WO2018019184A1 - 网络切片方法和系统 - Google Patents

网络切片方法和系统 Download PDF

Info

Publication number
WO2018019184A1
WO2018019184A1 PCT/CN2017/093856 CN2017093856W WO2018019184A1 WO 2018019184 A1 WO2018019184 A1 WO 2018019184A1 CN 2017093856 W CN2017093856 W CN 2017093856W WO 2018019184 A1 WO2018019184 A1 WO 2018019184A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
network slice
service flow
resource
transmitted
Prior art date
Application number
PCT/CN2017/093856
Other languages
English (en)
French (fr)
Inventor
黄勇
林栋�
汪漪
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP17833487.6A priority Critical patent/EP3471341B1/en
Publication of WO2018019184A1 publication Critical patent/WO2018019184A1/zh
Priority to US16/239,269 priority patent/US11381452B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/0816—Configuration setting characterised by the conditions triggering a change of settings the condition being an adaptation, e.g. in response to network events
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0806—Configuration setting for initial configuration or provisioning, e.g. plug-and-play
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0893—Assignment of logical groups to network elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0894—Policy-based network configuration management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0895—Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/20—Traffic policing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/74—Admission control; Resource allocation measures in reaction to resource unavailability
    • H04L47/745—Reaction in network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network

Definitions

  • the present application relates to the field of communications, and in particular, to a network slicing method, a network slicing system, a network node, and a terminal device.
  • a narrowband network carries voice services
  • a synchronous digital hierarchy (SDH) network carries wireless backhaul services.
  • SDH synchronous digital hierarchy
  • Each network can achieve efficient management and unification of business performance for a defined business.
  • TCP/IP transmission control protocol/internet protocol
  • IP transmission control protocol/internet protocol
  • IP Internet Protocol
  • Physical network to host. On a physical network, how to ensure the performance of each business becomes a common concern. It is generally accepted in the industry that the future network should be a plurality of virtual network slices coexisting on a physical network, so as to achieve the best balance between the efficient performance of the service and the low maintenance management cost.
  • vCPU virtual central process units
  • vStorage virtual storage
  • vCPU virtual central process units
  • vStorage virtual storage
  • the network resources of these fragments are matched and provided to different applications with sufficient flexibility, while ensuring the network requirements of various applications and the efficient use of the network.
  • DC networks and WAN networks all use differentiated services (DiffServe) and use large bandwidth to ensure service quality.
  • DiffServe differentiated services
  • SDN software defined network
  • the embodiment of the present application provides a network slicing method, which enables a network slice to meet different types of service pairs. The needs of the network.
  • the first aspect provides a network slicing method, including: acquiring metadata information of each of the plurality of service flows, where the metadata information of the service flow is used to indicate a traffic characteristic of the service flow;
  • the metadata information of the plurality of service flows determines a classification policy, where the classification policy includes a classification criterion of at least one service flow class, where the classification criterion includes a range or value description of the metadata of the service flow class;
  • the classification policy determines a network slice policy, where the network slice policy is used to divide the network into at least one network slice, where the network slice policy includes a mapping relationship between a service flow class and a network slice, and a slice weight of the network slice, The slice weight represents a ratio of resources of the network slice to resources of the network; determining a resource node policy according to the network slice policy, the resource node policy including at least one resource node in the network and the at least one network slice a mapping relationship, and a resource weight of each resource node, the resource weight representation of the resource
  • the network slicing method in the embodiment of the present application can obtain a classification policy based on the metadata information of the service flow, and according to the classification standard in the classification policy, an accurate requirement of different types of service flows on the network can be obtained.
  • the network slice weight of the network slice can be determined based on the service flow to the network requirement, so that the network slice can meet the requirements of different types of services to the network while satisfying the efficient utilization of the network.
  • the network slicing policy can be dynamically adjusted, so that the network slicing can meet the requirements of the dynamically changing different types of services to the network.
  • the method further includes: sending, to the resource node, configuration information, where the configuration information includes an identifier of the at least one network slice corresponding to the resource node, and a resource of the resource node.
  • the weight information is used to indicate a resource weight of the resource node.
  • the determining, according to the classification policy, a network slicing policy comprising: determining the network slicing policy according to the classification policy and a business support system (BSS) business policy; or The network slicing policy is determined according to the classification policy and a business policy of an Operation Support System (OSS).
  • BSS business support system
  • OSS Operation Support System
  • the fragmentation decision collaboration module is based on a business strategy for a business segmentation, which can control the business segmentation to meet commercial needs and bring commercial value.
  • the resource node includes a distributed network node, where the resources of the network slice include the number of packet processing units of the distributed network node, the number and length of the packet sending queue, and the physical port bandwidth. And/or said resource node comprises a distributed computing node, said network sliced resource comprising a virtual central processor number of the computing node; and/or said resource node comprising a distributed storage node, said network sliced resource The number of storage units including storage nodes.
  • the resource weight of the resource node includes a ratio of a bandwidth occupied by a physical port of the resource node to a total bandwidth of the physical port of each network slice corresponding to the resource node.
  • the metadata information of the service flow includes quintuple information of a packet of the service flow.
  • the metadata information of the service flow includes at least one of the following information: The establishment time of the service flow, the burstiness of the message in the service flow, the average length of the message in the service flow, and the transmission time interval between adjacent messages in the service flow, The total length of the traffic flow, the total duration of the traffic flow, and the average rate of the traffic flow.
  • the determining, according to the metadata information of the multiple service flows, the classification policy including: according to an average length of the packets of each of the multiple service flows, Having an average interval of adjacent packets of each of the plurality of service flows, and a total length of each of the plurality of service flows, classifying the plurality of service flows; determining, according to a result of the classification Classification strategy.
  • the method further includes: receiving, by the terminal device, a request message, where the request message carries quintuple information of the message in the service flow to be transmitted, where the request message is used for requesting The network slice information corresponding to the service flow to be transmitted; determining the network slice information according to the quintuple information and the network slice policy; sending a response message to the terminal device, where the response message carries the network And slicing information, so that the terminal device transmits the service flow to be transmitted according to the network slice information.
  • the method further includes: receiving a report message sent by the terminal device, where the report message carries metadata information of the service flow to be transmitted.
  • the second aspect provides a method for transmitting a service flow, including: receiving a service flow to be transmitted, where the service flow to be transmitted carries a network slice identifier, the network slice identified by the network slice identifier, and the to-be-processed Corresponding to the service flow class to which the transmitted service flow belongs; matching the flow table according to the network slice identifier; sending the service flow to be transmitted according to the matched flow table and the resource weight, where the resource weight is the network slice The bandwidth weight of the corresponding physical port of the network node.
  • the method for transmitting a service flow in the embodiment of the present application the service flow is scheduled by the network slice to which the service flow belongs, to avoid mutual interference between service flows of different service types, and to meet performance requirements of service flows of different service types. .
  • the matched flow table includes an execution field, where the execution field includes a physical port and a queue identifier corresponding to the network slice, where the sending is performed according to the matched flow table and bandwidth weight.
  • the service flow to be transmitted is sent to the queue identified by the queue identifier according to the matched flow table, and the physical port is configured according to the bandwidth weight according to the bandwidth The traffic to be transmitted is scheduled.
  • the method before receiving the service flow to be transmitted, the method further includes: receiving configuration information sent by the centralized controller, where the configuration information carries network slice information and the resource weight, the network The slice information is used to indicate the network slice.
  • the third aspect provides a method for transmitting a service flow, which includes: receiving a service flow to be transmitted; and sending a request message to the centralized controller, where the request message carries the report in the service flow to be transmitted
  • the quintuple information of the text the request message is used to request network slice information corresponding to the service flow to be transmitted; and receive network slice information sent by the centralized controller according to the quintuple information, the network
  • the network slice indicated by the slice information corresponds to the service class to which the service flow to be transmitted belongs; the service flow to be transmitted is sent to the network node, and the packet in the service flow to be transmitted carries the identifier of the network slice Network slice ID.
  • the method for transmitting a service flow in the embodiment of the present application the service flow is scheduled by the network slice to which the service flow belongs, to avoid mutual interference between service flows of different service types, and to meet performance requirements of service flows of different service types. .
  • the method further includes: sending a report message to the centralized controller, where The report message carries metadata information of the service flow to be transmitted.
  • the method further includes: saving a correspondence between the service flow to be transmitted and the network slice.
  • a network slicing system for performing the method of the first aspect or any possible implementation of the first aspect.
  • the network slicing system comprises means for performing the method of the first aspect or any of the possible implementations of the first aspect.
  • the application provides a network node for performing the method in any of the possible implementations of the second aspect or the second aspect.
  • the network node comprises means for performing the method of the second aspect or any of the possible implementations of the second aspect.
  • the application provides a terminal device for performing the method in any of the possible implementations of the third aspect or the third aspect.
  • the terminal device comprises means for performing the method of any of the third aspect or any of the possible implementations of the third aspect.
  • a network slicing system comprising a receiver, a transmitter, a processor, a memory, and a bus system.
  • the receiver, the transmitter, the processor and the memory are connected by a bus system, the memory is used for storing instructions, and the processor is configured to execute instructions stored in the memory to control the receiver to receive signals and control the transmitter to send signals.
  • the processor executes the instructions stored in the memory, the method of causing the processor to perform the first aspect or any of the possible implementations of the first aspect is performed.
  • the application provides a network node that includes a receiver, a transmitter, a processor, a memory, and a bus system.
  • the receiver, the transmitter, the processor and the memory are connected by a bus system, the memory is used for storing instructions, and the processor is configured to execute instructions stored in the memory to control the receiver to receive signals and control the transmitter to send signals.
  • the processor executes the instructions stored in the memory, the method of causing the processor to perform the second aspect or any of the possible implementations of the second aspect is performed.
  • the present application provides a terminal device including a receiver, a transmitter, a processor, a memory, and a bus system.
  • the receiver, the transmitter, the processor and the memory are connected by a bus system, the memory is used for storing instructions, and the processor is configured to execute instructions stored in the memory to control the receiver to receive signals and control the transmitter to send signals.
  • the processor executes the instructions stored in the memory, the method of causing the processor to perform the third aspect or any of the possible implementations of the third aspect is performed.
  • the application provides a computer readable medium for storing a computer program, the computer program comprising instructions for performing the method of the first aspect or any of the possible implementations of the first aspect.
  • the present application provides a computer readable medium for storing a computer program, the computer program comprising instructions for performing the method of any of the second aspect or any of the possible implementations of the second aspect.
  • the present application provides a computer readable medium for storing a computer program, the computer program comprising instructions for performing the method of any of the third aspect or any of the possible implementations of the third aspect.
  • FIG. 1 is a schematic block diagram of a system architecture in accordance with an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a method of network slicing according to an embodiment of the present application.
  • FIG. 3 is a schematic block diagram of another network architecture in accordance with an embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a method for transmitting a service flow according to an embodiment of the present application.
  • FIG. 5 is a schematic block diagram of another network architecture in accordance with an embodiment of the present application.
  • FIG. 6 is a schematic block diagram of a network slicing system in accordance with an embodiment of the present application.
  • FIG. 7 is a schematic block diagram of a network node in accordance with an embodiment of the present application.
  • FIG. 8 is a schematic block diagram of a terminal device according to an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of a network slicing system according to an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a network node according to an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
  • Network slicing technology divides a physical network virtualization into multiple logical networks that can coexist at the same time, providing the most suitable resource configuration and network topology for different types of services. Specifically, it is based on a physical network to flexibly assemble and orchestrate capabilities, and end-to-end network slicing to meet the needs of massive applications for network differentiation.
  • industrial control slicing can provide ultra-low latency to achieve precise control and interaction of industrial robots in disaster relief, power maintenance and other scenarios; utility slicing can provide low-power, ultra-compact connections to meet smart water meters, gas meters, Demand for public services such as electricity meters.
  • the embodiment of the present application provides a method for network segmentation.
  • the method of network slicing in the embodiment of the present application can be applied to the system architecture shown in FIG. 1.
  • the control plane of the system shown in FIG. 1 is composed of an application (App) sensing analysis module, a fragmentation decision coordination module, and a slice control module.
  • the control plane can perform a network slicing method based on an embodiment of the present application.
  • the App-aware analysis module can obtain metadata information of a service flow in a data plane, and output a classification policy according to the metadata information.
  • the data plane here includes resource nodes in the infrastructure network between the end system-transmitting device, the end system-receiving device, and the end system-transmitting device and the end system-receiving device.
  • the fragmentation decision coordination module can determine the network slicing strategy based on the classification strategy, and can output the network slicing policy to the fragmentation control module.
  • the fragmentation control module can determine the resource node policy based on the network slicing strategy. And configure end-to-end network slicing by sending configuration information to the network node.
  • the network is divided into network slice 1 and network slice 2.
  • Network slice 1 and network slice 2 respectively rely on forwarding, computing and storage resources serving network slices in a plurality of network nodes, computing nodes and storage nodes distributed in the base network.
  • the network node, the computing node, and the storage node may be referred to as resource nodes.
  • the network slice 1 and the network slice 2 are respectively composed of a forwarding function, a processing function, and a control function serving these forwarding functions and processing functions in a network composed of these resources.
  • the BSS business policy or the OSS business strategy provided by the BSS/OSS system shown in FIG. 1 can be used as a reference information for controlling the slice to face the network.
  • the control plane can divide the network into network slice 1 and network slice 2 based on the reference BSS business strategy or OSS business strategy.
  • Business strategies including web applications for a business user, require a specified number of network slice guarantees. For a business user or a certain type of application, specify the total amount of resources occupied by the network, the minimum or maximum ratio.
  • Fragmentation decision coordination module under the premise of business strategy When making a sharding decision, the network slicing strategy that guarantees the decision output is consistent with these business strategies.
  • network slice 1 and network slice 2 can be regarded as two virtual networks established between the end system with the communication relationship - the transmitting device to the end system - the receiving device.
  • Each virtual network carries the traffic of a particular business performance requirement.
  • Distributed-based congestion control of the data plane is called fast control.
  • fast control refer to more existing technologies, such as explicit congestion notification (ECN) technology and TCP explicit congestion echo (ECE) feedback mechanism (RFC3168) technology.
  • App-aware analysis module the slice decision coordination module, and the slice control module in the system architecture described in FIG. 1 may be three physical devices respectively, or may be three functional modules deployed on one entity.
  • the App-Awareness Analysis Module and the Fragmentation Decision Collaboration Module may be integrated on one physical device, or the fragmentation decision collaboration module and the slice control module may be integrated on one physical device, which is not limited in the embodiment of the present application.
  • FIG. 2 is a schematic flow chart of a network slicing method according to an embodiment of the present application.
  • a network slicing method according to an embodiment of the present application will be described in detail with reference to FIG. 1 and FIG.
  • S210 Acquire metadata information of each of the plurality of service flows.
  • the metadata information is used to indicate the traffic characteristics of the traffic flow.
  • the service flow involved in the embodiment of the present application refers to an order defined by a quintuple information, that is, an internet protocol (IP) address, a destination IP address, a protocol number, a source port number, and a destination port number.
  • IP internet protocol
  • the service flow mentioned here may be a historical service flow or a service flow to be transmitted.
  • the service flow is the service flow to be transmitted, the information of the first few messages in the service flow can be obtained, and the metadata information of the service flow extracted from the first few messages can support the classification of the service flow.
  • the App-awareness analysis module may collect metadata information of multiple service flows in the network from the data plane.
  • the App-awareness analysis module can periodically collect metadata information of the service flow in the data plane.
  • the App-aware analysis module can analyze the traffic characteristics of the service flow, for example, the service flow bandwidth characteristics, the delay characteristics, and the like.
  • the metadata information may include quintuple information of the traffic flow. Further, the metadata information may further include: a setup time of the service flow, an average length of the packet in the service flow, a burst of the packet in the service flow, and an interval between two adjacent packets in the service flow. The sending interval, the total length of the traffic, the total duration of the traffic, and the average rate of the traffic.
  • the packet burst of the service flow involved in the embodiment of the present application refers to the sum of the maximum number of consecutive packets arriving in the queue or the length of consecutive packets in the same service flow during queue scheduling.
  • the App-aware analysis module can identify the traffic characteristics of the service flow according to the metadata information, and can aggregate the flows with the similar traffic characteristics into one class. For example, the App-awareness analysis module can determine the service type by using the destination address and the source address of the packet in the service flow, or can estimate the application type by using the well-known port number carried in the packet in the service flow. For example, file transfer protocol (FTP) signaling port number TCP21, domain name system (DNS) port number UDP67, hypertext transfer protocol (HTTP) port number TCP80, and so on.
  • FTP file transfer protocol
  • TCP21 domain name system (DNS) port number UDP67
  • HTTP hypertext transfer protocol
  • the App-awareness analysis module aggregates flows with similar traffic characteristics into one type of process, which can be understood as a process of classifying multiple service flows.
  • a classification algorithm in the prior art can be used.
  • a clustering analysis algorithm in machine learning may be employed, but the comparison of the embodiments of the present application is not limited.
  • the App Perception Analysis module can obtain a classification strategy by classifying multiple service flows.
  • the classification strategy may include the classification criteria of the business flow class, that is, the classification criteria of each type of business flow.
  • a type of service flow includes a description of traffic flow characteristics from a plurality of metadata dimensions, and a metadata dimension is represented by a value range or a specific value of the service flow in the metadata.
  • the average length of a packet is a kind of metadata. Based on the dimension, the average length of the packet is greater than 200 bytes (bytes), and less than 400 bytes belongs to the service flow.
  • the classification strategy may also include the number of classifications.
  • the classification criteria reflect the traffic characteristics of this type of business flow, or the type of application of this type of business flow.
  • the classification criteria may include features such as message length characteristics of the traffic flow, duration characteristics of the traffic flow, burst characteristics of the traffic flow, and the like.
  • the classification standard may further include distribution data characteristics of the service flow, such as mean, variance, local peak, and the like.
  • multiple service flow classes may be obtained, and the multiple service flow classes may include F-type service flows.
  • F-type service flow is only a generation name, and the F-type service flow may also be referred to as a type A service flow, and may also be referred to as a class B service flow.
  • type A service flow and may also be referred to as a class B service flow.
  • the traffic characteristic information of the F-type service flow may further include the following contents:
  • the classification criteria of the foregoing F-type service flows are only examples.
  • the embodiments of the present application are not limited thereto, and the classification criteria of the traffic characteristics of other standard traffic volumes fall within the scope of the present application.
  • the average length of the packets of each of the multiple service flows, and the neighboring of each of the multiple service flows may be The average interval of the packets and the total length of each of the plurality of service flows classify the plurality of service flows; and then determine the classification policy according to the result of the classification.
  • the App-aware analysis module may classify multiple service flows based on the average length of the packets in the service flow, the average interval of adjacent packets, and the total length of the service flow.
  • the classification criteria of each type of service flow obtained may include the average length feature of the message of the service flow, the average interval feature of the adjacent message, and the total length feature of the service flow.
  • the fragmentation decision collaboration module may receive the classification policy sent by the App-aware analysis module, and the network slice policy may be determined according to the classification criteria of each type of service flow.
  • the network slicing policy may include a correspondence between a service flow class to which the service flow belongs and a network slice, and a slice weight of each network slice.
  • the slice weight of the network slice indicates the ratio of the resources of the network slice to the resources of the network.
  • the resources of the network refer to the overall resources of the network.
  • the network overall resources may include distributed network node resources, distributed computing resources, and distributed storage resources.
  • the resources of the network slice may include network node resources, distributed computing resources, and distributed storage resources.
  • the distributed network node resource may include the number of packet processing units of the network node, the number and length of the packet sending queues, and the physical port bandwidth. Distributed computing resources can be counted Count the number of virtual CPUs for the node.
  • a distributed storage resource may include the number of storage units of a storage node.
  • the number of slices of the network may be determined according to the number of classifications and the classification criteria of each type of service flow, that is, the network is divided into several network slices, and each type of service flow corresponding to each network slice (ie, the service in the above) Stream class) or which type of application, and the slice weight of each network slice.
  • a network slice can correspond to a type of service flow.
  • a network slice can also correspond to multiple types of service flows.
  • the slice weight it may be determined according to the classification standard of the service flow class or the application type corresponding to the service flow class.
  • the network resources can be allocated according to the bandwidth of 80% of the peak bandwidth; when it is determined that the other type of service flow belongs to low delay and large burst service (for example, the average bandwidth of the service flow is less than 100K, The burst is greater than 2Kbytes, the average packet length is less than 300bytes, and the stream length is less than 2kbytes. Network resources can be allocated at 3 times the average bandwidth.
  • the network policy when determining the network slicing policy according to the classification policy, may be determined according to the classification policy and the business support system BSS; or may be determined according to the classification policy and the business policy of the operation support system OSS; or, according to the classification policy, the BSS Business strategy, as well as the business strategy of OSS. Further, the network slicing policy can also be determined according to the available resource conditions of the network.
  • the fragmentation decision cooperation module is based on a business strategy for a business segment, and can control the business segment to meet commercial needs and bring commercial value.
  • the resource node policy may include a mapping relationship between resource nodes and network slices, and resource weights.
  • the resource weight of the resource node represents a ratio of resources of each network slice in at least one network slice corresponding to the resource node to resources of the resource node.
  • the resource nodes may include distributed network nodes, distributed computing nodes, and distributed storage nodes.
  • S240 can be executed by the slice control module shown in FIG. Specifically, after the fragmentation decision coordination module determines the network slicing policy, the network slicing policy may be sent to the fragmentation control module.
  • the fragmentation control module determines the resource node policy based on the network slicing strategy, and completes the decomposition of the slice weights to the distributed network resources, the distributed computing resources, and the distributed storage resources.
  • the slice control module After the slice control module according to the slice weight of each network slice, it is also required to determine a network node resource node corresponding to the specific network slice, that is, to determine which network node resource node the specific network slice falls on. on. Moreover, the fragmentation control module further needs to determine the resource weight, that is, the ratio of the resources of the specific network slice to the resources of the network node resource node, or each of the network node resource nodes respectively. The ratio of resources on. It should be understood that a particular network slice network as referred to herein may refer to any of a plurality of network slices.
  • the network slicing method in the embodiment of the present application can obtain a classification policy based on the metadata information of the service flow, and according to the classification standard in the classification policy, an accurate requirement of different types of service flows on the network can be obtained.
  • the network slice weight of the network slice can be determined according to the network flow according to the service flow, so as to meet the network efficient utilization rate, It enables network slicing to meet the needs of different types of services for the network.
  • the network slicing policy can be dynamically adjusted, so that the network slicing can meet the requirements of the dynamically changing different types of services to the network.
  • the resource weight of the resource node includes a ratio of a bandwidth occupied by a physical port of the resource node to a total bandwidth of the physical port of each network slice corresponding to the resource node.
  • multiple network slices can fall on the same network node.
  • the plurality of network slices occupy a corresponding ratio of the bandwidth of the physical port of the network node.
  • the method may further include: S250, sending configuration information to the resource node.
  • the configuration information may include an identifier of each network slice in the at least one network slice corresponding to the resource node, and resource weight information, where the resource weight information is used to indicate the resource weight of the resource node.
  • S250 can be executed by the slice control module shown in FIG.
  • the fragmentation control module sends the configuration information to the infrastructure node where the corresponding resource is located, that is, the resource node described above. , resource configuration.
  • the fragmentation control module needs to configure each resource node after determining the resource node policy.
  • the resource node can perform resource configuration according to the configuration information.
  • the resource node may determine a network slice corresponding thereto according to the network slice identifier in the configuration information, and according to the resource weight in the configuration information, configure a corresponding ratio resource for the network slice, for example, the resource node is a network.
  • the slice configures the bandwidth resource of the corresponding ratio. In this way, the mapping of the physical network to the logical network is realized.
  • the fragmentation control module can control the bandwidth weight of the port of the part of the network node to achieve dynamic coordination of the network resources of the entire network, thereby satisfying the high bandwidth utilization while satisfying the performance requirements of each service.
  • the method may further include: receiving, by the terminal device, a request message, where the request message carries quintuple information of the packet in the service flow to be transmitted, where the request message is used to request a network slice corresponding to the service flow to be transmitted. And determining, according to the quintuple information and the network splicing policy, network slice information corresponding to the service flow to be transmitted; sending a response message to the terminal device, where the response message carries the network slice information, so that the terminal device transmits the service to be transmitted according to the network slice information. flow.
  • the terminal device may be a network edge device, and specifically, may be a terminal device or a server at the edge of the network.
  • the terminal device may be the end system-sending device shown in FIG. 1.
  • the traffic to be transmitted may be transmitted according to a network slicing policy determined by method 100. Specifically, after the fragmentation decision coordination module has determined the network slicing policy, the fragmentation control module may determine, according to the received quintuple information of the service flow to be transmitted, the service flow class to which the service flow to be transmitted belongs. Further, the network slice corresponding to the service flow to be transmitted may be determined by the mapping relationship between the service flow class and the network slice in the network slicing policy. After determining the network slice corresponding to the service flow to be transmitted, the fragmentation control module may send the network slice information to the terminal device, so that the terminal device transmits the service flow to be transmitted according to the network slice information.
  • the method may further include: receiving a report message sent by the terminal device, where the report message carries metadata information of the service flow to be transmitted.
  • the metadata information of the service flow to be transmitted may be sent to the App-aware analysis module.
  • the App-aware analysis module can dynamically determine the classification policy by analyzing the traffic characteristics of the traffic flow.
  • the fragmentation decision coordination module may determine the network slice policy according to the dynamic classification policy, so that the network slice satisfies different types of services whose states change.
  • FIG. 3 is a schematic diagram of another system architecture according to an embodiment of the present application.
  • the network slicing method of the embodiment of the present application may Used in the system shown in Figure 3.
  • the control plane is composed of an analyzer, a synthesizer and a controller, and can respectively implement the functions of the APP-aware analysis module, the slice decision-making cooperation module, and the slice control module shown in FIG.
  • the analyzer, the synthesizer and the controller are coupled using a database.
  • the network slicing method according to an embodiment of the present application is explained below with reference to FIG. 3. It should be understood that the related terms or words appearing in the content described below in connection with FIG. 3 can be referred to the above description. For the sake of brevity, it will not be elaborated below.
  • each edge terminal device reports service flow metadata information to the analyzer.
  • the terminal device 1 and the terminal device 2 report the metadata information of the service flow to the analyzer.
  • the analyzer aggregates the reports of the terminal device 1 and the terminal device 2, classifies the service flows, obtains a classification policy, and sends the classification policy to the database.
  • the collaborator reads the classification policy in the database to determine the network slicing policy. Further, the collaborator can determine the network slicing strategy by combining the classification policy, the BSS business policy, or the OSS business strategy.
  • the coordinator sends the network slicing policy to the database, and the controller can read the network slicing policy from the database, determine the resource node policy according to the network slicing policy, and then send the configuration information to the corresponding network node. After receiving the configuration information, the network node may configure corresponding resources for the corresponding network slice according to the configuration information.
  • a database is used to implement an interface between entities, and an entity A (for example, an analyzer) sends information I (for example, metadata information) to a database, and entity B (for example, a collaborator) reads from a database.
  • entity A for example, an analyzer
  • entity B for example, a collaborator
  • the process of taking the information I is equivalent to the process in which the entity A sends the information I to the entity B.
  • each entity device on the control plane can interact with each other through a database.
  • analyzers and collaborators can interact through a database.
  • the physical device on the control plane and the terminal device on the data plane can also interact through the database.
  • the analyzer and the terminal device 1 can interact through a database.
  • the network device, the computing device, or the storage device may also interact with each other through a database, which is not limited by the embodiment of the present application.
  • the network slicing method in the embodiment of the present application can obtain a classification policy based on the metadata information of the service flow, and according to the classification standard in the classification policy, an accurate requirement of different types of service flows on the network can be obtained.
  • the network slice weight of the network slice can be determined based on the service flow to the network requirement, so that the network slice can meet the requirements of different types of services to the network while satisfying the efficient utilization of the network.
  • the network slicing policy can be dynamically adjusted, so that the network slicing can meet the requirements of the dynamically changing different types of services to the network.
  • FIG. 4 is a method of transmitting a traffic flow in accordance with an embodiment of the present application. It should be understood that the related terms or words appearing in the content described below in connection with FIG. 4 can be referred to the above description. For the sake of brevity, it will not be elaborated below.
  • the terminal device receives the service flow to be transmitted.
  • the service flow to be transmitted may be one or multiple.
  • the terminal device is a network edge device.
  • the terminal device may be the end system-sending device shown in FIG. 1, and may also be the terminal device 1 shown in FIG.
  • the terminal device sends a request message to the centralized controller.
  • the request message carries the quintuple information of the message in the service flow to be transmitted.
  • the request message is used to request network slice information corresponding to the service flow to be transmitted.
  • the centralized controller herein is only a generation, and the centralized controller may also be referred to as a centralized manager, a slice controller, and the like.
  • the centralized controller can be understood as integrating the App-aware analysis module and the slice decision coordination mode shown in Figure 1.
  • Block and slice control function module A physical device of three functional modules, the centralized controller can also be regarded as a physical device integrating the functions of the three physical devices of the analyzer, the synergizer and the controller shown in FIG.
  • the terminal device sends the quintuple information of the packet in the service flow to be transmitted to the centralized controller.
  • the terminal device may be the fragment control module shown in FIG. 1 or FIG.
  • the controller shown transmits quintuple information of the message in the service flow to be transmitted.
  • the centralized controller may determine network slice information corresponding to the service flow to be transmitted according to the quintuple information and the network slicing policy.
  • the centralized controller stores a network slicing policy, and the network slicing policy stores a correspondence between the service flow class and the network slice.
  • the centralized controller can determine the service flow class to which the service flow to be transmitted belongs by analyzing the traffic characteristics of the service flow to be transmitted. Then, according to the correspondence between the service flow class and the network slice, the network slice information corresponding to the service flow to be transmitted may be determined. Further, the centralized controller may determine a network slice identifier corresponding to the traffic flow to be transmitted.
  • the centralized controller sends a response message to the terminal device, where the response message is used to indicate the network slice information.
  • the centralized controller after determining the network slice corresponding to the service flow to be transmitted, the centralized controller sends a response message to the terminal device to notify the terminal device of the network slice corresponding to the service flow to be transmitted.
  • the slice identifier of the network slice may be carried in the response message.
  • the terminal device sends the service flow to be transmitted to the network node.
  • the network slice identifier may be carried in the packet of the service flow to be transmitted.
  • network node herein may be the distributed network node above.
  • the terminal device may save the correspondence between the service flow to be transmitted and the network slice. Further, the terminal device may save the correspondence between the service flow to be transmitted and the network slice identifier. Then, when the terminal device receives the subsequent packet of the service flow, the network slice identifier can be directly found and the service flow packet carrying the network slice identifier is forwarded based on the saved correspondence. In this way, the forwarding efficiency can be improved while saving the overhead.
  • the centralized controller may send a service flow identification standard corresponding to one slice, that is, corresponding to one or more of the foregoing service flow classification standards, to the terminal device.
  • the terminal device can complete the classification of the service flow and add the network slice identifier to the packet.
  • the network node matches the flow table according to the network slice identifier.
  • the network node may pre-configure a fuzzy flow table with a network slice identifier as a matching item. After the traffic flow to be transmitted arrives, the network node can generate an accurate flow table with the network slice identifier as a match based on the mode matching. On the other hand, the SDN method can also be adopted. The network node sends the first few packets of the service flow to be transmitted to the controller, and the controller sends an accurate flow table with the network slice identifier as a matching item.
  • the network node sends the service flow according to the matched flow table and the resource weight, where the resource weight is the bandwidth weight of the network slice at the network node.
  • the method may further include: 401: the network node receives configuration information sent by the centralized controller, where the configuration information carries an identifier of the network node and the resource weight.
  • the network node may obtain the resource weight corresponding to the network slice by querying the centralized controller.
  • the resource weights can also be obtained through the configuration information that is actively sent by the centralized controller.
  • the network node performs forwarding of the service flow to be transmitted according to the matched flow table and resource weight.
  • the matched flow table includes an execution field, and the execution field may include a physical port and a queue identifier corresponding to the network slice, where the network node sends the service flow to be transmitted according to the matched flow table and the bandwidth weight, including: the network node According to the matched flow table, the service flow to be transmitted is sent to the queue identified by the queue identifier, and the service flow to be transmitted according to the bandwidth weight is scheduled on the physical port.
  • the matching flow table is expressed as follows: ⁇ match:flow2&tagC,action:output to port:queue NE3:C>.
  • the matching match field is used by the network node to match the received service flow flow2 to be transmitted, where tagC is the identifier of the network slice and network slice C.
  • the execution action field is used to indicate that the network node needs to process flow2.
  • one network node may correspond to multiple network slices, and multiple network slices may correspond to multiple queues of the same physical port.
  • a network slice may correspond to a queue, which is not limited in this embodiment of the present application.
  • the network node can use the weighted fair queue (WFQ) algorithm to schedule the traffic to be transmitted at the physical port that is sent to the NE3.
  • the number of queues used by the WFQ may be the number of network slices; the scheduling weight of the WFQ and the queue may be the bandwidth weight of the slice of each network slice corresponding to the physical port of the network.
  • WRR weighted round robin
  • the terminal device sends a report message to the centralized controller.
  • the report message is used to report metadata information corresponding to the service flow to be transmitted.
  • the terminal device may send the quintuple information of the service flow to be transmitted to the centralized controller.
  • the centralized controller can dynamically determine the classification policy by analyzing the traffic characteristics of the traffic flow. Further, the centralized controller may determine the network slicing policy according to the dynamic classification policy, so that the network slice satisfies different types of services whose states change.
  • the method for transmitting a service flow in the embodiment of the present application the service flow is scheduled by the network slice to which the service flow belongs, to avoid mutual interference between service flows of different service types, and to meet performance requirements of service flows of different service types. .
  • the centralized controller 510 shown in FIG. 5 can be used to perform the respective steps performed by the centralized controller in the method illustrated in FIG. NE530 and NE540 may correspond to the network nodes described above.
  • the communication processing procedure for the terminal device 520 to the terminal device 540 and the interaction process between the data plane and the control plane are described in detail below with reference to FIG.
  • the request message may carry the quintuple information of the message in the flow1, and is used to request the centralized controller 510 for the network slice information corresponding to the flow1.
  • the centralized controller 510 can determine that the flow1 corresponds to the network slice B based on the mapping relationship between the service flow class and the network slice in the previously generated network slice policy. The centralized controller 510 transmits this information of the flow1 corresponding network slice B to the terminal device 520 through a response message.
  • the terminal 1 can store the corresponding relationship of the network slice B corresponding to the flow1, and add a tag tagB to the message in the flow1, indicating that the flow1 corresponds to the network slice B.
  • the terminal device 520 then transmits the tagged flow1 to the network, that is, the network device NE530 shown in FIG.
  • NE530 matches the flow table to obtain a matching flow table ⁇ match:flow1&tagB,action:output to port:queue NE540:B>. According to the matched flow table, the NE 530 sends the flow1 to the queue of the corresponding network slice B of the physical port connected to the network device NE540.
  • the flow table matching method and the scheduling of the message can refer to the description above, and for brevity, no further details are provided herein.
  • the NE 530 sends a flow1 to the next network node NE 540 based on the schedule, and the NE 540 performs an operation similar to the match flow table of the NE 530 and the scheduling of the flow 1 . Finally, flow1 is sent to the terminal device 540. After the terminal device 540 receives the flow1, the transfer of flow1 in the network ends.
  • the terminal device 520 sends the metadata information of the packet corresponding to the flow flow1 to the centralized controller 510.
  • the metadata information here may include the settling time of flow1, the quintuple of the packet in flow1, the length of the packet, the current maximum burst of the packet, the receiving interval of the packet and the previous packet, and the current flow1. Total length, current total duration of flow1, etc.
  • the NE 530 and the NE 540 can send the physical port to the centralized controller 510 corresponding to the occupied bandwidth of each network slice and the queue length information of each network slice.
  • the terminal device 550 sends the metadata of the packet in the flow1 corresponding to the flow1, the current maximum burst of the packet, the receiving time interval of the packet and the previous packet, and the current total duration of the flow1 to the centralized information. Controller 510.
  • the centralized controller 510 receives metadata information of a plurality of service flows (including flow1), performs classification calculation, and finally generates a network slice policy.
  • the specific classification method can be referred to the description in the foregoing, and will not be described in detail here.
  • the centralized controller 510 further determines the resource node policy. For details, refer to the foregoing.
  • the centralized controller 510 delivers the configuration information determined according to the resource node policy to the corresponding network forwarding in the data plane, and performs network slice isolation guarantee on the computing and storage infrastructure.
  • the terminal device 550 counts the congestion degree information of the network slice B corresponding to the flow1 on the logical pipe of the terminal device 520 to the terminal device 550.
  • the specific statistics mode can be obtained by using the NE530 and NE540 in the egress queue scheduling based on the corresponding congestion degree information carried in the packet.
  • the terminal device 550 transmits the congestion degree information to the terminal device 520 by carrying the congestion degree information or by using a dedicated message packet.
  • the terminal device 520 adjusts the transmission rate currently corresponding to the network slice B based on the congestion degree information. Specifically, when the feedback is not congested, the transmission rate is increased according to the bandwidth weight of the network slice B; when the feedback is congested, the transmission rate is reduced according to the bandwidth weight and the congestion degree of the network slice B.
  • the method for transmitting a service flow in the embodiment of the present application the service flow is scheduled by the network slice to which the service flow belongs, to avoid mutual interference between service flows of different service types, and to meet performance requirements of service flows of different service types. .
  • the network slicing method and the method for transmitting a service flow according to an embodiment of the present application are described in detail above with reference to FIG. 1 to FIG. 5.
  • a network slicing system according to an embodiment of the present application will be described with reference to FIG.
  • FIG. 6 illustrates a network slicing system 600 in accordance with an embodiment of the present application.
  • the network slicing system 600 includes an application sensing analysis unit 610, a fragmentation decision coordination unit 620, and a fragmentation control unit 630.
  • the application-awareness analysis unit 610 is configured to acquire metadata information of each of the plurality of service flows, where the metadata information of the service flow is used to indicate a traffic characteristic of the service flow;
  • the classification policy includes a classification criterion of at least one service flow class, where the classification criterion includes a range or a value description of the metadata of the service flow class.
  • the fragmentation decision coordination unit 620 receives the classification policy sent by the application-aware analysis unit 610, and is configured to determine a network slice policy according to the classification policy, where the network slice policy is used to divide the network into at least one network slice, where the network slice policy includes a service flow class.
  • the fragmentation control unit 630 receives the network slice policy sent by the fragmentation decision coordination unit 620, and is configured to determine a resource node policy according to the network slice policy, where the resource node policy includes at least one resource node in the network and the at least one network slice a mapping relationship, and a resource weight of each resource node, the resource weight of the resource node indicating a ratio of resources of each network slice in the at least one network slice corresponding to the resource node to resources of the resource node.
  • the application-aware analysis unit 610 in the network slicing system 600 may implement a corresponding process that may be performed by the application-aware analysis module in the method 100.
  • the fragmentation decision-making cooperation unit 620 may implement the implementation method 100.
  • the corresponding process performed by the fragmentation decision coordination module; the slice control unit 630 can implement a corresponding process in the implementation method 100 that can be executed by the slice control single module.
  • various units in network slicing system 600 may also implement other operations or functions that may be implemented by corresponding units in method 100. For the sake of brevity, it will not be repeated here.
  • the network slicing system can obtain a classification policy based on the metadata information of the service flow, and according to the classification standard in the classification policy, an accurate requirement of different types of service flows to the network can be obtained.
  • the network slice weight of the network slice can be determined based on the service flow to the network requirement, so that the network slice can meet the requirements of different types of services to the network while satisfying the efficient utilization of the network.
  • the network slicing system can dynamically adjust the network slicing policy according to the metadata information of the real-time collected service flow, so that the network slicing can meet the dynamic demand of different types of services to the network.
  • FIG. 7 shows a network node 700 in accordance with an embodiment of the present application.
  • the network node 700 includes a first receiving unit 710, a matching unit 720, and a transmitting unit 730.
  • the first receiving unit 710 is configured to receive a service flow to be transmitted, where the service flow to be transmitted carries a network slice identifier, and the network slice identified by the network slice identifier and the service to which the service flow to be transmitted belongs The stream class corresponds.
  • the matching unit 720 is configured to match the flow table according to the network slice identifier received by the first receiving unit.
  • the sending unit 730 is configured to send, according to the flow table and the resource weight matched by the matching unit, the service flow to be transmitted, where the resource weight is a bandwidth weight of the network slice at a corresponding physical port of the network node.
  • the units in the signal network node 700 and the other operations or functions described above in accordance with embodiments of the present application are respectively for the respective flows performed by the network nodes in the method illustrated in FIG. For the sake of brevity, it will not be repeated here.
  • the network node in the embodiment of the present application schedules the service flow by using the network slice to which the service flow belongs, so as to avoid mutual interference between the service flows of different service types, and can meet the performance requirements of service flows of different service types.
  • FIG. 8 shows a terminal device 800 in accordance with an embodiment of the present application.
  • the terminal device 800 includes a first receiving unit 810, a first transmitting unit 820, a second receiving unit 830, and a second transmitting unit 840.
  • the first receiving unit 810 is configured to receive a service flow to be transmitted.
  • the first sending unit 820 is configured to send a request message to the centralized controller, where the request message carries the quintuple information of the packet in the service flow to be transmitted received by the first receiving unit 810, where the request message is used by the first sending unit 810. And requesting network slice information corresponding to the service flow to be transmitted.
  • the second receiving unit 830 is configured to receive network slice information that is sent by the centralized controller according to the quintuple information sent by the first sending unit 820, where the network slice indicated by the network slice information corresponds to a service class to which the service flow to be transmitted belongs.
  • the second sending unit 840 is configured to send, to the network node, a service flow to be transmitted, where the packet in the service flow to be transmitted carries a network slice identifier that identifies the network slice.
  • the terminal device in the embodiment of the present application schedules the service flow by using the network slice to which the service flow belongs, and avoids mutual interference between the service flows of different service types, and can meet the performance requirements of service flows of different service types.
  • FIG. 9 illustrates a network slicing system 900 in accordance with another embodiment of the present application.
  • the network slicing system 900 includes a receiver 910, a transmitter 920, a processor 930, a memory 940, and a bus system 950.
  • the receiver 910, the transmitter 920, the processor 930, and the memory 940 are connected by a bus system 950 for storing instructions for executing instructions stored by the memory 940 to control the receiver 910 to receive.
  • a classification policy Determining, according to the metadata information of the multiple service flows, a classification policy, where the classification policy includes a classification criterion of at least one service flow class, where the service flow class includes at least one of the service flows;
  • a network slicing policy Determining, according to the classification policy, a network slicing policy, where the network slicing policy is used to divide the network into at least one network slice, where the network slicing policy includes a mapping relationship between a service flow class and a network slice, and a slice weight of the network slice.
  • the slice weight represents a ratio of resources of the network slice to resources of the network;
  • the resource node policy includes a mapping relationship between at least one resource node in the network and the at least one network slice, and a resource weight of each resource node, the resource node
  • the resource weight represents a ratio of resources of each network slice in at least one network slice corresponding to the resource node to resources of the resource node.
  • the processor 930 may be a central processing unit (CPU), and the processor 930 may also be other general-purpose processors, digital signal processors (DSPs), and application specific integrated circuits. (ASIC), off-the-shelf programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and more.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the memory 940 can include read only memory and random access memory and provides instructions and data to the processor 930. A portion of the memory 940 can also include a non-volatile random access memory. For example, the memory 940 can also store information of the device type.
  • the bus system 950 may include a power bus, a control bus, a status signal bus, and the like in addition to the data bus. However, for clarity of description, various buses are labeled as bus system 950 in the figure.
  • each step of the foregoing method may be completed by an integrated logic circuit of hardware in the processor 930 or an instruction in a form of software.
  • the steps of the method for combining the network slice disclosed in the embodiment of the present application may be directly implemented by the hardware processor, or may be performed by a combination of hardware and software modules in the processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in memory 940, and processor 930 reads the information in memory 940 and, in conjunction with its hardware, performs the steps of the above method. To avoid repetition, it will not be described in detail here.
  • the respective units in the network slicing system 900 and the other operations or functions described above according to embodiments of the present application may implement respective processes performed by the application-aware analysis module, the slice decision coordination module, and the slice control module in the method 100. For the sake of brevity, it will not be repeated here.
  • the classification policy can be obtained based on the metadata information of the service flow, and the precise requirements of different types of service flows to the network can be obtained according to the classification criteria in the classification policy.
  • the network slice weight of the network slice can be determined based on the service flow to the network requirement, so that the network slice can meet the requirements of different types of services to the network while satisfying the efficient utilization of the network.
  • the control plane can dynamically adjust the network slicing policy according to the metadata information of the traffic flow collected in real time, so that the network slicing can meet the dynamic demand of different types of services to the network.
  • FIG. 10 shows a network node 1000 in accordance with an embodiment of the present application.
  • the network node 1000 includes a receiver 1010, a transmitter 1020, a processor 1030, a memory 1040, and a bus system 1050.
  • the receiver 1010, the transmitter 1020, the processor 1030, and the memory 1040 are connected by a bus system 1050 for storing instructions for executing the instructions stored by the memory 1040 to control the receiver 1010 to receive.
  • Signaling and controlling the transmitter 1020 to send a signal wherein
  • the receiver 1010 is configured to receive a service flow to be transmitted, where the service flow to be transmitted carries a network slice identifier, and the network slice identified by the network slice identifier and the service flow class to which the service flow to be transmitted belongs correspond.
  • the processor 1030 is configured to match the flow table according to the network slice identifier received by the receiving unit.
  • the transmitter 1020 is configured to send, according to the flow table and the resource weight matched by the matching unit, the service flow to be transmitted, where the resource weight is a bandwidth weight of the network slice at a corresponding physical port of the network node.
  • the processor 1030 may be a central processing unit (CPU), and the processor 1030 may also be other general-purpose processors, digital signal processors (DSPs), and application specific integrated circuits. (ASIC), off-the-shelf programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and more.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the memory 1040 can include read only memory and random access memory and provides instructions and data to the processor 1030. A portion of the memory 1040 may also include a non-volatile random access memory. For example, the memory 1040 can also store information of the device type.
  • the bus system 1050 may include a power bus, a control bus, a status signal bus, and the like in addition to the data bus. However, for clarity of description, various buses are labeled as bus system 1050 in the figure.
  • each step of the above method may be completed by an integrated logic circuit of hardware in the processor 1030 or an instruction in a form of software.
  • the steps of the method for transmitting a service flow disclosed in the embodiments of the present application may be directly implemented by the hardware processor, or may be performed by a combination of hardware and software modules in the processor.
  • the software module can be located in a random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, Registers and other mature storage media in the field.
  • the storage medium is located in the memory 1040, and the processor 1030 reads the information in the memory 1040 and performs the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.
  • the units in the signal network node 1000 according to the embodiment of the present application and the other operations or functions described above are respectively for the respective processes performed by the network node in the method shown in FIG. For the sake of brevity, it will not be repeated here.
  • the network node in the embodiment of the present application schedules the service flow by using the network slice to which the service flow belongs, so as to avoid mutual interference between the service flows of different service types, and can meet the performance requirements of service flows of different service types.
  • FIG. 11 shows a terminal device 1100 according to an embodiment of the present application.
  • the terminal device 1100 includes a receiver 1110, a transmitter 1120, a processor 1130, a memory 1140, and a bus system 1150.
  • the receiver 1110, the transmitter 1120, the processor 1130, and the memory 1140 are connected by a bus system 1150.
  • the memory 1140 is configured to store instructions for executing the instructions stored by the memory 1140 to control the receiver 1110 to receive. Signaling and controlling the transmitter 1120 to send a signal, wherein
  • the receiver 1110 is configured to receive a service flow to be transmitted.
  • the sender 1120 is configured to send a request message to the centralized controller, where the request message carries the quintuple information of the message in the service flow to be transmitted received by the receiver 1110, where the request message is used for requesting
  • the network slice information corresponding to the transmitted service flow is mentioned.
  • the receiver 1110 is further configured to receive network slice information that is sent by the centralized controller according to the quintuple information, where the network slice indicated by the network slice information corresponds to a service class to which the service flow to be transmitted belongs.
  • the transmitter 1120 is further configured to send, to the network node, a service flow to be transmitted, where the packet in the service flow to be transmitted carries a network slice identifier that identifies the network slice.
  • the processor 1130 may be a central processing unit (CPU), and the processor 1130 may also be other general-purpose processors, digital signal processors (DSPs), and application specific integrated circuits. (ASIC), off-the-shelf programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and more.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the memory 1140 can include read only memory and random access memory and provides instructions and data to the processor 1130. A portion of the memory 1140 can also include a non-volatile random access memory. For example, the memory 1140 can also store information of the device type.
  • the bus system 1150 may include a power bus, a control bus, a status signal bus, and the like in addition to the data bus. However, for clarity of description, various buses are labeled as bus system 1150 in the figure.
  • each step of the foregoing method may be completed by an integrated logic circuit of hardware in the processor 1130 or an instruction in a form of software.
  • the steps of the method for transmitting a service flow disclosed in the embodiments of the present application may be directly implemented by the hardware processor, or may be performed by a combination of hardware and software modules in the processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory 1140, and the processor 1130 reads the information in the memory 1140 and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.
  • the terminal device in the embodiment of the present application schedules the service flow by using the network slice to which the service flow belongs, and avoids mutual interference between the service flows of different service types, and can meet the performance requirements of service flows of different service types.
  • the size of the sequence numbers of the foregoing processes does not mean the order of execution sequence, and the order of execution of each process should be determined by its function and internal logic, and should not be applied to the embodiment of the present application.
  • the implementation process constitutes any limitation.
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product.
  • the technical solution of the present application which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .

Abstract

本申请公开了一种网络切片方法，能够使得网络切片满足不同类型的业务对网络的需求。该方法包括：获取多个业务流中的每个业务流的元数据信息，业务流的元数据信息用于指示业务流的流量特征；根据多个业务流的元数据信息确定分类策略；根据分类策略确定网络切片策略，网络切片策略用于将网络分为至少一个网络切片，网络切片策略包括业务流类与网络切片的映射关系，以及网络切片的切片权重；根据网络切片策略确定资源节点策略，资源节点策略包括网络中的至少一个资源节点与至少一个网络切片的映射关系，以及每个资源节点的资源权重，资源节点的资源权重表示与资源节点对应的网络切片中的每个网络切片的资源占资源节点的资源的比率。

Description

网络切片方法和系统
本申请要求于2016年07月25日提交中国专利局、申请号为201610589662.6、发明名称为“网络切片方法和系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及通信领域，更具体地，涉及一种网络切片方法、网络切片系统、网络节点和终端设备。
背景技术
传统网络承载的业务比较单一，如窄带网络承载语音业务，同步数字体系(synchronous digital hierarchy，SDH)网络承载无线回传业务。每张网络针对确定的业务，可以实现高效的管理和业务性能最大化的统一。随着传输控制协议/互联网络协议(transmission control protocol/internet protocol，TCP/IP)技术与互联网的发展，网络运营商希望降低网络的管理维护成本，希望将越来越多的业务，采用一张物理网络来承载。在一张物理网络上，如何保证各业务的性能，成为一个普遍关心的问题。当前业界普遍认可的是，未来网络应是多种虚拟的网络切片共存于一张物理网络上，这样可以达到业务的高效性能与低维护管理成本的最佳平衡。由于业务的种类较多，并且不同种类业务对网络的需求是变化的，所以，最关键的问题在于，如何将分布式的网络资源，像类似云计算中的虚拟中央处理器(virtual central process unit，vCPU)，虚拟存储器(virtual storage，vStorage)一样进行虚拟化分片管理。并把这些分片的网络资源以足够的弹性，匹配并提供给不同应用使用，同时保证各应用变化的网络需求和网络的高效利用。
长期以来，数据中心(data centre，DC)网络，WAN网络都采用差分服务(DiffServe)方式，并采用大带宽来保证业务质量。该方式之所以普及，是因为它方法简单，不需要复杂的控制。也正因为如此，在这种方式下的带宽利用率很低并且带宽分配不均衡，对于高突发业务、低时延业务等不能够很好的保证质量。
近些年，人们认识到集中粗放式网络管理越来越不能满足日益增长的网络各类需求。随着软件定义网络(software defined network，SDN)技术的产生，基于SDN思路的集中流量控制技术也有不少研究。该技术的主要思路就是通过收集网络流量流向，构建动态的流量矩阵，并根据流量矩阵来调整网络中的路径，使得网络达到最优的性能和利用率。
然而，在多业务共存的网络中，这种追求资源最优的方案难以兼顾不同业务类型的业务流的性能要求。
发明内容
本申请实施例提供了一种网络切片方法，能够使得网络切片满足不同类型的业务对网 络的需求。
第一方面，提供了一种网络切片方法，包括：获取多个业务流中的每个业务流的元数据信息，所述业务流的元数据信息用于指示所述业务流的流量特征；根据所述多个业务流的元数据信息确定分类策略，所述分类策略包括至少一个业务流类的分类标准，所述分类标准包括对所述业务流类的元数据的范围或取值描述；根据所述分类策略确定网络切片策略，所述网络切片策略用于将网络分为至少一个网络切片，所述网络切片策略包括业务流类与网络切片的映射关系，以及网络切片的切片权重，所述切片权重表示网络切片的资源占所述网络的资源的比率；根据所述网络切片策略确定资源节点策略，所述资源节点策略包括所述网络中的至少一个资源节点与所述至少一个网络切片的映射关系，以及每个资源节点的资源权重，所述资源节点的资源权重表示与所述资源节点对应的网络切片中的每个网络切片的资源占所述资源节点的资源的比率。
因此，本申请实施例的网络切片方法，基于业务流的元数据信息可以得到分类策略，根据分类策略中的分类标准可以得到不同类型的业务流对网络的精确需求。这样，可以基于业务流对网络需求确定网络切片的网络切片权重，从而在满足网络高效利用率的同时，能够使得网络切片满足不同类型的业务对网络的需求。另一方面，根据实时收集的业务流的元数据信息，可以动态调整网络切片策略，从而使得网络切片能够满足动态变化的不同类型的业务对网络的需求。
在一种可能的实现方式中，所述方法还包括：向所述资源节点发送配置信息，所述配置信息包括所述资源节点对应的所述至少一个网络切片的标识和所述资源节点的资源权重信息，所述资源权重信息用于指示所述资源节点的资源权重。
这样，在相同的物理基础设施上，通过管理编排，可以为每种业务分配虚拟的专有资源。在每种业务看来，为其分配到的资源是独享的，和其它业务之间是相互隔离的。通过多个业务共享相同的物理基础设施，提高了物理资源利用率，能够降低网络成本。
在一种可能的实现方式中，所述根据所述分类策略确定网络切片策略，包括：根据所述分类策略和业务支持系统(Business support system，BSS)的商业策略确定所述网络切片策略；或根据所述分类策略和运营支持系统(Operation Support System，OSS)的商业策略确定所述网络切片策略。
分片决策协同模块基于对某业务分片的商业策略，可以控制使该业务分片满足商业需求，带来商业价值。
在一种可能的实现方式中，所述资源节点包括分布式网络节点，所述网络切片的资源包括分布式网络节点的报文处理单元数量、报文发送队列个数与长度、及物理端口带宽大小；和/或所述资源节点包括分布式计算节点，所述网络切片的资源包括计算节点的虚拟中央处理器数量；和/或所述资源节点包括分布式存储节点，所述网络切片的资源包括存储节点的存储单元数量。
在一种可能的实现方式中，所述资源节点的资源权重包括所述资源节点对应的每个网络切片在所述资源节点的物理端口占用的带宽与所述物理端口的总带宽的比率。
在一种可能的实现方式中，所述业务流的元数据信息包括所述业务流的报文的五元组信息。
在一种可能的实现方式中，所述业务流的元数据信息包括以下信息中的至少一种：所 述业务流的建立时间、所述业务流中的报文的突发度、所述业务流中的报文的平均长度、所述业务流中的相邻报文之间的发送时间间隔、所述业务流的总长度，所述业务流的总持续时间、所述业务流的平均速率。
在一种可能的实现方式中，所述根据所述多个业务流的元数据信息确定分类策略，包括：根据所述多个业务流中的每个业务流的报文的平均长度、所述多个业务流中的每个业务流的相邻报文平均间隔、所述多个业务流中的每个业务流的总长度对所述多个业务流进行分类；根据分类的结果确定所述分类策略。
在一种可能的实现方式中，所述方法还包括：接收终端设备发送请求消息，所述请求消息携带待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息；根据所述五元组信息和所述网络切片策略确定所述网络切片信息；向所述终端设备发送响应消息，所述响应消息携带所述网络切片信息，以使所述终端设备根据所述网络切片信息传输所述待传输的业务流。
在一种可能的实现方式中，所述方法还包括：接收所述终端设备发送的报告消息，所述报告消息携带所述待传输的业务流的元数据信息。
第二方面，提供了一种传输业务流的方法，包括：接收待传输的业务流，所述待传输的业务流携带有网络切片标识，所述网络切片标识所标识的网络切片与所述待传输的业务流所归属的业务流类对应；根据所述网络切片标识匹配流表；根据匹配的流表和资源权重发送所述待传输的业务流，所述资源权重为所述网络切片在所述网络节点的对应物理端口的带宽权重。
本申请实施例的传输业务流的方法，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
在一种可能的实现方式中，所述匹配的流表包括执行字段，所述执行字段包括物理端口和与所述网络切片对应的队列标识，其中，根据匹配的流表和带宽权重发送所述待传输的业务流，包括：根据所述匹配的流表，将所述待传输的业务流发送至所述队列标识所标识的队列中，并在所述物理端口根据所述带宽权重对所述待传输的业务流进行调度。
在一种可能的实现方式中，在接收待传输的业务流之前，所述方法还包括：接收集中控制器发送的配置信息，所述配置信息携带网络切片信息和所述资源权重，所述网络切片信息用于指示所述网络切片。
第三方面，提供了一种传输业务流的方法，其特征在于，包括：接收待传输的业务流；向集中控制器发送请求消息，所述请求消息携带所述待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息；接收所述集中控制器根据所述五元组信息发送的网络切片信息，所述网络切片信息指示的网络切片与所述待传输的业务流所属的业务类对应；向网络节点发送所述待传输的业务流，所述待传输的业务流中的报文携带标识所述网络切片的网络切片标识。
本申请实施例的传输业务流的方法，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
在一种可能的实现方式中，所述方法还包括：向所述集中控制器发送报告消息，所述 报告消息携带所述待传输的业务流的元数据信息。
在一种可能的实现方式中，所述方法还包括：保存所述待传输的业务流与所述网络切片的对应关系。
第四方面，提供了一种网络切片系统，用于执行第一方面或第一方面的任意可能的实现方式中的方法。具体地，该网络切片系统包括用于执行第一方面或第一方面的任意可能的实现方式中的方法的单元。
第五方面，本申请提供一种网络节点，用于执行第二方面或第二方面的任意可能的实现方式中的方法。具体地，该网络节点包括用于执行第二方面或第二方面的任意可能的实现方式中的方法的单元。
第六方面，本申请提供一种终端设备，用于执行第三方面或第三方面的任意可能的实现方式中的方法。具体地，该终端设备包括用于执行第三方面或第三方面的任意可能的实现方式中的方法的单元。
第七方面，提供了一种网络切片系统，该切片系统包括接收器、发送器、处理器、存储器和总线系统。其中，接收器、发送器、处理器和存储器通过总线系统相连，存储器用于存储指令，处理器用于执行存储器存储的指令，以控制接收器接收信号和控制发送器发送信号。并且当处理器执行存储器存储的指令时，执行使得处理器执行第一方面或第一方面的任意可能的实现方式中的方法。
第八方面，本申请提供一种网络节点，该网络节点包括接收器、发送器、处理器、存储器和总线系统。其中，接收器、发送器、处理器和存储器通过总线系统相连，存储器用于存储指令，处理器用于执行存储器存储的指令，以控制接收器接收信号和控制发送器发送信号。并且当处理器执行存储器存储的指令时，执行使得处理器执行第二方面或第二方面的任意可能的实现方式中的方法。
第九方面，本申请提供一种终端设备，该终端设备包括接收器、发送器、处理器、存储器和总线系统。其中，接收器、发送器、处理器和存储器通过总线系统相连，存储器用于存储指令，处理器用于执行存储器存储的指令，以控制接收器接收信号和控制发送器发送信号。并且当处理器执行存储器存储的指令时，执行使得处理器执行第三方面或第三方面的任意可能的实现方式中的方法。
第十方面，本申请提供一种计算机可读介质，用于存储计算机程序，该计算机程序包括用于执行第一方面或第一方面的任意可能的实现方式中的方法的指令。
第十一方面，本申请提供一种计算机可读介质，用于存储计算机程序，该计算机程序包括用于执行第二方面或第二方面的任意可能的实现方式中的方法的指令。
第十二方面，本申请提供一种计算机可读介质，用于存储计算机程序，该计算机程序包括用于执行第三方面或第三方面的任意可能的实现方式中的方法的指令。
附图说明
图1是根据本申请实施例的系统架构的示意性框图。
图2是根据本申请实施例的网络切片的方法的示意性流程图。
图3是根据本申请实施例的另一网络架构的示意性框图。
图4是根据本申请实施例的传输业务流的方法的示意性流程图。
图5是根据本申请实施例的另一网络架构的示意性框图。
图6是根据本申请实施例的网络切片系统的示意性框图。
图7是根据本申请实施例的网络节点的示意性框图。
图8是根据本申请实施例的终端设备的示意性框图。
图9是根据本申请实施例的网络切片系统的示意性结构图。
图10是根据本申请实施例的网络节点的示意性结构图。
图11是根据本申请实施例的终端设备的示意性结构图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述。
首先，简要介绍与本申请相关的网络切片技术。
网络切片技术是将一个物理网络虚拟化划分为多个能同时共存的逻辑网络，为各类不同服务提供最合适的资源配置与网络拓扑。具体来说，就是基于一张物理网络对能力进行灵活组装和编排，通过端到端的网络切片，满足海量应用对网络差异化的需求。例如，工业控制切片可以提供超低时延来实现在救灾、电力检修等场景下对工业机器人的精准操控与交互；公用事业切片可以提供低功耗、超密连接来满足智能水表、气表、电表等公用业务需求。
由于业务的种类较多，并且不同种类业务对网络的需求是变化的，所以，如何实现适应多业务动态环境下的网络切片，如何协同各网络切片之间的资源成为一个问题。
为了解决上述问题，本申请实施例提供了一种网络切片的方法。
本申请实施例的网络切片的方法可以应用于图1所示的系统架构中。图1所示的系统的控制面，由应用(application，App)感知分析模块、分片决策协同模块和分片控制模块构成。控制面可以执行基于本申请实施例的网络切片方法。简单来讲，App感知分析模块可以获取数据面中的业务流的元数据信息，并根据元数据信息输出分类策略。这里的数据面包括端系统-发送设备、端系统-接收设备和位于端系统-发送设备与端系统-接收设备之间的基础设施网络中的资源节点。分片决策协同模块基于分类策略可以确定网络切片策略，并可以将网络切片策略输出至分片控制模块。分片控制模块基于网络切片策略可以确定资源节点策略。并向通过向网络节点发送配置信息，配置端到端的网络切片。最后，将网络分为网络切片1和网络切片2。网络切片1和网络切片2，分别依托于分布于基础网络中的多个网络节点、计算节点和存储节点中的服务于网络切片的转发、计算和存储资源。这里，可以称上述的网络节点、计算节点和存储节点为资源节点。网络切片1和网络切片2分别由利用这些资源构成的网络中的转发功能、处理功能、以及服务于这些转发功能与处理功能的控制功能组成。
另外，图1所示的BSS/OSS系统提供的BSS商业策略或OSS商业策略，可以作为控制面对网络进行切片时的一个参考信息。控制面在参考BSS商业策略或OSS商业策略的基础上，可以将网络分为网络切片1和网络切片2。商业策略包括针对某个商业用户的网络应用，需要指定个数的网络切片保证。针对某个商业用户或某类应用，规定其占用网络总体资源的数量，比率的最小值或最大值。在存在商业策略的前提下，分片决策协同模块 在做分片决策时，将保证决策输出的网络切片策略符合这些商业策略。
在数据面，网络切片1和网络切片2可以看作是有通信关系的端系统-发送设备到端系统-接收设备之间建立的两个虚拟网络。每个虚拟网络承载特定业务性能要求的业务流。在网络中每个目的端执行独立的拥塞检测，并将拥塞状态反馈到源端。数据面的基于分布式的拥塞控制称为快控制。关于快控制，可以参照较多现有技术，如网络显式拥塞通知(explicit congestion notification，ECN)技术配合TCP显式拥塞提醒回应(explicit congestion echo，ECE)反馈机制(RFC3168)技术，这里不再详细描述。
应理解，图1所述的系统架构中的App感知分析模块、分片决策协同模块和分片控制模块可以分别是三个实体设备，也可以是部署在一个实体上的三个功能模块。或者，App感知分析模块和分片决策协同模块可以集成在一个实体设备上，或者，分片决策协同模块和分片控制模块可以集成在一个实体设备上，本申请实施例对比不作限定。
图2是根据本申请一个实施例的网络切片方法的示意性流程如图。下面，结合图1和图2，对根据本申请实施例的网络切片方法进行详细描述。
S210，获取多个业务流中的每个业务流的元数据信息。元数据信息用于指示业务流的流量特征。
本申请实施例中涉及到的业务流，是指由一个五元组信息，即源网际协议(internet protocol，IP)地址、目的IP地址、协议号、源端口号、目的端口号界定的有序报文集合。
需要说明的是，这里所说的业务流可以是历史业务流，也可以是待传输的业务流。当这里的业务流为待传输的业务流时，可以获取业务流中的前几个报文的信息，从这前几个报文提取的业务流的元数据信息可以支持业务流的分类。
S210可以由图1所示的App感知分析模块执行。具体地，App感知分析模块可以从数据面中分布式收集网络中的多个业务流的元数据信息。这里，App感知分析模块可以周期性地收集数据面中的业务流的元数据信息。根据元数据信息，App感知分析模块可以分析业务流的流量特征，例如，业务流带宽特性、时延特性等。
元数据信息可以包括业务流的五元组信息。进一步地，元数据信息还可以包括：业务流的建立时间、业务流中的报文的平均长度、业务流中的报文的突发度、业务流中的相邻两个报文之间的发送时间间隔、业务流的总长度，业务流的总持续时间、业务流的平均速率等。
本申请实施例中涉及到的业务流的报文突发度，是指在队列调度时，同一个业务流中的报文连续到达队列的最大数量或连续报文的长度之和。
S220，根据业务流的元数据信息确定分类策略。
S220可以由App感知分析模块执行。具体地，App感知分析模块根据元数据信息，可以识别业务流的流量特征，并且可以将具有相近流量特征的流聚合为一类。例如，App感知分析模块可以通过业务流中的报文的目的地址和源地址的报文即可判断其业务类型，或者可以通过业务流中的报文中携带的公知端口号来推测应用类型，如文件传输协议(file transfer protocol，FTP)信令端口号TCP21，域名系统(domain name system，DNS)端口号UDP67，超文本传输协议(hypertext transfer protocol，HTTP)端口号TCP80等。
App感知分析模块将具有相近流量特征的流聚合为一类的过程，可以理解为对多个业务流进行分类的过程。在对多个业务流进行分类时，可以采用现有技术中的分类算法，例 如，可以采用机器学习中的聚类分析算法，但本申请实施例对比不作限定。
App感知分析模块通过对多个业务流进行分类，可以得到分类策略。分类策略可以包括业务流类的分类标准，即每类业务流的分类标准。一类业务流包括从多个元数据维度对业务流流量特征的描述，一个元数据维度体现为该类业务流在这个元数据的取值范围或特定取值。举例说明：报文平均长度是一种元数据，基于该维度来讲，可以说，报文平均长度大于200字节(bytes)，小于400bytes属于该类业务流。进一步地，分类策略还可以包括分类数量。分类标准体现了这一类业务流的流量特征，或者说这一类业务流的应用类型。作为示例而非限定，分类标准可以包括业务流的报文长度特征、业务流的持续时间特征、业务流的突发度特征等特征。进一步地，分类标准还可以包括业务流的分布数据特征，如均值，方差，局部峰值等信息。
举例来说，在对多个业务流进行分类后，可以得到多个业务流类，这多个业务流类可以包括F类业务流。应理解，F类业务流仅是一个代称，F类业务流也可以称为A类业务流，也可以称为B类业务流。这里，仅仅是为了区别不同类别的业务流。
这里，F类业务流的分类标准举例说明如下：
400字节>业务流的平均报文长度>200字节
300ms>业务流的持续时间>50ms
业务流的突发度<3000字节
另外，F类业务流的流量特征信息还可以包括以下内容：
应用类型：MapReduce
业务流流数量：340个
业务流流平均带宽:150千位每秒(kbps)
业务流带宽分布方差：50kbps
上述F类业务流的分类标准仅是示例，本申请实施例不限于此，其它的标准业务量的流量特征的分类标准都落入本申请的范围。
可选地，根据多个业务流的元数据信息确定分类策略时，可以根据多个业务流中的每个业务流的报文的平均长度、多个业务流中的每个业务流的相邻报文平均间隔、多个业务流中的每个业务流的总长度对多个业务流进行分类；然后根据分类的结果确定分类策略。
具体地，App感知分析模块可以基于业务流中的报文的平均长度、相邻报文平均间隔、业务流的总长度对多个业务流进行分类。这样，得到的每类业务流的分类标准可以包括业务流的报文的平均长度特征、相邻报文平均间隔特征、业务流的总长度特征。
S230，根据分类策略确定网络切片策略。
S230可以由图1所示的分片决策协同模块执行。具体地，分片决策协同模块可以接收App感知分析模块发送的分类策略，根据每类业务流的分类标准可以确定网络切片策略。网络切片策略可以包括业务流所属的业务流类与网络切片的对应关系，以及每个网络切片的切片权重。网络切片的切片权重表示网络切片的资源占网络的资源的比率。
这里，网络的资源指的是网络总体资源。网络总体资源可以包括分布式网络节点资源，分布式计算资源和分布式存储资源。相对应地，网络切片的资源可以包括网络节点资源，分布式计算资源和分布式存储资源。其中，分布式网络节点资源可以包括网络节点的报文处理单元数量、报文发送队列个数与长度、及物理端口带宽大小。分布式计算资源可以计 算节点的虚拟中央处理器数量。分布式存储资源可以包括存储节点的存储单元数量。
在本申请实施例中，可以根据分类数量和每类业务流的分类标准确定网络的切片数量，即将网络分为几个网络切片，每一个网络切片对应哪类业务流(即，上文中的业务流类)或者说哪类应用，以及每个网络切片的切片权重。这里，一个网络切片可以对应一类业务流。当有多类业务流的分类标准相近，或者网络的资源有限时，一个网络切片也可以对应多类业务流。在确定切片权重时，可以根据业务流类的分类标准或者业务流类对应的应用类型确定。例如，当根据业务流类的分类标准确定某一类业务流属于大带宽，但突发度不高的业务时(判定方法举例如：业务流的平均带宽大于2M，突发小于5kbytes，平均报文长度大于500bytes)，则可以按峰值带宽的80％带宽分配网络资源；当确定另一类业务流属于低时延，大突发业务时(判定方法举例如：业务流的平均带宽小于100K，突发大于2Kbytes,平均报文长度小于300bytes,流长度小于2kbytes)，可以按平均带宽的3倍带宽分配网络资源。
可选地，在根据分类策略确定网络切片策略时，可以根据分类策略和业务支持系统BSS的商业策略确定；还可以根据分类策略和运营支持系统OSS的商业策略确定；或者，根据分类策略、BSS的商业策略、以及OSS的商业策略确定。进一步地，还可以根据网络的可用的资源情况，来确定网络切片策略。
这里，分片决策协同模块基于对某业务分片的商业策略，可以控制使该业务分片满足商业需求，带来商业价值。
S240，根据网络切片策略确定资源节点策略。
资源节点策略可以包括资源节点与网络切片的映射关系，以及资源权重。资源节点的资源权重表示与资源节点对应的至少一个网络切片中的每个网络切片的资源占资源节点的资源的比率。
这里，资源节点可以包括分布式网络节点、分布式计算节点和分布式存储节点。
S240可以由图1所示的分片控制模块执行。具体地，在分片决策协同模块确定网络切片策略后，可以向分片控制模块发送网络切片策略。分片控制模块基于网络切片策略确定资源节点策略，完成切片权重到各分布式网络资源、分布式计算资源和分布式存储资源的分解。
具体来说，分片控制模块根据每个网络切片的切片权重后，还需要可以确定与特定网络切片对应的网络节点资源节点，即确定特定网络切片具体落在哪个或哪几个网络节点资源节点上。并且，分片控制模块还需要可以确定资源权重，即这个特定网络切片的资源占这个网络节点资源节点上的资源的比率，或者分别占这几个网络节点资源节点中的每个网络节点资源节点上的资源的比率。应理解，这里所说的特定网络切片网络可以是指的是多个网络切片中的任一网络切片。
这样，在相同的物理基础设施上，通过管理编排，可以为每种业务分配虚拟的专有资源。在每种业务看来，为其分配到的资源是独享的，和其它业务之间是相互隔离的。通过多个业务共享相同的物理基础设施，提高了物理资源利用率，能够降低网络成本。
因此，本申请实施例的网络切片方法，基于业务流的元数据信息可以得到分类策略，根据分类策略中的分类标准可以得到不同类型的业务流对网络的精确需求。这样，可以基于业务流对网络需求确定网络切片的网络切片权重，从而在满足网络高效利用率的同时， 能够使得网络切片满足不同类型的业务对网络的需求。另一方面，根据实时收集的业务流的元数据信息，可以动态调整网络切片策略，从而使得网络切片能够满足动态变化的不同类型的业务对网络的需求。
可选地，资源节点的资源权重包括资源节点对应的每个网络切片在资源节点的物理端口占用的带宽与物理端口的总带宽的比率。具体地，多个网络切片可落到同一网络节点上。这多个网络切片占用相应比率的该网络节点的物理端口的带宽。
可选地，该方法还可以包括：S250，向资源节点发送配置信息。
配置信息可以包括与资源节点对应的至少一个网络切片中的每个网络切片的标识、以及资源权重信息，资源权重信息用于指示资源节点的资源权重。
S250可以由图1所示的分片控制模块执行。这里，分片控制模块完成切片权重到各分布式网络节点资源、分布式计算资源和分布式存储资源的分解后，向对应资源所在的基础设施节点，即上文中所述的资源节点发送配置信息，进行资源配置。
具体地，分片控制模块在确定资源节点策略后需要对各资源节点进行配置。通过下发配置信息，资源节点可以根据配置信息可以进行资源配置。具体地，资源节点根据通过该配置信息中的网络切片标识，可以确定与其对应的网络切片，并且根据该配置信息中的资源权重，可以为该网络切片配置相应比率资源，例如，资源节点为网络切片配置相应比率的带宽资源。这样，便实现了物理网络到逻辑网络的映射。
分片控制模块可以控制调整部分网络节点的端口的带宽权重，达到整个网络的网络资源动态协同，从而在满足各业务的性能需求时，同时满足高的带宽利用率。
可选地，该方法还可以包括：接收终端设备发送请求消息，请求消息携带待传输的业务流中的报文的五元组信息，请求消息用于请求与待传输的业务流对应的网络切片信息；根据五元组信息和网络切片策略确定待传输的业务流对应的网络切片信息；向终端设备发送响应消息，响应消息携带网络切片信息，以使终端设备根据网络切片信息传输待传输的业务流。
这里，终端设备可以是网络边缘设备，具体地，可以是网络边缘的终端设备或服务器。例如，终端设备可以是图1所示的端系统-发送设备。
待传输的业务流可以根据通过方法100确定的网络切片策略进行传输。具体地，在分片决策协同模块已经确定网络切片策略后，分片控制模块可以根据接收到的待传输的业务流的五元组信息，确定该待传输的业务流所归属的业务流类。进一步地，通过网络切片策略中的业务流类与网络切片的映射关系，可以确定与该待传输的业务流对应的网络切片。分片控制模块在确定与该待传输的业务流对应的网络切片后，可以向终端设备发送网络切片信息，以使终端设备根据网络切片信息传输待传输的业务流。
可选地，该方法还可以包括：接收终端设备发送的报告消息，报告消息携带该待传输的业务流的元数据信息。
具体地，在终端设备发送业务流的过程中，可以向App感知分析模块发送该待传输的业务流的元数据信息。这样，App感知分析模块可以通过分析业务流的流量特征，动态确定分类策略。进一步地，分片决策协同模块可以根据动态的分类策略，确定网络切片策略，以使网络切片满足态变化的不同类型的业务。
图3是根据本申请实施例的另一系统架构示意图。本申请实施例的网络切片方法可以 应用于图3所示的系统中。
如图3所示，控制面由分析器，协同器和控制器组成，可以分别实现图1所示的APP感知分析模块、分片决策协同模块及分片控制模块的功能。分析器，协同器和控制器之间采用数据库进行耦合。下面结合图3，阐述根据本申请实施例的网络切片方法。应理解，下文中结合图3所描述的内容中出现的相关术语或词语等，均可以参照上文中的描述。为了简洁，下文中不再详细阐述。
具体来说，各边缘终端设备报告业务流元数据信息到分析器。在本申请实施例中，即终端设备1和终端设备2报告业务流的元数据信息到分析器。分析器汇集终端设备1和终端设备2的报告，对业务流进行分类，得到分类策略，并将分类策略发送至数据库。协同器读取数据库中的分类策略，可以确定网络切片策略。进一步地，协同器可以同时结合分类策略、BSS商业策略或OSS商业策略，确定网络切片策略。协同器将网络切片策略发送到数据库，控制器可以从数据库中读取网络切片策略，根据网络切片策略确定资源节点策略，然后向对应的网络节点发送配置信息。网络节点在接收到配置信息后，可以根据配置信息，对相应的网络切片配置对应的资源。
本申请实施例中，采用数据库实现各实体之间的接口，在实体A(例如，分析器)发送信息I(例如，元数据信息)到数据库，实体B(例如，协同器)从数据库中读取信息I的过程，等同于实体A发送信息I到实体B的过程。在本申请实施例中，控制面上的各实体设备之间可以通过数据库进行交互。例如，分析器和协同器可以通过数据库进行交互。控制面上的实体设备和数据面上的终端设备也可以通过数据库进行交互。例如，分析器和终端设备1可以通过数据库进行交互。另外，本申请实施例中，网络设备、计算设备或存储设备之间也可以通过数据库进行交互，本申请实施例对此不作限定。
因此，本申请实施例的网络切片方法，基于业务流的元数据信息可以得到分类策略，根据分类策略中的分类标准可以得到不同类型的业务流对网络的精确需求。这样，可以基于业务流对网络需求确定网络切片的网络切片权重，从而在满足网络高效利用率的同时，能够使得网络切片满足不同类型的业务对网络的需求。另一方面，根据实时收集的业务流的元数据信息，可以动态调整网络切片策略，从而使得网络切片能够满足动态变化的不同类型的业务对网络的需求。
图4是根据本申请实施例的一种传输业务流的方法。应理解，下文中结合图4所描述的内容中出现的相关术语或词语等，均可以参照上文中的描述。为了简洁，下文中不再详细阐述。
402，终端设备接收待传输的业务流。
这里，待传输的业务流可以是一个，也可以是多个。
应理解，终端设备为网络边缘设备。例如，终端设备可以是图1所示的端系统-发送设备，还可以是图3所示的终端设备1。
403，终端设备向集中控制器发送请求消息。
该请求消息携带待传输的业务流中的报文的五元组信息。请求消息用于请求与待传输的业务流对应的网络切片信息。
应理解，这里的集中控制器仅是一个代称，集中控制器还可以称为集中管理器、切片控制器等。集中控制器可以理解为集成了图1所示的App感知分析模块、分片决策协调模 块和分片控制功能模块三个功能模块的一个实体设备，集中控制器也可以看成是集成了图3所示的分析器、协同器和控制器三个实体装置的功能的一个实体设备。
还应理解，终端设备向集中控制器发送该待传输的业务流中的报文的五元组信息，在具体实现时，可以是终端设备向图1所示的分片控制模块，或图3所示的控制器发送该待传输的业务流中的报文的五元组信息。
405，集中控制器根据五元组信息和网络切片策略可以确定与待传输的业务流对应的网络切片信息。
具体地，集中控制器存储有网络切片策略，网络切片策略保存有业务流类与网络切片的对应关系。集中控制器接收到待传输的业务流的五元组信息后，通过分析待传输的业务流的流量特征，可以确定待传输的业务流所属的业务流类。然后，根据业务流类与网络切片的对应关系，可以确定与待传输的业务流对应的网络切片信息。进一步地，集中控制器可以确定与待传输的业务流对应的网络切片标识。
407，集中控制器向终端设备发送响应消息，响应消息用于指示该网络切片信息。
具体地，集中控制器在确定与待传输的业务流对应的网络切片后，向终端设备发送响应消息，以告知终端设备与待传输的业务流对应的网络切片。作为一个实施例，响应消息中可以携带该网络切片的切片标识。
409，终端设备向网络节点发送待传输的业务流。该待传输的业务流的报文中可以携带该网络切片标识。
应理解，这里的网络节点可以是上文中的分布式网络节点。
作为一个实施例，终端设备可以保存待传输的业务流与该网络切片的对应关系。进一步地，终端设备可以保存待传输的业务流与该网络切片标识的对应关系。那么，在终端设备接收到该业务流的后续报文时，则基于所保存的对应关系，可以直接查到网络切片标识，并转发携带有网络切片标识的业务流报文。这样，可以在节省开销的同时，能够提高转发效率。
另一种方式，集中控制器可以将对应一个切片的业务流识别标准，即对应一个或多个上述业务流分类标准，发送到终端设备。这样终端设备可以自行完成业务流的分类与对报文增加网络切片标识。
411，网络节点根据网络切片标识匹配流表。
在本申请实施例中，网络节点可以预先配置带有网络切片标识作为匹配项的模糊流表。待传输的业务流到来后，网络节点可以基于模匹配产生带有网络切片标识作为匹配项的精确流表。另一方面，也可以采用SDN的方法，网络节点将待传输的业务流的前几个报文发送到控制器，由控制器下发带有网络切片标识作为匹配项的精确流表。
413，网络节点根据匹配的流表和资源权重发送业务流，资源权重为网络切片在网络节点的带宽权重。
可选地，在402之前，该方法还可以包括401：网络节点接收集中控制器发送的配置信息，所述配置信息携带所述网络节点的标识和所述资源权重。
具体地，网络节点可以通过向集中控制器查询的方式，获得与该网络切片对应的资源权重。也可以通过集中控制器主动下发的配置信息获得资源权重。在这里，网络节点根据匹配的流表和资源权重执行待传输的业务流的转发。
可选地，匹配的流表包括执行字段，执行字段可以包括物理端口和与网络切片对应的队列标识，其中，网络节点根据匹配的流表和带宽权重发送待传输的业务流，包括：网络节点根据匹配的流表，将待传输的业务流发送至队列标识所标识的队列中，并在物理端口根据带宽权重对待传输的业务流进行调度。
作为一例，匹配的流表表示如下：<match:flow2&tagC,action:output to port:queue NE3:C>。这里，匹配match字段用于网络节点对接收到的待传输的业务流flow2进行匹配，其中，tagC为该网络切片，网络切片C的标识。执行action字段用于指示网络节点需要对flow2进行的处理。在这里，表示网络节点需要将待传输的业务流发送到连接网络节点NE(Network Element，网元)3的物理端口的对应网络切片C的队列中。应理解，一个网络节点可以对应多个网络切片，多个网络切片可以对应于同一物理端口的多个队列中。例如，一个网络切片可以对应于一个队列，本申请实施例对此不作限定。
网络节点在通向NE3的物理端口发送处，可以采用加权公平队列(weighted fair queue，WFQ)算法对待传输的业务流进行调度。WFQ采用的队列数，可以是网络切片数；WFQ和队列的调度权重，可以是各网络切片的切片权重对应到该网络物理端口的带宽权重。在实现对待传输的业务流的调度时，考虑到平衡队列调度的成本，也可以用加权循环调度算法(weighted round robin，WRR)近似实现。
415，终端设备向集中控制器发送报告消息。报告消息用于报告与待传输的业务流对应的元数据信息。
具体来说，终端设备传输业务流的过程中，可以向集中控制器发送该待传输的业务流的五元组信息。这样，集中控制器可以通过分析业务流的流量特征，动态确定分类策略。进一步地，集中控制器可以根据动态的分类策略，确定网络切片策略，以使网络切片满足态变化的不同类型的业务。
本申请实施例的传输业务流的方法，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
下面结合图5，对根据本申请实施例的传输业务流的方法进行详细描述。
图5所示的集中控制器510可以用于执行图4所示的方法中的由集中控制器执行的相应步骤。NE530和NE540可以对应于前文所述的网络节点。下面结合图5，详细描述了针对终端设备520到终端设备540的通信处理过程，及数据面与控制面的交互流程。
(A1)：当终端设备520需要发送一条新的业务流(或者称，待传输的业务流)flow1(流1)到终端终端设备520时，终端设备520首先可以向集中控制器510发送请求消息，请求消息可以携带flow1中的报文的五元组信息，用于向集中控制器510请求flow1对应的网络切片信息。
(A2)：集中控制器510基于之前已生成的网络切片策略中的业务流类与网络切片的映射关系，可以确定flow1对应网络切片B。集中控制器510将flow1对应网络切片B的这个信息，通过响应消息发送到终端设备520。
(A3)：终端1可以保存flow1对应网络切片B的这个对应关系，同时，给flow1中的报文增加一个标签tagB，指示flow1与网络切片B对应。然后终端设备520将带有标签的flow1发送到网络中，即图5所示的网络设备NE530。
(A4)：NE530通过匹配流表，可以得到匹配的流表<match:flow1&tagB,action:output to port:queue NE540:B>。根据该匹配的流表，NE530将flow1发送到连接网络设备NE540的物理端口的对应网络切片B的队列中。这里，流表匹配方法以及对报文的调度可以参照上文中的描述，为了简洁，此处不再赘述。
(A5)：NE530基于调度发送flow1到下一个网络节点NE540，NE540执行类似NE530的匹配流表的操作及对flow1的调度。最后，将flow1发送至终端设备540。终端设备540接收到flow1后，flow1在网络中的传送结束。
在上述过程中存在如下的慢循环控制：
(B1)：终端设备520将流flow1对应报文的元数据信息，送到集中控制器510。这里的元数据信息可以包括flow1的建立时间，flow1中的报文的五元组，报文长度，报文的当前最大突发度，报文与上一个报文的接收时间间隔，flow1的当前总长度，flow1的当前总持续时间等。
NE530，NE540可以将物理端口对应各网络切片的占用带宽，对应各网络切片的队列长度信息发送到集中控制器510。
终端设备550将流flow1对应的接收flow1中的报文时间，报文当前最大突发度，报文与上一个报文的接收时间间隔，flow1的当前总持续时间等元数据信息，发送到集中控制器510。
集中控制器510接收多个业务流(其中，包括flow1)的元数据信息，进行分类计算，最后生成网络切片策略。具体的分类方法，可以参照前文中的描述，这里不再详细介绍。基于网络切片策略，集中控制器510进一步确定资源节点策略，具体可以参见前文介绍。集中控制器510将根据资源节点策略确定的配置信息下发到数据面中对应的网络转发，计算与存储基础设施上实施网络切片隔离保证。
以上过程中，存在快控制循环为如下步骤：
(C1)：终端设备550统计flow1对应的网络切片B在终端设备520到终端设备550的逻辑管道上的拥塞程度信息。具体统计方式，可借助NE530，NE540在出口队列调度时，基于队列长度在报文中携带的对应的拥塞程度信息来得到。
(C2)：终端设备550将拥塞程度信息通过报文携带，或通过专用的消息报文，将拥塞程度信息发送到终端设备520.
(C3)：终端设备520基于拥塞程度信息，调整当前对应于网络切片B的发送速率。具体地，当反馈为不拥塞时，按网络切片B的带宽权重增大发送速率；当反馈为拥塞时，按网络切片B的带宽权重及拥塞程度，减小发送速率。
以上的快控制循环可以参照现有技术，这里不再详述。
本申请实施例的传输业务流的方法，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
上文中结合图1至图5，详细描述了根据本申请实施例的网络切片方法和传输业务流的方法，以下，将结合图6，介绍根据本申请实施例的网络切片系统。
图6示出了根据本申请实施例的网络切片系统600。所述网络切片系统600包括应用感知分析单元610、分片决策协同单元620和分片控制单元630。
应用感知分析单元610，用于获取多个业务流中的每个业务流的元数据信息，所述业务流的元数据信息用于指示所述业务流的流量特征；
根据所述多个业务流的元数据信息确定分类策略，所述分类策略包括至少一个业务流类的分类标准，所述分类标准包括对所述业务流类的元数据的范围或取值描述。
分片决策协同单元620接收应用感知分析单元610发送的分类策略，用于根据分类策略确定网络切片策略，网络切片策略用于将网络分为至少一个网络切片，所述网络切片策略包括业务流类与网络切片的映射关系，以及网络切片的切片权重，所述切片权重表示网络切片的资源占所述网络的资源的比率。
分片控制单元630接收分片决策协同单元620发送的网络切片策略，用于根据网络切片策略确定资源节点策略，资源节点策略包括所述网络中的至少一个资源节点与所述至少一个网络切片的映射关系，以及每个资源节点的资源权重，所述资源节点的资源权重表示与所述资源节点对应的至少一个网络切片中的每个网络切片的资源占所述资源节点的资源的比率。
应理解，根据本申请实施例的网络切片系统600中的应用感知分析单元610可以实现方法100中可以由应用感知分析模块执行的相应流程；分片决策协同单元620可以实现实现方法100中可以由分片决策协同模块执行的相应流程；分片控制单元630可以实现实现方法100中可以由分片控制单模块执行的相应流程。并且，网络切片系统600中的各单元还可以实现在方法100中可以由对应单元实现的其它操作或功能。为了简洁，此处不再赘述。
因此，在本申请实施例中，网络切片系统基于业务流的元数据信息可以得到分类策略，根据分类策略中的分类标准可以得到不同类型的业务流对网络的精确需求。这样，可以基于业务流对网络需求确定网络切片的网络切片权重，从而在满足网络高效利用率的同时，能够使得网络切片满足不同类型的业务对网络的需求。另一方面，网络切片系统可以根据实时收集的业务流的元数据信息，动态调整网络切片策略，从而使得网络切片能够满足动态变化的不同类型的业务对网络的需求。
图7示出了根据本申请实施例的网络节点700。所述网络节点700包括第一接收单元710、匹配单元720和发送单元730。
第一接收单元710，用于接收待传输的业务流，所述待传输的业务流携带有网络切片标识，所述网络切片标识所标识的网络切片与所述待传输的业务流所归属的业务流类对应。
匹配单元720，用于根据所述第一接收单元接收的所述网络切片标识匹配流表。
发送单元730，用于根据所述匹配单元匹配的流表和资源权重发送所述待传输的业务流，所述资源权重为所述网络切片在所述网络节点的对应物理端口的带宽权重。
根据本申请实施例的信号网络节点700中的各单元和上述其它操作或功能分别为了图4所示的方法中由网络节点执行的相应流程。为了简洁，此处不再赘述。
本申请实施例的网络节点，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
图8示出了根据本申请实施例的终端设备800。终端设备800包括第一接收单元810、第一发送单元820、第二接收单元830和第二发送单元840。
第一接收单元810，用于接收待传输的业务流。
第一发送单元820，用于向集中控制器发送请求消息，所述请求消息携带第一接收单元810接收的所述待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息。
第一接收单元810接收的待传输的业务流中的报文的五元组信息。
第二接收单元830，用于接收集中控制器根据第一发送单元820发送的五元组信息发送的网络切片信息，网络切片信息指示的网络切片与待传输的业务流所属的业务类对应。
第二发送单元840，,用于向网络节点发送待传输的业务流，待传输的业务流中的报文携带标识所述网络切片的网络切片标识。
根据本申请实施例的信号终端设备800中的各单元和上述其它操作或功能分别为了图4所示的方法中由终端设备执行的相应流程。为了简洁，此处不再赘述。
本申请实施例的终端设备，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
图9示出了根据本申请另一实施例的网络切片系统900。所述网络切片系统900包括接收器910、发送器920、处理器930、存储器940和总线系统950。其中，接收器910、发送器920、处理器930和存储器940通过总线系统950相连，该存储器940用于存储指令，该处理器930用于执行该存储器940存储的指令，以控制接收器910接收信号，并控制发送器920发送信号，其中，处理器930用于，
获取多个业务流中的每个业务流的元数据信息，所述业务流的元数据信息用于指示所述业务流的流量特征；
根据所述多个业务流的元数据信息确定分类策略，所述分类策略包括至少一个业务流类的分类标准，所述业务流类包括至少一个所述业务流；
根据所述分类策略确定网络切片策略，所述网络切片策略用于将网络分为至少一个网络切片，所述网络切片策略包括业务流类与网络切片的映射关系，以及网络切片的切片权重，所述切片权重表示网络切片的资源占所述网络的资源的比率；
根据所述网络切片策略确定资源节点策略，所述资源节点策略包括所述网络中的至少一个资源节点与所述至少一个网络切片的映射关系，以及每个资源节点的资源权重，所述资源节点的资源权重表示与所述资源节点对应的至少一个网络切片中的每个网络切片的资源占所述资源节点的资源的比率。
应理解，在本申请实施例中，该处理器930可以是中央处理单元(central processing unit，CPU)，该处理器930还可以是其他通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现成可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
该存储器940可以包括只读存储器和随机存取存储器，并向处理器930提供指令和数据。存储器940的一部分还可以包括非易失性随机存取存储器。例如，存储器940还可以存储设备类型的信息。
该总线系统950除包括数据总线之外，还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见，在图中将各种总线都标为总线系统950。
在实现过程中，上述方法的各步骤可以通过处理器930中的硬件的集成逻辑电路或者软件形式的指令完成。结合本申请实施例所公开的网络切片的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器940，处理器930读取存储器940中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
根据本申请实施例的网络切片系统900中的各单元和上述其它操作或功能可以实现方法100中由应用感知分析模块、分片决策协同模块和分片控制模块执行的相应流程。为了简洁，此处不再赘述。
因此，在本申请实施例的网络切片系统，基于业务流的元数据信息可以得到分类策略，根据分类策略中的分类标准可以得到不同类型的业务流对网络的精确需求。这样，可以基于业务流对网络需求确定网络切片的网络切片权重，从而在满足网络高效利用率的同时，能够使得网络切片满足不同类型的业务对网络的需求。另一方面，控制面可以根据实时收集的业务流的元数据信息，动态调整网络切片策略，从而使得网络切片能够满足动态变化的不同类型的业务对网络的需求。
图10示出了根据本申请实施例的网络节点1000。所述网络节点1000包括接收器1010、发送器1020、处理器1030、存储器1040和总线系统1050。其中，接收器1010、发送器1020、处理器1030和存储器1040通过总线系统1050相连，该存储器1040用于存储指令，该处理器1030用于执行该存储器1040存储的指令，以控制接收器1010接收信号，并控制发送器1020发送信号，其中，
接收器1010，用于接收待传输的业务流；所述待传输的业务流携带有网络切片标识，所述网络切片标识所标识的网络切片与所述待传输的业务流所归属的业务流类对应。
处理器1030，用于根据所述接收单元接收的所述网络切片标识匹配流表。
发送器1020，用于根据所述匹配单元匹配的流表和资源权重发送所述待传输的业务流，所述资源权重为所述网络切片在所述网络节点的对应物理端口的带宽权重。
应理解，在本申请实施例中，该处理器1030可以是中央处理单元(central processing unit，CPU)，该处理器1030还可以是其他通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现成可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
该存储器1040可以包括只读存储器和随机存取存储器，并向处理器1030提供指令和数据。存储器1040的一部分还可以包括非易失性随机存取存储器。例如，存储器1040还可以存储设备类型的信息。
该总线系统1050除包括数据总线之外，还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见，在图中将各种总线都标为总线系统1050。
在实现过程中，上述方法的各步骤可以通过处理器1030中的硬件的集成逻辑电路或者软件形式的指令完成。结合本申请实施例所公开的传输业务流的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、 寄存器等本领域成熟的存储介质中。该存储介质位于存储器1040，处理器1030读取存储器1040中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
根据本申请实施例的信号网络节点1000中的各单元和上述其它操作或功能分别为了图4所示的方法中由网络节点执行的相应流程。为了简洁，此处不再赘述。
本申请实施例的网络节点，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
图11示出了根据本申请实施例的终端设备1100。所述终端设备1100包括接收器1110、发送器1120、处理器1130、存储器1140和总线系统1150。其中，接收器1110、发送器1120、处理器1130和存储器1140通过总线系统1150相连，该存储器1140用于存储指令，该处理器1130用于执行该存储器1140存储的指令，以控制接收器1110接收信号，并控制发送器1120发送信号，其中，
接收器1110，用于接收待传输的业务流。
发送器1120，用于向集中控制器发送请求消息，所述请求消息携带接收器1110接收的所述待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息。
接收器1110还用于，接收集中控制器根据五元组信息发送的网络切片信息，网络切片信息指示的网络切片与待传输的业务流所属的业务类对应。
发送器1120还用于，向网络节点发送待传输的业务流，待传输的业务流中的报文携带标识所述网络切片的网络切片标识。
应理解，在本申请实施例中，该处理器1130可以是中央处理单元(central processing unit，CPU)，该处理器1130还可以是其他通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现成可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
该存储器1140可以包括只读存储器和随机存取存储器，并向处理器1130提供指令和数据。存储器1140的一部分还可以包括非易失性随机存取存储器。例如，存储器1140还可以存储设备类型的信息。
该总线系统1150除包括数据总线之外，还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见，在图中将各种总线都标为总线系统1150。
在实现过程中，上述方法的各步骤可以通过处理器1130中的硬件的集成逻辑电路或者软件形式的指令完成。结合本申请实施例所公开的传输业务流的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器1140，处理器1130读取存储器1140中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
根据本申请实施例的信号终端设备1100中的各单元和上述其它操作或功能分别为了图4所示的方法中由终端设备执行的相应流程。为了简洁，此处不再赘述。
本申请实施例的终端设备，通过业务流所归属的网络切片对业务流进行调度，避免不同业务类型的业务流之间产生的相互干扰，能够满足不同业务类型的业务流的性能要求。
应理解，在本申请的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (31)

  1. 一种网络切片方法，其特征在于，包括：
    获取多个业务流中的每个业务流的元数据信息，所述业务流的元数据信息用于指示所述业务流的流量特征；
    根据所述多个业务流的元数据信息确定分类策略，所述分类策略包括至少一个业务流类的分类标准，所述分类标准包括对所述业务流类的元数据的范围或取值描述；
    根据所述分类策略确定网络切片策略，所述网络切片策略用于将网络分为至少一个网络切片，所述网络切片策略包括业务流类与网络切片的映射关系，以及网络切片的切片权重，所述切片权重表示网络切片的资源占所述网络的资源的比率；
    根据所述网络切片策略确定资源节点策略，所述资源节点策略包括所述网络中的至少一个资源节点与所述至少一个网络切片的映射关系，以及每个资源节点的资源权重，所述资源节点的资源权重表示与所述资源节点对应的网络切片中的每个网络切片的资源占所述资源节点的资源的比率。
  2. 如权利要求1所述的方法，其特征在于，所述方法还包括：
    向所述资源节点发送配置信息，所述配置信息包括所述资源节点对应的所述至少一个网络切片的标识和所述资源节点的资源权重信息，所述资源权重信息用于指示所述资源节点的资源权重。
  3. 如权利要求1或2所述的方法，其特征在于，所述根据所述分类策略确定网络切片策略，包括：
    根据所述分类策略和业务支持系统BSS的商业策略确定所述网络切片策略；或
    根据所述分类策略和运营支持系统OSS的商业策略确定所述网络切片策略。
  4. 如权利要求1-3中任一项所述的方法，其特征在于，所述资源节点包括分布式网络节点，所述网络切片的资源包括分布式网络节点的报文处理单元数量、报文发送队列个数与长度、及物理端口带宽大小；和/或
    所述资源节点包括分布式计算节点，所述网络切片的资源包括计算节点的虚拟中央处理器数量；和/或
    所述资源节点包括分布式存储节点，所述网络切片的资源包括存储节点的存储单元数量。
  5. 如权利要求1-4中任一项所述的方法，其特征在于，所述资源节点的资源权重包括所述资源节点对应的每个网络切片在所述资源节点的物理端口占用的带宽与所述物理端口的总带宽的比率。
  6. 如权利要求1-5中任一项所述的方法，其特征在于，所述业务流的元数据信息包括所述业务流的报文的五元组信息。
  7. 如权利要求1-6中任一项所述的方法，其特征在于，所述业务流的元数据信息包括以下信息中的至少一种：
    所述业务流的建立时间、所述业务流中的报文的突发度、所述业务流中的报文的平均长度、所述业务流中的相邻报文之间的发送时间间隔、所述业务流的总长度，所述业务流 的总持续时间、所述业务流的平均速率。
  8. 如权利要求1-7中任一项所述的方法，其特征在于，所述根据所述多个业务流的元数据信息确定分类策略，包括：
    根据所述多个业务流中的每个业务流的报文的平均长度、所述多个业务流中的每个业务流的相邻报文平均间隔、所述多个业务流中的每个业务流的总长度对所述多个业务流进行分类；
    根据分类的结果确定所述分类策略。
  9. 如权利要求1-8中任一项所述的方法，其特征在于，所述方法还包括：
    接收终端设备发送请求消息，所述请求消息携带待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息；
    根据所述五元组信息和所述网络切片策略确定所述网络切片信息；
    向所述终端设备发送响应消息，所述响应消息携带所述网络切片信息，以使所述终端设备根据所述网络切片信息传输所述待传输的业务流。
  10. 如权利要求9所述的方法，其特征在于，所述方法还包括：
    接收所述终端设备发送的报告消息，所述报告消息携带所述待传输的业务流的元数据信息。
  11. 一种传输业务流的方法，其特征在于，包括：
    接收待传输的业务流，所述待传输的业务流携带有网络切片标识，所述网络切片标识所标识的网络切片与所述待传输的业务流所归属的业务流类对应；
    根据所述网络切片标识匹配流表；
    根据匹配的流表和资源权重发送所述待传输的业务流，所述资源权重为所述网络切片在所述网络节点的对应物理端口的带宽权重。
  12. 如权利要求11所述的方法，其特征在于，所述匹配的流表包括执行字段，所述执行字段包括物理端口和与所述网络切片对应的队列标识，
    其中，根据匹配的流表和带宽权重发送所述待传输的业务流，包括：
    根据所述匹配的流表，将所述待传输的业务流发送至所述队列标识所标识的队列中，并在所述物理端口根据所述带宽权重对所述待传输的业务流进行调度。
  13. 如权利要求11或12所述的方法，其特征在于，在接收待传输的业务流之前，所述方法还包括：
    接收集中控制器发送的配置信息，所述配置信息携带网络切片信息和所述资源权重，所述网络切片信息用于指示所述网络切片。
  14. 一种传输业务流的方法，其特征在于，包括：
    接收待传输的业务流；
    向集中控制器发送请求消息，所述请求消息携带所述待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息；
    接收所述集中控制器根据所述五元组信息发送的网络切片信息，所述网络切片信息指示的网络切片与所述待传输的业务流所属的业务类对应；
    向网络节点发送所述待传输的业务流，所述待传输的业务流中的报文携带标识所述网络切片的网络切片标识。
  15. 如权利要求14所述的方法，其特征在于，所述方法还包括：
    向所述集中控制器发送报告消息，所述报告消息携带所述待传输的业务流的元数据信息。
  16. 如权利要求14或15所述的方法，其特征在于，所述方法还包括：
    保存所述待传输的业务流与所述网络切片的对应关系。
  17. 一种网络切片系统，其特征在于，所述系统包括应用感知分析单元、分片决策协同单元和分片控制单元；
    所述应用感知分析单元，用于获取多个业务流中的元数据信息，所述业务流的元数据信息用于指示所述业务流的流量特征；
    所述应用感知分析单元，还用于根据所述多个业务流的元数据信息确定分类策略，所述分类策略包括至少一个业务流类的分类标准，所述分类标准包括对所述业务流类的元数据的范围或取值描述；
    所述分片决策协同单元接收所述应用感知分析单元发送的所述分类策略，用于根据所述分类策略确定网络切片策略，所述网络切片策略用于将网络分为至少一个网络切片，所述网络切片策略包括业务流类与网络切片的映射关系，以及网络切片的切片权重，所述切片权重表示网络切片的资源占所述网络的资源的比率；
    所述分片控制单元接收所述分片决策协同单元发送的所述网络切片策略，用于根据所述网络切片策略确定资源节点策略，所述资源节点策略包括所述网络中的至少一个资源节点与所述至少一个网络切片的映射关系，以及每个资源节点的资源权重，所述资源节点的资源权重表示与所述资源节点对应的网络切片中的每个网络切片的资源占所述资源节点的资源的比率。
  18. 如权利要求17所述的系统，其特征在于，所述分片控制单元还用于：
    向所述资源节点发送配置信息，所述配置信息包括所述资源节点对应的所述至少一个网络切片的标识，和所述资源节点的资源权重信息，所述资源权重信息指示所述资源节点的资源权重。
  19. 如权利要求17或18所述的系统，其特征在于，所述分片决策协同单元具体用于：
    根据所述分类策略和业务支持系统BSS的商业策略确定所述网络切片策略；或
    根据所述分类策略和运营支持系统OSS的商业策略确定所述网络切片策略。
  20. 如权利要求17-19中任一项所述的系统，其特征在于，所述资源节点包括分布式网络节点，所述网络切片的资源包括分布式网络节点的报文处理单元数量、报文发送队列个数与长度、及物理端口带宽大小；和/或
    所述资源节点包括分布式计算节点，所述网络切片的资源包括计算节点的虚拟中央处理器数量；和/或
    所述资源节点包括分布式存储节点，所述网络切片的资源包括存储节点的存储单元数量。
  21. 如权利要求17-20中任一项所述的系统，其特征在于，所述资源节点的资源权重包括所述资源节点对应的每个网络切片在所述资源节点的物理端口占用的带宽与所述物理端口的总带宽的比率。
  22. 如权利要求17-21中任一项所述的系统，其特征在于，所述业务流的元数据信息 包括所述业务流的报文的五元组信息。
  23. 如权利要求17-22中任一项所述的系统，其特征在于，所述应用感知分析单元具体用于：
    根据所述多个业务流中的每个业务流的报文的平均长度、所述多个业务流中的每个业务流的相邻报文平均间隔、所述多个业务流中的每个业务流的总长度对所述多个业务流进行分类；
    根据分类的结果确定所述分类策略。
  24. 如权利要求17-23中任一项所述的系统，其特征在于，所述分片控制单元还用于：
    接收终端设备发送请求消息，所述请求消息携带待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息；
    根据所述五元组信息和所述网络切片策略确定所述待传输的业务流对应的网络切片信息；
    向所述终端设备发送响应消息，所述响应消息携带所述网络切片信息，以使所述终端设备根据所述网络切片信息传输所述待传输的业务流。
  25. 如权利要求17-23中任一项所述的系统，其特征在于，所述应用感知分析单元还用于，
    接收所述终端设备发送的报告消息，所述报告消息携带所述待传输的业务流的元数据信息。
  26. 一种网络节点，其特征在于，包括：
    第一接收单元，用于接收待传输的业务流，所述待传输的业务流携带有网络切片标识，所述网络切片标识所标识的网络切片与所述待传输的业务流所归属的业务流类对应；
    匹配单元，用于根据所述第一接收单元接收的所述网络切片标识匹配流表；
    发送单元，用于根据所述匹配单元匹配的流表和资源权重发送所述待传输的业务流，所述资源权重为所述网络切片在所述网络节点的对应物理端口的带宽权重。
  27. 如权利要求26所述的网络节点，其特征在于，所述匹配的流表包括执行字段，所述执行字段包括物理端口和与所述网络切片对应的队列标识，
    其中，所述发送单元具体用于：
    根据所述匹配的流表，将所述待传输的业务流发送至所述队列标识所标识的队列中，并在所述物理端口根据所述带宽权重对所述待传输的业务流进行调度。
  28. 如权利要求26或27所述的网络节点，其特征在于，所述网络节点还包括：
    第二接收单元，用于接收集中控制器发送的配置信息，所述配置信息携带网络切片信息和所述资源权重，所述网络切片信息用于指示所述网络切片。
  29. 一种终端设备，其特征在于，包括：
    第一接收单元，用于接收待传输的业务流；
    第一发送单元，用于向集中控制器发送请求消息，所述请求消息携带所述待传输的业务流中的报文的五元组信息，所述请求消息用于请求与所述待传输的业务流对应的网络切片信息；
    第二接收单元，用于接收所述集中控制器根据所述第一发送单元发送的所述五元组信息发送的网络切片信息，所述网络切片信息指示的网络切片与所述待传输的业务流所属的 业务类对应；
    第二发送单元，用于向网络节点发送所述待传输的业务流，所述待传输的业务流中的报文携带标识所述网络切片的网络切片标识。
  30. 如权利要求29所述的终端设备，其特征在于，所述第一发送单元还用于：
    向所述集中控制器发送报告消息，所述报告消息携带所述待传输的业务流的元数据信息。
  31. 如权利要求29或30所述的终端设备，其特征在于，所述终端设备还还包括：
    存储单元，用于保存所述待传输的业务流与所述网络切片的对应关系。
PCT/CN2017/093856 2016-07-25 2017-07-21 网络切片方法和系统 WO2018019184A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP17833487.6A EP3471341B1 (en) 2016-07-25 2017-07-21 Network slice method and system
US16/239,269 US11381452B2 (en) 2016-07-25 2019-01-03 Network slicing method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610589662.6 2016-07-25
CN201610589662.6A CN107659419B (zh) 2016-07-25 2016-07-25 网络切片方法和系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/239,269 Continuation US11381452B2 (en) 2016-07-25 2019-01-03 Network slicing method and system

Publications (1)

Publication Number Publication Date
WO2018019184A1 true WO2018019184A1 (zh) 2018-02-01

Family

ID=61015669

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/093856 WO2018019184A1 (zh) 2016-07-25 2017-07-21 网络切片方法和系统

Country Status (4)

Country Link
US (1) US11381452B2 (zh)
EP (1) EP3471341B1 (zh)
CN (1) CN107659419B (zh)
WO (1) WO2018019184A1 (zh)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110661636A (zh) * 2018-06-30 2020-01-07 华为技术有限公司 用于获取网络切片的方法、装置和系统
CN110662261A (zh) * 2018-06-30 2020-01-07 华为技术有限公司 一种网络切片的资源分配方法及装置
US10681589B1 (en) 2018-12-03 2020-06-09 At&T Itellectual Property I, L.P. Vertical slice management and selection leveraging dynamic alliances of access core and service network resources
CN112205058A (zh) * 2018-03-27 2021-01-08 诺基亚通信公司 基于一个或多个令牌计数器的网络切片
CN113541980A (zh) * 2020-04-14 2021-10-22 中国移动通信集团浙江有限公司 网络切片故障根因定位方法、装置、计算设备及存储介质
CN113726847A (zh) * 2021-07-15 2021-11-30 阿里巴巴新加坡控股有限公司 网络系统、网络切分系统、网络切分方法及电子设备
RU2767314C1 (ru) * 2018-10-09 2022-03-17 Телефонактиеболагет Лм Эрикссон (Пабл) Устройства, способы, компьютерные программы и компьютерные программные продукты для поддержки взаимоисключающего доступа к сетевым срезам
US11343333B2 (en) 2018-11-16 2022-05-24 Tencent Technology (Shenzhen) Company Limited Service data transmission method and apparatus, computer device, and computer-readable storage medium
US11516113B2 (en) 2018-03-20 2022-11-29 Microsoft Technology Licensing, Llc Systems and methods for network slicing

Families Citing this family (52)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3520553B1 (en) * 2016-09-29 2023-03-01 Telefonaktiebolaget LM Ericsson (PUBL) Quality of service differentiation between network slices
WO2018170922A1 (zh) * 2017-03-24 2018-09-27 华为技术有限公司 一种配置网络切片的方法、装置和系统
CN108810903B (zh) * 2017-05-04 2020-08-18 中国移动通信有限公司研究院 一种数据传输配置及数据传输方法和装置
WO2019164518A1 (en) * 2018-02-25 2019-08-29 Nokia Solutions And Networks Oy Method and system for automated dynamic network slice deployment using artificial intelligence
CN108599994B (zh) * 2018-03-26 2021-04-20 华南理工大学 一种基于流量聚类的sdn切片构造方法
CN110324284B (zh) * 2018-03-30 2020-10-27 华为技术有限公司 接入ims的方法和通信装置
CN110621045B (zh) * 2018-06-20 2022-05-13 华为云计算技术有限公司 一种物联网业务路由的方法
CN109039742A (zh) * 2018-08-03 2018-12-18 西安电子科技大学 一种服务不同业务类型的网络切片及其切换方法
CN110798415B (zh) * 2018-08-03 2022-02-18 中兴通讯股份有限公司 一种业务传输的方法、设备及计算机存储介质
CN110838926A (zh) * 2018-08-15 2020-02-25 中国移动通信有限公司研究院 管理网络切片的方法及系统
CN111050361B (zh) * 2018-10-12 2023-07-25 中国移动通信有限公司研究院 一种报文传送方法、装置及系统
CN111147271B (zh) * 2018-11-06 2022-09-09 中兴通讯股份有限公司 一种设备管理的方法、设备及系统
CN109600259A (zh) * 2018-12-11 2019-04-09 浙江工商大学 一种基于软件可定义的实时传输机制
KR102641254B1 (ko) * 2019-01-08 2024-02-29 삼성전자 주식회사 무선 통신 시스템에서 종단 간 네트워크를 제어하는 관리 장치 및 방법
US10979314B2 (en) 2019-01-24 2021-04-13 Vmware, Inc. Dynamic inter-cloud placement of virtual network functions for a slice
US10944647B2 (en) 2019-01-24 2021-03-09 Vmware, Inc. Dynamic inter-cloud placement of virtual network functions for a slice
EP3957100A4 (en) * 2019-04-15 2022-10-26 Nokia Solutions and Networks Oy ASSIGNMENT OF RESOURCES IN A NETWORK SLOT
US10897423B2 (en) 2019-05-14 2021-01-19 Vmware, Inc. Congestion avoidance in a slice-based network
US10892994B2 (en) * 2019-05-14 2021-01-12 Vmware, Inc. Quality of service in virtual service networks
US11012288B2 (en) 2019-05-14 2021-05-18 Vmware, Inc. Congestion avoidance in a slice-based network
US10958579B2 (en) 2019-05-14 2021-03-23 Vmware, Inc. Congestion avoidance in a slice-based network
US11588733B2 (en) 2019-05-14 2023-02-21 Vmware, Inc. Slice-based routing
CN110324260B (zh) * 2019-06-21 2020-10-09 北京邮电大学 一种基于流量识别的网络功能虚拟化智能调度方法
CN112152899B (zh) * 2019-06-28 2023-04-07 中兴通讯股份有限公司 一种基于网络切片的数据处理方法及装置
CN110505101B (zh) * 2019-09-05 2021-11-12 无锡北邮感知技术产业研究院有限公司 一种网络切片编排方法及装置
CN112448838B (zh) * 2019-09-05 2023-05-16 中国电信股份有限公司 流量转发方法、装置和计算机可读存储介质
KR20220052343A (ko) * 2019-09-11 2022-04-27 후아웨이 테크놀러지 컴퍼니 리미티드 데이터 전송 제어 방법 및 장치
US11070422B2 (en) 2019-09-16 2021-07-20 Cisco Technology, Inc. Enabling enterprise segmentation with 5G slices in a service provider network
WO2021062873A1 (zh) * 2019-10-03 2021-04-08 华为技术有限公司 一种无线资源分配策略的配置方法、装置及系统
CN112804687B (zh) * 2019-11-14 2023-04-11 华为技术有限公司 一种资源分配方法、装置、系统及存储介质
CN116567736A (zh) * 2019-11-14 2023-08-08 华为技术有限公司 一种资源分配方法、装置、系统及存储介质
CN111242232B (zh) * 2020-01-17 2023-11-14 广州欧赛斯信息科技有限公司 数据分片处理方法、装置及学分银行服务器
CN113395612B (zh) * 2020-03-11 2022-11-22 华为技术有限公司 一种光纤通信中的数据转发方法以及相关装置
US11202234B1 (en) 2020-06-03 2021-12-14 Dish Wireless L.L.C. Method and system for smart operating bandwidth adaptation during power outages
US11265135B2 (en) 2020-06-03 2022-03-01 Dish Wireless Llc Method and system for slicing assigning for load shedding to minimize power consumption where gNB is controlled for slice assignments for enterprise users
CN112511329A (zh) * 2020-06-12 2021-03-16 中兴通讯股份有限公司 信息通告方法及设备、可读介质
US11470549B2 (en) 2020-07-31 2022-10-11 Dish Wireless L.L.C. Method and system for implementing mini-slot scheduling for all UEs that only are enabled to lower power usage
US11405941B2 (en) 2020-07-31 2022-08-02 DISH Wireless L.L.C Method and system for traffic shaping at the DU/CU to artificially reduce the total traffic load on the radio receiver so that not all the TTLs are carrying data
CN114205239A (zh) * 2020-08-26 2022-03-18 中国移动通信集团终端有限公司 网络切片服务确定方法、装置、设备及计算机存储介质
CN114125882A (zh) * 2020-08-31 2022-03-01 华为技术有限公司 一种资源传输保障方法、装置及系统
CN112399609B (zh) * 2020-12-03 2023-08-11 中国联合网络通信集团有限公司 一种资源配置方法及装置
US20220239690A1 (en) * 2021-01-27 2022-07-28 EMC IP Holding Company LLC Ai/ml approach for ddos prevention on 5g cbrs networks
US11564159B2 (en) * 2021-02-01 2023-01-24 T-Mobile Usa, Inc. Orchestration and mediation stack
CN115348202A (zh) * 2021-05-12 2022-11-15 中国移动通信有限公司研究院 基于网络切片的数据传输方法、装置、设备及存储介质
CN113240020B (zh) * 2021-05-19 2022-11-01 浙江凡双科技有限公司 一种基于超图分割的5g无线通信网络切片分类方法
CN115473818B (zh) * 2021-06-10 2024-03-26 维沃移动通信有限公司 流量特征分析结果的获取方法、装置及网络侧设备
CN114125947A (zh) * 2021-06-30 2022-03-01 北京中电飞华通信有限公司 基于切片管理的无线专网基站及其资源分配方法
US11871318B2 (en) * 2021-11-08 2024-01-09 Verizon Patent And Licensing Inc. Systems and methods for tiered network slice design and management in a wireless network
CN114268537B (zh) * 2021-11-26 2023-12-22 华中科技大学 面向确定性网络的网络切片生成及动态配置系统及方法
CN113993167B (zh) * 2021-12-14 2022-07-12 荣耀终端有限公司 数据流调度方法、终端及存储介质
US11876877B2 (en) * 2022-03-29 2024-01-16 United States Of America As Represented By The Secretary Of The Navy Reinforcement-learning based queue-management method for undersea networks
US20240004933A1 (en) * 2022-06-29 2024-01-04 Tigergraph, Inc. Minhash signatures as vertices for fuzzy string match on graph

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102972080A (zh) * 2010-05-17 2013-03-13 创网协会公司 用于网络虚拟化的方法和系统
CN103269282A (zh) * 2013-04-25 2013-08-28 杭州华三通信技术有限公司 网络配置自动部署方法和装置
US20150154257A1 (en) * 2013-12-04 2015-06-04 Nec Laboratories America, Inc. System and method for adaptive query plan selection in distributed relational database management system based on software-defined network

Family Cites Families (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6295532B1 (en) * 1999-03-02 2001-09-25 Nms Communications Corporation Apparatus and method for classifying information received by a communications system
US6594268B1 (en) * 1999-03-11 2003-07-15 Lucent Technologies Inc. Adaptive routing system and method for QOS packet networks
US7095759B1 (en) * 1999-03-19 2006-08-22 At Comm Corporation Combined telephone PBX and computer data router with pooled resources
US6778530B1 (en) * 1999-11-08 2004-08-17 Juniper Networks, Inc. Method and apparatus for multiple field matching in network device
US7042848B2 (en) * 2001-05-04 2006-05-09 Slt Logic Llc System and method for hierarchical policing of flows and subflows of a data stream
US7440403B2 (en) * 2002-10-10 2008-10-21 P-Cube Ltd. Method and software for calculating rate constraints
US20050089054A1 (en) * 2003-08-11 2005-04-28 Gene Ciancaglini Methods and apparatus for provisioning connection oriented, quality of service capabilities and services
JP4761365B2 (ja) * 2005-12-28 2011-08-31 Ｋｄｄｉ株式会社 通信スケジューリング方法
US7525969B2 (en) * 2006-08-25 2009-04-28 Cisco Technology, Inc. NAT processing in a VRF environment
US8000251B2 (en) * 2007-11-14 2011-08-16 Cisco Technology, Inc. Instrumenting packet flows
US20110125748A1 (en) * 2009-11-15 2011-05-26 Solera Networks, Inc. Method and Apparatus for Real Time Identification and Recording of Artifacts
US20140115126A1 (en) * 2012-10-19 2014-04-24 Electronics And Telecommunications Research Institute System for controlling and verifying open programmable network and method thereof
GB2510345A (en) * 2013-01-30 2014-08-06 Nec Corp Sharing base station resources among plural network operators
US9413667B2 (en) * 2013-02-15 2016-08-09 Telefonaktiebolaget Lm Ericsson (Publ) Methods and network nodes for traffic steering based on per-flow policies
US20140330937A1 (en) * 2013-05-03 2014-11-06 Microsoft Corporation End-to-end classification of storage traffic streams
US9100326B1 (en) * 2013-06-13 2015-08-04 Narus, Inc. Automatic parsing of text-based application protocols using network traffic data
EP2830275A1 (en) * 2013-07-23 2015-01-28 Thomson Licensing Method of identification of multimedia flows and corresponding apparatus
US10069683B2 (en) * 2013-09-27 2018-09-04 Nxp Usa, Inc. Apparatus for optimising a configuration of a communications network device
EP3059903B1 (en) * 2013-10-18 2019-08-21 NEC Corporation System and method for controlling network
CN104679665A (zh) * 2013-12-02 2015-06-03 中兴通讯股份有限公司 一种实现分布式文件系统块存储的方法及系统
US10050892B2 (en) * 2014-01-14 2018-08-14 Marvell International Ltd. Method and apparatus for packet classification
CN104363159B (zh) * 2014-07-02 2018-04-06 北京邮电大学 一种基于软件定义网络的开放虚拟网络构建系统和方法
US9882814B2 (en) * 2014-09-25 2018-01-30 Intel Corporation Technologies for bridging between coarse-grained and fine-grained load balancing
US9722935B2 (en) * 2014-10-16 2017-08-01 Huawei Technologies Canada Co., Ltd. System and method for transmission management in software defined networks
US9594592B2 (en) * 2015-01-12 2017-03-14 International Business Machines Corporation Dynamic sharing of unused bandwidth capacity of virtualized input/output adapters
US10044620B2 (en) * 2015-05-01 2018-08-07 Hughes Network Systems, Llc Multi-phase IP-flow-based classifier with domain name and HTTP header awareness
US9762402B2 (en) * 2015-05-20 2017-09-12 Cisco Technology, Inc. System and method to facilitate the assignment of service functions for service chains in a network environment
EP3314827B1 (en) * 2015-06-25 2022-08-03 NEC Corporation Method and system for managing data traffic in a computing network
CN106375987B (zh) * 2015-07-22 2021-08-20 中兴通讯股份有限公司 一种网络切片的选择方法及系统
EP3363233B1 (en) * 2015-10-12 2020-01-01 Telefonaktiebolaget LM Ericsson (publ) Load balancing of data traffic in a communications network
US11006309B2 (en) * 2015-11-18 2021-05-11 Telefonaktiebolaget Lm Ericsson (Publ) Network node, radio network node and methods performed therein
US10863560B2 (en) * 2015-11-18 2020-12-08 Telefonaktiebolaget Lm Ericsson (Publ) First network node, receiving network node and methods performed therein
WO2017113109A1 (en) * 2015-12-29 2017-07-06 Telefonaktiebolaget Lm Ericsson (Publ) Method and apparatus for virtualized network service provision
US11044203B2 (en) * 2016-01-19 2021-06-22 Cisco Technology, Inc. System and method for hosting mobile packet core and value-added services using a software defined network and service chains
CN108781447B (zh) * 2016-03-23 2022-09-06 日本电气株式会社 基站、用户设备和其中的方法
EP3424242B1 (en) * 2016-04-01 2021-06-30 Sony Corporation Terminal device, infrastructure equipment, wireless telecommunications network, methods and integrated circuitry
WO2017204067A1 (ja) * 2016-05-26 2017-11-30 京セラ株式会社 ネットワーク装置
KR102358918B1 (ko) * 2016-07-04 2022-02-07 삼성전자 주식회사 무선 통신 시스템에서 서비스에 따른 보안 관리 방법 및 장치

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102972080A (zh) * 2010-05-17 2013-03-13 创网协会公司 用于网络虚拟化的方法和系统
CN103269282A (zh) * 2013-04-25 2013-08-28 杭州华三通信技术有限公司 网络配置自动部署方法和装置
US20150154257A1 (en) * 2013-12-04 2015-06-04 Nec Laboratories America, Inc. System and method for adaptive query plan selection in distributed relational database management system based on software-defined network

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
"NGMN 5G WHITE PAPER", NGMN AUIANCE, 17 February 2015 (2015-02-17), pages 46 - 47, XP055561266 *
"Study on Architecture for Next Generation System", 3GPP TR 23. 799 V0.6.0, 22 July 2016 (2016-07-22), pages 44 - 52, XP 055459528 *
See also references of EP3471341A4 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11516113B2 (en) 2018-03-20 2022-11-29 Microsoft Technology Licensing, Llc Systems and methods for network slicing
CN112205058A (zh) * 2018-03-27 2021-01-08 诺基亚通信公司 基于一个或多个令牌计数器的网络切片
US11395220B2 (en) 2018-06-30 2022-07-19 Huawei Technologies Co., Ltd. Method and apparatus for allocating resource to network slice
CN110662261A (zh) * 2018-06-30 2020-01-07 华为技术有限公司 一种网络切片的资源分配方法及装置
CN110661636A (zh) * 2018-06-30 2020-01-07 华为技术有限公司 用于获取网络切片的方法、装置和系统
US11533239B2 (en) 2018-06-30 2022-12-20 Huawei Technologies Co., Ltd. Method, apparatus, and system for obtaining network slice
US11271818B2 (en) 2018-06-30 2022-03-08 Huawei Technologies Co., Ltd. Method, apparatus, and system for obtaining network slice
CN110662261B (zh) * 2018-06-30 2022-06-28 华为技术有限公司 一种网络切片的资源分配方法及装置
US11812370B2 (en) 2018-10-09 2023-11-07 Telefonaktiebolaget Lm Ericsson (Publ) Apparatuses, methods, computer programs, and computer program products for supporting mutually exclusive access to network slices
RU2767314C1 (ru) * 2018-10-09 2022-03-17 Телефонактиеболагет Лм Эрикссон (Пабл) Устройства, способы, компьютерные программы и компьютерные программные продукты для поддержки взаимоисключающего доступа к сетевым срезам
US11343333B2 (en) 2018-11-16 2022-05-24 Tencent Technology (Shenzhen) Company Limited Service data transmission method and apparatus, computer device, and computer-readable storage medium
US10887798B2 (en) 2018-12-03 2021-01-05 At&T Intellectual Property I, L.P. Vertical slice management and selection leveraging dynamic alliances of access core and service network resources
US10681589B1 (en) 2018-12-03 2020-06-09 At&T Itellectual Property I, L.P. Vertical slice management and selection leveraging dynamic alliances of access core and service network resources
CN113541980A (zh) * 2020-04-14 2021-10-22 中国移动通信集团浙江有限公司 网络切片故障根因定位方法、装置、计算设备及存储介质
CN113726847A (zh) * 2021-07-15 2021-11-30 阿里巴巴新加坡控股有限公司 网络系统、网络切分系统、网络切分方法及电子设备

Also Published As

Publication number Publication date
CN107659419A (zh) 2018-02-02
EP3471341A4 (en) 2019-05-15
EP3471341B1 (en) 2021-06-30
US11381452B2 (en) 2022-07-05
US20190140904A1 (en) 2019-05-09
EP3471341A1 (en) 2019-04-17
CN107659419B (zh) 2021-01-01

Similar Documents

Publication Publication Date Title
WO2018019184A1 (zh) 网络切片方法和系统
US20190014053A1 (en) Network Flow Control Method And Network Device
JP6559670B2 (ja) ネットワーク機能仮想化情報コンセントレータのための方法、システム、およびコンピュータ読取可能媒体
WO2017005223A1 (zh) 一种智能电表的信息共享方法、智能电表及采集路由器
EP4024778A1 (en) Method for determining required bandwidth for data stream transmission, and devices and system
US11165716B2 (en) Data flow processing method and device
WO2012145841A1 (en) Hierarchical profiled scheduling and shaping
CN112491619A (zh) 一种基于sdn的服务定制网络资源自适应分配技术
WO2022111329A1 (zh) 报文处理方法及设备
EP4037270A1 (en) Service traffic adjusting method and apparatus
WO2016150020A1 (zh) 基于调度流标识的报文调度方法和装置
WO2016109970A1 (zh) 网络实体及服务策略管理方法
CN115037694A (zh) 一种数据传输方法、装置、电子设备和存储介质
WO2020143785A1 (zh) 无线接入网中心单元和通信方法
CN110417656A (zh) 一种基于sdn网络的服务质量保障系统以及方法
Silva et al. Extending OpenFlow with industrial grade communication services
Tayyaba et al. NPRA: Novel policy framework for resource allocation in 5G software defined networks
Velasco et al. Pervasive monitoring and distributed intelligence for 6G near real-time operation
US11757742B2 (en) System and method to distribute traffic flows among a plurality of applications in a data center system
Rashid et al. Traffic intensity based efficient packet schedualing
WO2023280004A1 (zh) 一种网络配置方法、设备和系统
WO2021238694A1 (zh) 确定传输业务流的带宽的方法、设备和系统
WO2023125995A1 (zh) 一种业务切片处理方法和装置
WO2022253194A1 (zh) 报文转发方法、装置及通信网络
WO2022253192A1 (zh) 报文转发方法、装置及通信网络

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17833487

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017833487

Country of ref document: EP

Effective date: 20190108

NENP Non-entry into the national phase

Ref country code: DE