Embodiments of the present invention provide a packet processing method, a forwarding plane device and a network device, the method includes: receiving, by a forwarding plane device, a first packet transmitted by a user, where an identity of the user is comprised in the first packet, and a forwarding table is comprised in the forwarding plane device; determining, by the forwarding plane device, an identity of a service according to a corresponding relationship between the identity of the user and the identity of the service as well as the identity of the user in the first packet; generating, by the forwarding plane device, a second packet by encapsulating the first packet with the identity of the service; and transmitting the second packet to a network device, to enable the network device to manage the service according to the identity of the service in the second packet.

A service chain accommodation device includes an influence coefficient calculation unit that calculates an influence coefficient indicating that an influence at the time of processing failure of a service chain is greater for a VNF located in a subsequent stage of a service chain and a VNF shared among a plurality of service chains, a residual resource calculation unit that corrects an amount of residual resources that can be accommodated for each of the VNFs through which the service chain passes, and an accommodation design unit that assigns a new service chain on the basis of the amount of the residual resources.

Hybrid edge computing that includes a nimble framework that identifies services for available in a marketplace. The nimble framework defines a location for computing the services selected from the group consisting of a center server, an edge provision server and an edge node. The hybrid edge computing further includes a third party provider making are request for a service to the nimble framework. The hybrid edge computing further includes a virtualized service being provided by the nimble framework to the third party provider including a matched service to the third party provider request for the service, and an optimal location for computing.

The present disclosure is directed to methods and apparatus for tuning implementations of machine learning models for resource-constrained devices. In various embodiments, computer-readable instructions that utilize a trained machine learning model during execution of the computer-readable instructions on a resource-constrained device may be evaluated. Based on the evaluating, it may be determined that an amount of a first computing resource required by the resource-constrained device to execute the computer-readable instructions fails to satisfy a constraint related to the first computing resource. Based on the constraint, one or more candidate alterations to the computer-readable instructions may be identified.

The present disclosure is directed to methods and apparatus for tuning implementations of machine learning models for resource-constrained devices. In various embodiments, computer-readable instructions that utilize a trained machine learning model during execution of the computer-readable instructions on a resource-constrained device may be evaluated. Based on the evaluating, it may be determined that an amount of a first computing resource required by the resource-constrained device to execute the computer-readable instructions fails to satisfy a constraint related to the first computing resource. Based on the constraint, one or more candidate alterations to the computer-readable instructions may be identified.

Techniques for preventing concurrent execution of an infrastructure orchestration service are described. Worker nodes can receive instructions, or tasks, for deploying infrastructure resources and can provide heartbeat notifications to scheduler nodes, also considered a lease. A signing proxy can track the heartbeat notifications sent from the worker nodes to the scheduler node. The signing proxy can receive requests corresponding to a performance of the tasks assigned to the worker nodes. The signing proxy can determine whether the lease between each worker node and the scheduler is valid. If the lease is valid, the signing proxy may make a call to services on behalf of the worker node, and if the lease is not valid, the signing proxy may not make a call to services on behalf of the worker node. Instead, the signing proxy may cut off all outgoing network traffic, blocking access of the worker node to services.

Techniques for preventing concurrent execution of an infrastructure orchestration service are described. Worker nodes can receive instructions, or tasks, for deploying infrastructure resources and can provide heartbeat notifications to scheduler nodes, also considered a lease. A signing proxy can track the heartbeat notifications sent from the worker nodes to the scheduler node. The signing proxy can receive requests corresponding to a performance of the tasks assigned to the worker nodes. The signing proxy can determine whether the lease between each worker node and the scheduler is valid. If the lease is valid, the signing proxy may make a call to services on behalf of the worker node, and if the lease is not valid, the signing proxy may not make a call to services on behalf of the worker node. Instead, the signing proxy may cut off all outgoing network traffic, blocking access of the worker node to services.

A device determines, from a network slice template associated with a network slice, a quality of service (QoS) profile for the network slice that includes performance metrics for corresponding QoS parameters associated with providing a service. The device monitors performance of the network slice in association with the QoS profile, and determines, based on the performance, that a performance indicator for a QoS parameter of the network slice is outside a threshold range of a performance metric. The device determines, based on the performance information and the QoS profile, a slice modification to the network slice template for the network slice, where the slice modification is configured to cause the performance indicator to be within the threshold range of the performance metric. The device causes a network slice orchestrator to update an instantiation of the network slice according to the slice modification and the network slice template.

Techniques are described with regard to application integration management. An associated computer-implemented method includes receiving from at least one user a request to map a plurality of data fields associated with a workflow during an application integration session and automatically mapping all data fields among the plurality of data fields for which a mapping confidence value is greater than or equal to a confidence level threshold set at a default value. The method further includes receiving mapping evaluations from each of the at least one user for each automatic data field mapping based upon the confidence level threshold set at the default value, processing at least one confidence level threshold adjustment input selection from one or more of the at least one user, and deriving at least one candidate default confidence level threshold value based upon the at least one confidence level threshold adjustment input selection.

Techniques are described for enabling users of a cloud provider network to define and use custom “virtual regions” comprised of selected sets of cloud provider network infrastructure locations. A cloud provider network enables users to obtain information about available infrastructure locations and to identify locations that satisfy a set of performance characteristics and other parameters. Once a set of desirable infrastructure locations have been identified, users can request the creation of a virtual region to be associated with their user account, where the request specifies a set of infrastructure locations to comprise the virtual region. Once a virtual region is created, users can cause computing resources to be launched into or otherwise associated with the virtual region in a manner similar to the use of existing regions and availability zones.