set of volumes in response to determining that a ratio of the second number of volumes to the first number of volumes is greater than a performance threshold.

Abstract: Some embodiments provide a method for generating a multi-layer network map from network configuration data. The method receives network configuration data that defines network components and connections between the network components for a network that spans one or more datacenters. Based on the received network configuration data, the method generates multiple data layers for a multi-layer interactive map of the network. Different data layers include different network components and connections. The method generates a visual representation of the network for each data layer. Each visual representation includes a map of the network at a different level of hierarchy.

Abstract: Disclosed embodiments provide systems and methods related to updating creatives generation models. The system may include at least one memory unit for storing instructions and at least one processor configured to execute the instructions to perform operations.

Abstract: Systems and methods provide techniques for more effective and efficient predictive monitoring of a software application framework. For example, various embodiments of the present invention provide methods, apparatuses, systems, computing devices, and/or the like that are configured to enable effective and efficient predictive monitoring of a software application framework using alert signatures for the software application that are generated by using at least one of first-order alert classification machine learning models, temporal-sequential alert classification machine learning models, categorical-sequential alert classification machine learning models, alert classification machine learning models, and second-order alert classification machine learning models.

Abstract: Techniques for automated machine learning (ML) pipeline exploration and deployment are described. An automated ML pipeline generation system allows users to easily construct optimized ML pipelines by providing a dataset, identifying a target column in the dataset, and providing an exploration budget. Multiple candidate ML pipelines can be identified and evaluated through an exploration process, and a best ML pipeline can be provided to the requesting user or deployed for production inference. Users can configure, monitor, and adapt the exploration at multiple points in time throughout.