Data is transmitted in accordance with a parameter. For a metric associated with transmission of the data, a response to a stochastic error state of the metric includes making a first adjustment to the parameter in a first direction. A response to a deterministic error state of the metric includes making a second adjustment to the parameter in a second direction, where the second direction is in opposition to the first direction. A transition point between the two states is identified, and a response to the identification is made.

Data is transmitted in accordance with a parameter. For a metric associated with transmission of the data, a response to a stochastic error state of the metric includes making a first adjustment to the parameter in a first direction. A response to a deterministic error state of the metric includes making a second adjustment to the parameter in a second direction, where the second direction is in opposition to the first direction. A transition point between the two states is identified, and a response to the identification is made.

A disclosed example to determine a migration recommendation of a service between geographic regions includes: a graph generator to generate an interaction graph, the interaction graph including first and second nodes and an edge therebetween, the first node representative of a first service in a first geographic region, the second node representative of a second service in a second geographic region, and the edge representative of a network path of interactions between the first and second services; a weighing engine to determine a weight value of the edge between the first and second services based on a count of network interactions between the first and second services and a real-time latency between the first and second services; and a recommendation engine to generate a migration recommendation to migrate the first service to the second geographic region based on the weight value of the edge.

The invention relates to a communication network which is configured to enable instantiation of network slices which represent virtual networks. The invention further relates to a content application server (CAS) for providing an application service via a network slice to at least one user equipment (UE). A network function (SREF) is provided which on the one hand may have access to one or more slice management network functions (CSMF, NSMF, NSSMF) and on the other hand may be accessible to the CAS. The network function (SREF) may abstract the properties of the network slice using a data structure representing a slice object and expose these abstracted properties to the CAS by providing access to the slice object. The slice object may comprise modifiable properties which represent modifiable properties of the corresponding network slice. The CAS may at least in part manage the network slice by modifying one or more modifiable properties, and the SREF may then translate the modified properties into one or more requests for the slice management network functions. The SREF may thereby enable the CAS to perform slice management, while at the same time reducing the complexity of the slice management.

The invention relates to a communication network which is configured to enable instantiation of network slices which represent virtual networks. The invention further relates to a content application server (CAS) for providing an application service via a network slice to at least one user equipment (UE). A network function (SREF) is provided which on the one hand may have access to one or more slice management network functions (CSMF, NSMF, NSSMF) and on the other hand may be accessible to the CAS. The network function (SREF) may abstract the properties of the network slice using a data structure representing a slice object and expose these abstracted properties to the CAS by providing access to the slice object. The slice object may comprise modifiable properties which represent modifiable properties of the corresponding network slice. The CAS may at least in part manage the network slice by modifying one or more modifiable properties, and the SREF may then translate the modified properties into one or more requests for the slice management network functions. The SREF may thereby enable the CAS to perform slice management, while at the same time reducing the complexity of the slice management.

An apparatus, systems, methods, and the like, for autonomous scaling of Internet services, such as scaling Internet access speeds, provided by a communications network to one or more connected networks is provided. The service-providing network may include one or more virtual interface devices through which the connected network or device may access the Internet. The connected networks may utilize these virtual devices to communicate with the broader Internet to exchange data. For example, the connected networks may be associated with a common access point to the service providing network, such as an on-net building or other common location that utilizes the same device or devices to access the network to receive services from the network. In some implementations, the virtual network access devices may be instantiated on one or more network computing devices, such as an application server or other configurable computing and networking device.

An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

Systems and methods are described herein for managing information transmitted between and/or within communication networks. Aspects discussed herein relate to monitoring and characterizing data flows with network and/or latency parameters, such as a time-to-buffer (TTB) parameter. Latency managers, network components, or other suitable devices operating in a communication network may utilize TTB parameter information as a management mechanism throughout the communication network to negotiate and schedule the delivery of data packets in view of a variety of factors, e.g., network performance, application priority, and the like. Such devices may be further configured to modify network or routing policies based on network performance and latency information obtained from and/or shared by various network components and devices in the communication network.

Systems and methods are described herein for managing information transmitted between and/or within communication networks. Aspects discussed herein relate to monitoring and characterizing data flows with network and/or latency parameters, such as a time-to-buffer (TTB) parameter. Latency managers, network components, or other suitable devices operating in a communication network may utilize TTB parameter information as a management mechanism throughout the communication network to negotiate and schedule the delivery of data packets in view of a variety of factors, e.g., network performance, application priority, and the like. Such devices may be further configured to modify network or routing policies based on network performance and latency information obtained from and/or shared by various network components and devices in the communication network.