The present invention provides a method and device for data shunting and relates to the technical field of communications. The present invention solves the problem that the requirements for the Service Quality can not be satisfied, because the shunted data can't be transmitted in the shunting network based on appointed Service Quality; and the reliability of data transmission and the system resource utilization are decreased. The method concretely comprises the following steps: the first network device of the first network determines the second Quality of Service parameter of the second network according to the first Quality of Service parameter of data to be transmitted in the first network; the first network device transmits some or all of data to be transmitted to the second network device of the second network according to the second Quality of Service parameter. The method can be applied to data shunting.

Various embodiments associated an inter-network policy that is implemented for use across multiple networks are described. Individual networks can have individual policies that govern how communications are handled, how resources are allocated, and other metrics. When individual networks work together, these networks can experience problems if their individual policies conflict with one another. Therefore, the inter-network policy can be generated that facilitates the individual networks working together.

The present disclosure provides a resource allocation method of a heterogeneous network under intermediate processing constraints. The method includes: generating a preference list for each intermediate node; obtaining network topology information of the heterogeneous network, and calculating a transmission path length of the data stream passing through the intermediate node according to the network topology information; obtaining a quasi-service list of each intermediate node; obtaining a mapping relationship between the intermediate nodes and the data flows according to the quasi-service list of each intermediate node, and sequentially scheduling the data flows according to the mapping relationship and the preference list of each intermediate node. A device for resource allocation of a heterogeneous network under intermediate processing constraints and a non-transitory computer storage medium are also disclosed.

A resource orchestration framework may include vertices representing physical network nodes and a resource orchestrator. The resource orchestrator may obtain a service function chain specifying, for multiple service functions, mappings of the service functions to respective physical nodes along a specified forwarding path, and a first starting time at which to instantiate resources for a first service function in the chain on the first mapped physical node. The resource orchestrator may instantiate the resources for the first service function on the first mapped physical node at the first starting time, prior to arrival of a first packet in a packet flow for the service function chain at the first mapped physical node and, subsequently, instantiate resources for a second service function in the chain on the second mapped physical node at a second starting time, prior to arrival of the first packet at the second mapped physical node.

Method for an automatic provisioning of a customized cloud stack comprising a customized infrastructure of servers, software and services, by using a number of domain specific languages, model-to-model transformations and code generators, wherein a first domain specific language is used to define a model of software and services to be provisioned on particular hosting units that are defined by a user, wherein the hosting units are mapped to a general model of the infrastructure of the customized cloud stack by an execution engine; and wherein the general model is generated by a second domain specific language, transformed by the execution engine and mapped to a model conforming to a metamodel of a third domain specific language which is used to provision the infrastructure according to the particular hosting units defined by the first domain specific language; wherein files for initialization of a particular server within the infrastructure of servers are generated by particular code generators according to the model defined by the third domain specific language and weaved into userdata for specifying particular software and services, wherein the userdata are passed when particular servers are started and wherein particular code generators are used to produce consumers of services generated by the third domain specific language for provisioning the infrastructure of the customized cloud stack as specified in respective hosting units.

Systems and methods for identifying a mapping solution for a virtual network request are disclosed. A resource orchestrator may receive a virtual network request specifying resources required on physical nodes in a network and generate a mapping solution for the request. Generating the mapping solution may include identifying vertices at which a first resource is available, mapping each identified vertex to the first resource in a respective candidate chaining solution, determining, for each identified vertex, a number of messages including the candidate chaining solution to be forwarded to respective neighbor vertices, and forwarding the determined number of messages to the respective neighbor vertices. Determining the number of messages to be forwarded may include applying a message forwarding policy to reduce the total number of messages forwarded between vertices when generating the mapping solution. The policy may be dependent on the topology of the virtual network request.

Systems and methods for performing distributed virtual network embedding are disclosed. A resource orchestrator may receive a virtual network request specifying a set of virtual nodes, a set of virtual links, each connecting two virtual nodes in a mesh topology, and resource requirements for some virtual nodes. The orchestrator may partition the virtual network request into multiple sub-requests, each specifying a linear topology for a subset of the virtual nodes and links within the mesh topology. The sub-requests may collectively include all virtual links within the mesh topology with no overlapping links. Resource orchestrators may collaborate to compute, independently for each sub-request, a respective chaining solution in which each virtual node is mapped to a physical node having resources sufficient to implementing the virtual node. A resource orchestrator may combine the respective chaining solutions for each of the sub-requests to generate a mapping solution for the virtual network request.

Hierarchical Software Defined Network (SDN) architectures can be used to reduce complexity of traffic engineering in large or diverse network environments. In hierarchical SDN architectures, a network is sub-divided into multiple regions, and each region is assigned to a different SDN controller. Network status information is collected and consolidated at a regional level, and fed upstream through the SDN control plane until it reaches a root SDN controller. The root-SDN controller computes cost-based parameters, which are distributed to regional SDN controllers for local provisioning. The cost-based parameters can include Lagrangian variables estimations or other parameters that constrain regional traffic engineering optimization in a manner that advances global traffic engineering objectives.

A system and method reducing grant uncertainty is described. The present system and method is implemented in a two network system having a front haul network, such as an LTE network, and a backhaul network, such as a DOCSIS network. In practice, the present system determines when requests for network resources are not complete satisfied and auto generates a new grant or modifies an existing grant to accommodate the portion of the previous request that was not satisfied. Embodiments are shown and described where implementation occurs in the front haul system and in the backhaul system.

Data transmission method includes a first mobile device linking a first wireless network with a first linking quality and a second wireless network with a second linking quality. The first mobile device acquires a sum of first bandwidth requirements and a sum of second bandwidth requirements of application programs currently running. The first mobile device acquires a bandwidth allocation result according to the first linking quality, the second linking quality, the sum of first bandwidth requirements, and the sum of second bandwidth requirements. The first mobile device performs data link to the first wireless network and/or the second wireless network according to the bandwidth allocation result, and then performs data transmission of the application programs. The sum of second bandwidth requirements is greater than the sum of first bandwidth requirements. The bandwidth allocation result corresponds to a bandwidth usage value for all application programs.