WO2019104788A1 - 一种基于可编程网络技术的多宿主网络路由转发方法 - Google Patents

一种基于可编程网络技术的多宿主网络路由转发方法 Download PDF

Info

Publication number
WO2019104788A1
WO2019104788A1 PCT/CN2017/117700 CN2017117700W WO2019104788A1 WO 2019104788 A1 WO2019104788 A1 WO 2019104788A1 CN 2017117700 W CN2017117700 W CN 2017117700W WO 2019104788 A1 WO2019104788 A1 WO 2019104788A1
Authority
WO
WIPO (PCT)
Prior art keywords
port
address
forwarding
destination address
routing
Prior art date
Application number
PCT/CN2017/117700
Other languages
English (en)
French (fr)
Inventor
王劲林
尤佳莉
麻朴方
王玲芳
叶晓舟
程钢
Original Assignee
中国科学院声学研究所
北京中科海力技术有限公司
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 中国科学院声学研究所, 北京中科海力技术有限公司 filed Critical 中国科学院声学研究所
Priority to US16/760,737 priority Critical patent/US11522784B2/en
Priority to JP2020524190A priority patent/JP6921322B2/ja
Priority to AU2017441267A priority patent/AU2017441267B2/en
Priority to EP17933683.9A priority patent/EP3709583B1/en
Priority to KR1020207015402A priority patent/KR102342723B1/ko
Publication of WO2019104788A1 publication Critical patent/WO2019104788A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/645—Splitting route computation layer and forwarding layer, e.g. routing according to path computational element [PCE] or based on OpenFlow functionality
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H04L45/04—Interdomain routing, e.g. hierarchical routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H04L45/03—Topology update or discovery by updating link state protocols
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/20—Hop count for routing purposes, e.g. TTL
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/38—Flow based routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/42—Centralised routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/54—Organization of routing tables
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/64—Routing or path finding of packets in data switching networks using an overlay routing layer
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/74—Address processing for routing
    • H04L45/745—Address table lookup; Address filtering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/60—Software-defined switches

Definitions

  • the invention relates to computer network technology, in particular to a multi-homed network routing and forwarding method based on programmable network technology.
  • the current user's terminal will typically be in multiple host networks, such as Wi-Fi and 4G networks.
  • the user can increase the transmission efficiency by transmitting in parallel through multiple host networks.
  • the network with good status can be selected according to the change of the host network to improve the transmission reliability.
  • the current 4G solution is to choose one of the hosts to communicate, which can not enjoy the advantages of multi-homed reliability, double transmission and better host selection according to network status. Therefore, in the face of the next generation of 5G networks, it is necessary to implement multi-homing functions.
  • multi-homing There are two main ways to implement multi-homing. One is the separation of addresses and identifiers, such as LISP, HIP, LIN6, and SHIM6. These solutions have the potential to solve mobility and multi-homing, but at the cost of modifying terminal devices. Or deploy a new network entity (mapping system or dedicated edge device). The other is the transport layer scheme, such as SCTP and MPTCP, which support concurrent multipath transmission, but the support for mobility is not good.
  • the existing solution is mainly an end-to-end multi-homing transmission scheme, which allows users to use multiple host networks at the same time, improving the quality of service.
  • the network information known by the end-end multi-homing scheme is limited, and the multi-homed path switching cannot be performed in time according to the state of the network.
  • the object of the present invention is to solve the problem that the traditional multi-homing technology cannot adjust the multi-homed network according to the network state in time, and consider that the programmable network device can realize the independent forwarding of the protocol and increase the flexibility of the operation of the data packet;
  • the multi-homed network routing and forwarding method based on network state dynamic adjustment can effectively improve the communication quality of users.
  • the present invention provides a routing and forwarding method for a multi-homed network based on a programmable network technology, the method comprising: the multi-homed network includes a controller and a programmable network device; The state of the network maintains information in the programmable network device that maintains two tables: a forwarding flow table and a link information table, and modifies and forwards the data packet; the method includes :
  • the address-to-port matching is performed according to the forwarding flow table address matching method, and the addresses are classified according to the matched ports, so that each port corresponds to one destination address group.
  • the destination address is used as the destination address, and the packet is forwarded to the destination address by the general route forwarding method. Otherwise, the destination address group is encapsulated according to its corresponding address.
  • the port forwards the packet and uses a single address selection method to determine a single destination address hop by hop until the destination address is reached.
  • the programmable network device is a network device having a processing custom message and a protocol-independent forwarding capability under an SDN architecture; the programmable network device is implemented in a forwarding layer by extending an Openflow protocol. Forwarding flow table processing and control of custom messages; the programmable network device is a router or switch supporting P4, POF or PIF technology.
  • the forwarding flow table is a flow table used for packet forwarding, and records forwarding information of the data packet to the destination;
  • the controller obtains the forwarding flow table through the routing protocol and delivers the forwarding flow table to the Programming the network device;
  • the link information table records each forwarding port port k in the programmable network device and a status information value of the corresponding related link, status k,1 ,status k,2, ..., the status information value includes The link state of the next hop or multi-hop and the link state of the possible path, including the port load, link bandwidth, and delay of the next hop or multi-hop.
  • the controller is a network policy control center, and manages a routing and forwarding policy of the network, which is obtained by an SDN controller; the controller includes a forwarding flow table control module and a link information table.
  • Control module
  • the forwarding flow table control module is configured to send, to the programmable network device, a command for creating, modifying, and deleting an entry of the forwarding flow table according to the routing rule;
  • the link information table control module is configured to send, by the device, the command for creating, modifying, and deleting the entry of the link information table to the programmable network device after acquiring, calculating, and learning the network state information.
  • the single address selection method is: if only one address or only one address satisfies the destination address matching condition, the address is selected as the destination address; if multiple addresses currently reach the destination address matching condition According to the selection policy, an address that satisfies the matching condition of the destination address is selected as the destination address.
  • the method specifically includes:
  • Step 1) Each data packet carries multiple addresses of the user and the first k hop information routed to the current address. If the user has n hosts, the corresponding set of multiple addresses is ⁇ addr 1 , addr 2 , ..., addr n ⁇ ;
  • Step 2 When performing routing and forwarding on the data packet, the address-to-port matching is performed according to the forwarding flow table address matching method, and the addresses are classified according to the matched port, so that each port corresponds to one destination address group;
  • Step 3) Query the information of all the ports in the link information table, and select the port with the good state as the forwarding port of the next hop through the port selection function, and cut out the destination address group corresponding to the port that is not selected;
  • Step 4) further cutting the remaining destination address group by using the address filtering function and the pre-hop information carried by the data; if a single destination address is obtained, proceeding to step 6); otherwise, proceeding to step 5); the single destination address Means that only one address or only one of the multiple addresses satisfies the destination address matching condition;
  • Step 5 If there are multiple addresses satisfying the destination address matching condition, select a single destination address that satisfies the destination address matching condition according to the selection policy, and proceed to step 6); otherwise, encapsulate the address group and set the data according to the corresponding port. Packet forwarding, go to step 3);
  • Step 6 The obtained single destination address is used as the destination address, and the general route forwarding method is used to forward the packet until the destination is reached.
  • the step 2) specifically includes:
  • Step 201) Matching a plurality of address data packets to the forwarding flow table, each address matching the corresponding forwarding port according to the forwarding flow table address matching method, and having n addresses ⁇ addr 1 , addr 2 , . . . , addr n ⁇
  • Step 202) classify the address according to the matched forwarding port; the address corresponding to each port is classified into an address group, and the mapping relationship is ⁇ port 1 :[addr 11 , addr 12 ,...], port 2 : [addr 21 , addr 22 ,...],...,port m :[addr m1 ,addr m2 ,...] ⁇ .
  • the forwarding flow table address matching method is: matching corresponding forwarding ports according to the address or identifier of the data packet, and different routing mechanisms have different matching methods, including: IP address matching and ICN The name matches.
  • Pack is the message content
  • F(x) is the port selection function based on the status information
  • the port selection function F(x) includes a state calculation function F 1 (x) and a selection function F 2 (x); the state calculation F 1 (x) passes a value of an item in the link information table, Or calculating a state value for each candidate port according to a plurality of values; the selection function F 2 (x) selects a next hop port according to the calculated state value, and selects the port with the highest state value as the next Jump forwarding port.
  • the address filtering function of the step 4) is used to filter the address group ADDR i corresponding to the port port i selected by the port selection function, and the implementation process is: filtering information according to the previous k hop address. If there is no address deletion within the k hop, the address information in the u ADDR i is deleted by the probability p. In the process, at least one address in the ADDR i is guaranteed.
  • the destination address matching condition is: calculating an address matching scoring function under the current routing mechanism as G(x), where x is an address, and G(x) is obtained according to different matching methods defined by different routing mechanisms.
  • the score indicates the degree of matching with the destination address; when G(x)>w, w is a preset or dynamically set threshold, indicating that the address x has reached the destination address matching condition.
  • the selection strategy of the step 5) is: random selection, selection according to link prediction information or subnet load selection.
  • the system and method of the present invention can timely adjust the transmission path of the data packet according to the dynamic change of the network, so that the user can fully enjoy the reliability brought by the multi-homing, double the transmission, and select a better host according to the network state.
  • FIG. 1 is a structural diagram of a multi-homed network of a multi-purpose address based on a programmable network technology of the present invention
  • FIG. 2 is a structural diagram of a switch in a multi-homed network based on a programmable network technology of the present invention and a flow chart for processing a multi-purpose address data packet.
  • the present invention will be further described by taking a multi-purpose IP address based on POF (Protocol-Oblivious Forwarding) as an example.
  • POF Protocol-Oblivious Forwarding
  • the invention provides a routing and forwarding method for a multi-homed network based on a programmable network device, in which each data packet of the transmitting end carries multiple IP addresses and the first k-hop information routed to the current address;
  • the maintenance information of the programmable network device two information tables are maintained, one is a forwarding flow table obtained based on the IP routing forwarding mechanism, and the other is a link information table, wherein the link information table provides each of the programmable network devices. Forwarding port and status information of the corresponding link.
  • the address-to-port matching is performed according to the forwarding flow table IP address matching method, and the addresses are classified according to the matched ports, so that each port corresponds to a destination address group, based on the state table.
  • the information and port selection function select the appropriate port and the corresponding destination address group, and trim the address group corresponding to the remaining ports; further modify the remaining address groups by using the IP address filtering function and the forward hop information carried by the data;
  • the address in the address group it is determined whether the IP access network address matching condition is met. If not, the address group is encapsulated, and the data packet is forwarded according to the corresponding port. If it is reached, an address is selected according to the single address selection method. As the destination address, delete other addresses, encapsulate the packets, and forward them according to the corresponding port. The above process needs to be judged and executed in each hop forwarding until it reaches the destination.
  • the programmable network device refers to a network device that processes a custom message and a protocol-independent forwarding capability under an SDN (Software Defined Network) architecture. Through the extension of the Openflow protocol, the flow table is used to process and control the customized packet at the forwarding layer.
  • Programmable network devices include, but are not limited to, routers and switches that support technologies such as P4, POF, and PIF. Each programmable network device can support multiple addresses.
  • the programmable network is a network formed by interconnecting programmable network devices.
  • the multi-homed network refers to that a terminal connects to a network through multiple access networks, and each terminal can send data through multiple networks, and can receive data from multiple networks at the same time.
  • the network can dynamically select a better access network to provide services for the terminal according to its own state.
  • the multi-homed network is based on the SDN architecture and consists mainly of controllers and programmable network devices. Among them, the controller mainly maintains the information in the programmable network device according to the state of the network, and the programmable network device is mainly responsible for completing the modification and forwarding of the data packet.
  • the method for matching the IP address of the forwarding flow table refers to matching the corresponding forwarding port according to the IP address in the data packet, and matching can be performed by using an IP address matching method.
  • the forwarding flow table refers to a flow table used for packet forwarding, where the flow table is a table for forwarding data packets in the SDN, wherein the forwarding information of the data packet to the destination is maintained, and the controller uses the routing protocol to Obtain and deliver to a programmable network device.
  • the link information table maintains status information of the link, the information including port load, current link status, such as bandwidth, delay, and amount related to network status.
  • the controller is a network policy control center, and manages a routing and forwarding policy of the network, which can be obtained by extending the SDN controller, and the controller sends a control command to the programmable network device through the southbound interface, and the programmable network device executes the commands.
  • the structure includes a forwarding flow table control module and a link information table control module.
  • the forwarding flow table control module runs a routing algorithm based on the obtained network state to generate a routing rule, and sends a command to create, modify, and delete a flow entry of the forwarding flow table to the programmable network device through the southbound interface; After obtaining, calculating, and learning the network status information, the information table control module sends a command to create, modify, and delete the entry of the link information table to the programmable network through the southbound interface, such as according to the dynamic change of the network. The learning of the network state in the past to obtain and predict the state of the link in the network, thereby maintaining the link information table in the programmable network device.
  • the information in the link information table is that the controller sends the information maintenance policy to the programmable network device, and the programmable network device obtains relevant information and maintains the link through statistical information, network measurement, and learning of dynamic changes of the network.
  • the information table reflects the dynamic changes of the link of the network.
  • the port port k corresponding status information is status k, 1 , status k, 2, ..., and its value includes but is not limited to the link status of the next hop or multi-hop, the link status of the possible path, etc., for example: One-hop or multi-hop port load, link bandwidth, delay, or other network-related information can be obtained by measurement, prediction, and the like.
  • the route forwarding method specifically includes the following processes:
  • Step 1) Each data packet carries multiple addresses of users, which are allocated by the host network where the user is located. If the user has n hosts, the corresponding set of multiple addresses is ⁇ IP 1 , IP 2 ,..., IP. n ⁇ .
  • Step 2) Matching a plurality of address data packets to the forwarding flow table, each address matching the corresponding forwarding port according to the forwarding flow table address matching method, and having n addresses ⁇ IP 1 , IP 2 , . . . , IP n ⁇
  • Each address is then matched to a port based on the forwarding flow table.
  • Step 3) Classify the address according to the matched forwarding port; the address corresponding to each port is classified into an address group, and the mapping relationship is ⁇ port 1 :[IP 11 , IP 12 ,...], port 2 : [IP 21 , IP 22 ,...],...,port m :[IP m1 , IP m2 ,...] ⁇ .
  • Step 4) Query the information of all the ports in the link information table, and select the port with the good state as the forwarding port of the next hop through the port selection function, and delete the address group corresponding to the port that is not selected.
  • the address filtering information of the middle k-hop is filtered by the address filtering function to further the destination address of the destination address group corresponding to the selected forwarding port, and the destination address group corresponding to the unselected port is tailored;
  • the calculation function for the state value is F 1 (x), here the historical load of the corresponding port, and the output is the state value calculated according to the defined function.
  • the appropriate forwarding port port x is obtained by F 2 (x) selection for next hop forwarding.
  • the port selection function F(x) includes two steps of state calculation F 1 (x) and selection F 2 (x).
  • the selection step F 2 (y) is to select the next hop port according to the state value y calculated by F 1 , where F 2 is calculated by sorting the ports according to the F 1 (x) state value from high to low, and selecting the state. The port with the highest value is used as the next hop forwarding port.
  • the address filtering function comprising the steps of: Group address ADDR i port selection function corresponding to the selected port is port i determination, the method of: filtering hop IP address information of a front k, if the hops are the k If no IP address is deleted, the IP address information in the U ADDR i is deleted by the probability p. In the process, at least one IP address in the ADDR i is guaranteed.
  • Step 5 For the remaining addresses in the destination address group, if a single destination address can be determined by using the single address selection method, the packet is forwarded to the destination address by using the general route forwarding method. Otherwise, the destination address group is encapsulated. The data packet is forwarded according to its corresponding port, and a single destination address is determined hop by hop by a single address selection method until reaching the destination address.
  • the single address selection method includes the following steps: if only one IP address or only one address satisfies the destination address matching condition, the address is selected as the destination address; if multiple addresses currently meet the destination address matching condition, the selection is based on the selection
  • the policy selects an address that satisfies the destination address matching condition as the destination address.
  • the selection strategy is: random selection, selection according to link prediction information or subnet load selection.
  • the destination address matching condition is: defining an address matching scoring function under the current routing mechanism as G(x), where x is an IP address, G(x) is an IP address matching length; when G(x)>L, If L is a preset matching length threshold, it indicates that the address x has reached the access destination address matching condition, that is, the data packet reaches the edge access network.
  • a single address selection method cannot determine a single destination address, that is, there are multiple addresses at present, but multiple addresses do not satisfy the address matching condition, only the destination address group can be encapsulated, and the data packet is forwarded according to the corresponding port. Use the single address selection method on the next port to continue looking for the destination address.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computing Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开一种基于可编程网络技术的多宿主网络的路由转发方法，所述方法包括：当进行数据包的路由转发时，根据转发流表地址匹配方法进行地址到端口的匹配，根据所匹配的端口对地址进行归类，使得每个端口都对应一个目的地址组，基于链路信息表中的信息和端口选择函数，选择合适的端口和相应的目的地址组，裁剪掉其余端口的目的地址组；利用地址过滤函数以及数据携带的前跳信息对剩下的目的地址组进行进一步裁剪；对于目的地址组中剩余的地址，如果采用单一地址选取方法能够确定单一目的地址，则将其作为目的地址，采用通用路由转发方法将报文转发至目的地址，否则封装目的地址组，采用单一地址选取方法逐跳确定单一目的地址，直至到达目的地址。

Description

一种基于可编程网络技术的多宿主网络路由转发方法
相关申请
本申请要求名称为“一种基于可编程网络技术的多宿主网络的路由转发方法”、于2017年11月28日提交的中国专利申请号为201711213654.2的优先权，在此通过引用包括该件申请。
技术领域
本发明涉及计算机网络技术，特别涉及一种基于可编程网络技术的多宿主网络路由转发方法。
背景技术
当前用户的终端通常会处在多个宿主网络中，比如Wi-Fi和4G网络。用户可以通过多个宿主网络并行传输来提高传输的效率，同时也可以根据宿主网络的变化选择状态好的网络，提高传输可靠性。但是目前的4G方案是选择其中一种宿主进行通信，这就无法享受到多宿主带来的可靠性、加倍传输和根据网络状态选择较优的宿主等优点。所以，面对下一代5G网络，有必要实现多宿主功能。
目前的多宿主的实现方式主要有两种，一种是地址和标志符分离的方法，比如LISP，HIP，LIN6和SHIM6等，这些方案有希望解决移动性和多宿主，但代价是修改终端设备或者部署新的网络实体(映射系统或者专用的边缘设备)。另一种是传输层的方案，比如SCTP和MPTCP，这些方法支持并发多路径传输，但是对移动性的支持并不好。已有的方案主要是端到端的多宿主传输方案，这可以让用户可以同时使用多个宿主网络，提高了服务质量。但是，端到端的多宿主方案中端系统所知道的网络信息是有限的，不能够很好地根据网络的状态及时实现多宿主路径的切换。
发明内容
本发明的目的在于解决传统的多宿主技术不能及时根据网络状态调整多宿主网络的问题，考虑到可编程网络设备能够实现协议独立转发，增加对数据包操作的灵活性的优势；提出了基于可编程网络设备的多目的地址的多宿主网络技术 方案，多个目的地址可以让网络根据多个通往目的地址的路径的状态来选择链路状态好的链路进行传输，平衡网络中的流量，解决网络拥塞等问题。通过基于网络状态动态调整的多宿主网络路由转发方法，可以有效提高用户的通信质量。
为了解决上述问题，本发明提出了一种基于可编程网络技术的多宿主网络的路由转发方法，所述方法包括：所述多宿主网络包括控制器和可编程网络设备；所述控制器用于根据网络的状态对可编程网络设备中的信息进行维护，所述可编程网络设备用于维护了两张表：转发流表和链路信息表，并对数据包进行修改和转发；所述方法包括：
当进行数据包的路由转发时，根据转发流表地址匹配方法进行地址到端口的匹配，根据所匹配的端口对地址进行归类，使得每个端口都对应一个目的地址组，
基于链路信息表中的信息和端口选择函数，选择合适的端口和相应的目的地址组，裁剪掉其余端口对应的目的地址组；
利用地址过滤函数以及数据携带的前跳信息对剩下的目的地址组进行进一步裁剪；
对于目的地址组中剩余的地址，如果采用单一地址选取方法能够确定单一目的地址，则将其作为目的地址，采用通用路由转发方法将报文转发至目的地址，否则封装目的地址组，根据其对应的端口将数据包转发，采用单一地址选取方法逐跳确定单一目的地址，直至到达目的地址。
作为上述方法的一种改进，所述可编程网络设备为在SDN架构下具有处理自定义报文和协议无关转发能力的网络设备；所述可编程网络设备通过对Openflow协议扩展，在转发层实现转发流表对自定义报文的处理和控制；所述可编程网络设备为支持P4，POF或PIF技术的路由器或交换机。
作为上述方法的一种改进，所述转发流表为报文转发所使用的流表，记录数据报文到目的地的转发信息；所述控制器通过路由协议获得转发流表并下发到可编程网络设备；所述链路信息表记录可编程网络设备中的每个转发端口port k和对应相关链路的状态信息值status k,1，status k,2……，所述状态信息值包括下一跳或多跳的链路状态和可能的路径的链路状态，所述链路状态包括下一跳或者多跳的端口负载、链路带宽和延迟。
作为上述方法的一种改进，所述控制器是网络策略管控中心，管理着网络的路由和转发策略，由SDN控制器经扩展得到；所述控制器包括转发流表控制模块和链路信息表控制模块；
所述转发流表控制模块，用于根据路由规则向可编程网络设备下发对转发流表的表项进行创建、修改和删除的命令；
所述链路信息表控制模块，用于通过对网络状态信息进行获取、计算和学习后，向可编程网络设备下发对链路信息表的表项进行创建、修改和删除的命令。
作为上述方法的一种改进，所述单一地址选择方法为：如果只有一个地址或者只有一个地址满足目的地址匹配条件，则选择该地址为目的地址；如果当前有多个地址达到了目的地址匹配条件，则依据选择策略选择一个满足目的地址匹配条件的地址作为目的地址。
作为上述方法的一种改进，所述方法具体包括：
步骤1)每个数据包携带用户多个地址以及路由到当前地址的前k跳信息，若用户有n个宿主，则对应的多个地址的集合为{addr 1,addr 2,…,addr n}；
步骤2)对数据包进行路由转发时，根据转发流表地址匹配方法进行地址到端口的匹配，根据所匹配的端口对地址进行归类，使得每个端口都对应一个目的地址组；
步骤3)查询所有端口在链路信息表中的信息，并通过端口选择函数选择状态好的端口作为下一跳的转发端口，裁剪掉没被选中的端口所对应的目的地址组；
步骤4)通过地址过滤函数及数据携带的前跳信息对剩下的目的地址组进行进一步裁剪；如果得到单一目的地址，转入步骤6)；否则，转入步骤5)；所述单一目的地址是指只有一个地址或者多个地址中只有一个地址满足目的地址匹配条件；
步骤5)如果有多个地址满足目的地址匹配条件，则依据选择策略选择一个满足目的地址匹配条件的地址作为的单一目的地址，转入步骤6)；否则，封装地址组，根据对应端口将数据包转发，转入步骤3)；
步骤6)将获取的单一目的地址作为目的地址，采用通用路由转发方法转发报文，直至到达目的地。
作为上述方法的一种改进，所述步骤2)具体包括：
步骤201)将包含多个地址数据包进行转发流表匹配，每个地址根据转发流表地址匹配方法匹配对应的转发端口，对有n个地址{addr 1,addr 2,…,addr n}的数据包和m个端口的可编程网络设备{port 1,port 2,…,port m}进行匹配，对于每个地址有映射关系Mapping(addr i)＝port j，其中，1≤i≤n,1≤j≤m，且port m为默认端口；则每个地址都会根据转发流表匹配到一个端口；
步骤202)对地址根据匹配到的转发端口进行归类；每个端口所对应的地址归类到一个地址组，其映射关系为{port 1:[addr 11,addr 12,…],port 2:[addr 21,addr 22,…],…,port m:[addr m1,addr m2,…]}。
作为上述方法的一种改进，所述转发流表地址匹配方法为：根据数据包的地址或标识匹配出对应的转发端口，不同路由机制有不同的匹配方法，包括：IP地址匹配和ICN中的名字匹配。
作为上述方法的一种改进，所述步骤3)的链路信息表的端口选择函数为port x＝F(Z,pack)＝F 2(F 1(Z,Pack))，其中，Z为链路信息表的端口与链路信息表之间的映射关系Z＝{port 1:[status 1,1,status 1,2……],port 2:[status 2,1,status 2,2……],…,port m:[status m,1,status m,2……]}，Pack为报文内容，F(x)为基于状态信息的端口选择函数；
所述端口选择函数F(x)，包含状态计算函数F 1(x)和选择函数F 2(x)；所述状态计算F 1(x)通过链路信息表中的某一项的值，或者根据多项的值计算，从而为每个候选端口计算状态值；所述选择函数F 2(x)则是根据计算得到的状态值选择下一跳端口，选择状态值最高的端口作为下一跳转发端口。
作为上述方法的一种改进，所述步骤4)的地址过滤函数用于对端口选择函数所选出的端口port i对应的地址组ADDR i进行过滤，实现过程为：根据前k跳地址过滤信息，如果在k跳之内都没有地址删除，则以概率p删除u个ADDR i中的地址信息，在该过程中，保证ADDR i中至少要有一个地址。
作为上述方法的一种改进，所述目的地址匹配条件为：计算当前路由机制下地址匹配打分函数为G(x)，其中x为地址，G(x)根据不同路由机制定义的不同匹配方法得到的得分，表示与目的地址的匹配程度；当G(x)>w时，w为预先设定或者动态设定的阈值，表明该地址x达到了目的地址匹配条件。
作为上述方法的一种改进，所述步骤5)的选择策略为：随机选择、根据链路预测信息选择或子网负载选择。
本发明的优势在于：
本发明的系统及方法能够根据网络的动态变化及时调整数据包的传输路径，使用户能够充分享受到多宿主带来的可靠性、加倍传输和根据网络状态选择较优的宿主等优点。
附图说明
图1是本发明的基于可编程网络技术的多目的地址的多宿主网路的结构图；
图2是本发明的基于可编程网络技术的多宿主网络中交换机的结构图和处理多目的地址数据包的流程图。
具体实施方式
现结合附图以基于POF(Protocol-Oblivious Forwarding)的多目的IP地址为例对本发明作进一步描述。
本发明提出了一种基于可编程网络设备的多宿主网络的路由转发方法，在该方法中，发送端每个数据包携带多个IP地址，以及路由到当前地址的前k跳信息；而在可编程网络设备的维护信息中，维护了两张信息表，一个是基于IP路由转发机制得到的转发流表，另一个是链路信息表，其中链路信息表提供可编程网络设备中每个转发端口和对应相关链路的状态信息。当进行数据包的路由转发时，首先根据转发流表IP地址匹配方法进行地址到端口的匹配，根据所匹配的端口对地址进行归类，使得每个端口都对应一个目的地址组，基于状态表中的信息和端口选择函数，选择合适的端口和相应的目的地址组，裁剪掉对应其余端口的地址组；利用IP地址过滤函数以及数据携带的前跳信息对剩下的地址组进行进一步裁剪；根据地址组中的地址，判断是否达到了IP接入网络地址匹配条件，如果没达到，则封装地址组，根据对应端口将数据包转发，如果达到，则根据单一地址选择方法，选出一个地址作为目的地址，删除其它地址，封装数据包，并根据对应端口转发。上述过程在每一跳转发都需判断和执行，直至到达目的地。
所述可编程网络设备是指在SDN(软件定义网络)架构下，具有处理自定义报文和协议无关转发能力的网络设备。通过对Openflow协议扩展，在转发层实现流表对自定义报文的处理和控制。可编程网络设备包括但不限于支持P4，POF，PIF等技术的路由器、交换机。每个可编程网络设备可支持多地址。所述的可编程网络，是由可编程网络设备相互连接构成的网络。
如图1所示，所述多宿主网络是指终端通过多个接入网连接网络，每个终端可以通过多个网络发送数据，可以同时从多个网络接收数据，当终端作为接收端时，可以从不同接入网络接收到同一个终端所发送出来的数据。网络能够根据自身的状态为终端动态选择较优的接入网提供服务。多宿主网络基于SDN架构，主要由控制器和可编程网络设备构成。其中，控制器主要是根据网络的状态对可编程网络设备中的信息进行维护，而可编程网络设备主要负责完成对数据包的修改 和转发。
所述转发流表IP地址匹配方法，是指根据数据包中的IP地址匹配出对应的转发端口，可用IP地址匹配方法进行匹配。
所述转发流表指的报文转发所使用的流表，流表即SDN中用于转发数据报文的表，其中维护着数据报文到目的地的转发信息，是控制器通过路由协议来获得并下发到可编程网络设备。所述链路信息表维护着链路的状态信息，其信息包含端口负载、当前链路状态，如带宽、延迟，以及与网络状态相关的量。
所述控制器是网络策略管控中心，管理着网络的路由和转发策略，可由SDN控制器经扩展而得到，控制器通过南向接口向可编程网络设备发送控制命令，可编程网络设备执行这些命令。其结构包括转发流表控制模块和链路信息表控制模块。转发流表控制模块主要根据获得的网络状态运行一些路由算法生成路由规则并通过南向接口向可编程网络设备下发对转发流表的流表项进行创建、修改和删除的命令；而链路信息表控制模块在对网络状态信息的获取、计算和学习后通过南向接口向可编程网络下发对链路信息表的表项进行创建、修改和删除的命令，如根据网络的动态变化和对以往网络状态的学习来获得和预测网络中链路的状态，从而对可编程网络设备中链路信息表进行维护。
链路信息表的信息是控制器将信息维护策略下发给可编程网络设备，可编程网络设备通过统计信息、网络的测量以及对网络动态变化的情况学习等方式获得相关信息并维护，链路信息表反映网络的链路动态变化。其中，端口port k对应状态信息为status k,1，status k,2……，其值包括但不限于下一跳或多跳的链路状态、可能的路径的链路状态等，例如:下一跳或者多跳的端口负载、链路带宽、延迟或其他网络相关信息，可通过测量、预测等方法获得。
如图2所示，所述路由转发方法具体包括以下流程：
步骤1)每个数据包携带用户多个地址，这些地址是用户所在的宿主网络所分配，假设用户有n个宿主，则对应的多个地址的集合为{IP 1,IP 2,…,IP n}。
步骤2)将包含多个地址数据包进行转发流表匹配，每个地址根据转发流表地址匹配方法匹配对应的转发端口，对有n个地址{IP 1,IP 2,…,IP n}的数据包和m个端口的可编程网络设备{port 1,port 2,…,port m}进行匹配，对于每个地址有映射关系Mapping(IP i)＝port j，其中，1≤i≤n,1≤j≤m，且port m为默认端口。则每个地址都会根据转发流表匹配到一个端口。
步骤3)对地址根据匹配到的转发端口进行归类；每个端口所对应的地址归 类到一个地址组，其映射关系为{port 1:[IP 11,IP 12,…],port 2:[IP 21,IP 22,…],…,port m:[IP m1,IP m2,…]}。
步骤4)查询所有端口在链路信息表中的信息，并通过端口选择函数选择状态好的端口作为下一跳的转发端口，删除没被选中的端口所对应的地址组；同时，考虑路由过程中前k跳的地址过滤信息，通过地址过滤函数，将所选择转发端口所对应的目的地址组进行地进一步目的地址过滤，裁剪掉没被选中的端口所对应的目的地址组；
基于链路信息表的端口选择函数为port x＝F(Z,pack)＝F 2(F 1(Z,Pack))，其中Z为链路信息表的端口与链路信息表之间的映射关系Z＝{port 1:[status 1,1,status 1,2……],port 2:[status 2,1,status 2,2……],…,port m:[status m,1,status m,2……]}，Pack为报文内容，F(x)为基于状态信息的端口选择函数。状态值的计算函数为F 1(x)，这里为对应端口的历史负载，输出为根据所定义的函数计算得到的状态值。通过F 2(x)选择得到合适转发端口port x，用于下一跳转发。
所述的端口选择函数F(x)，包含状态计算F 1(x)和选择F 2(x)两个步骤。状态值的计算函数为y＝F 1(x)＝load(x)，表示对应端口x的历史负载。选择步骤F 2(y)则是根据F 1计算得到的状态值y选择下一跳端口，这里F 2的计算方法为根据F 1(x)状态值由高到低对端口进行排序，选择状态值最高的端口作为下一跳转发端口。
所述地址过滤函数，其步骤包括：对端口选择函数所选出的端口port i对应的地址组ADDR i进行判断，其方法为：根据前k跳IP地址过滤信息，如果在k跳之内都没有IP地址删除，则以概率p删除u个ADDR i中的IP地址信息，在该过程中，保证ADDR i中至少要有一个IP地址。
步骤5)对于目的地址组中剩余的地址，如果采用单一地址选取方法能够确定单一目的地址，则将其作为目的地址，采用通用路由转发方法将报文转发至目的地址，否则，封装目的地址组，根据其对应的端口将数据包转发，采用单一地址选取方法逐跳确定单一目的地址，直至到达目的地址。
所述单一地址选择方法，其步骤包括：如果只有一个IP地址或者只有一个地址满足目的地址匹配条件，则选择该地址为目的地址；如果当前有多个地址达到了目的地址匹配条件，则依据选择策略选择一个满足目的地址匹配条件的地址作为目的地址。所述选择策略为：随机选择、根据链路预测信息选择或子网负载选择。
所述的目的地址匹配条件是指：定义当前路由机制下地址匹配打分函数为G(x)，其中x为IP地址，G(x)为IP地址匹配长度；当G(x)>L时，L为预先设定的匹配长度阈值，则说明该地址x达到了接入目的地址匹配条件，也即数据报文到达了边缘的接入网络。
当采用单一地址选取方法不能够确定单一目的地址，也就是当前有多个地址，但是多个地址都不满足地址匹配条件，则只能封装目的地址组，根据其对应的端口将数据包转发，在下一个端口利用单一地址选取方法继续寻找目的地址。
最后所应说明的是，以上实施例仅用以说明本发明的技术方案而非限制。尽管参照实施例对本发明进行了详细说明，本领域的普通技术人员应当理解，对本发明的技术方案进行修改或者等同替换，都不脱离本发明技术方案的精神和范围，其均应涵盖在本发明的权利要求范围当中。

Claims (12)

  1. 一种基于可编程网络技术的多宿主网络的路由转发方法，所述多宿主网络包括控制器和可编程网络设备；所述控制器用于根据网络的状态对可编程网络设备中的信息进行维护，所述可编程网络设备用于维护了两张表：转发流表和链路信息表，并对数据包进行修改和转发；所述方法包括：
    当进行数据包的路由转发时，根据转发流表地址匹配方法进行地址到端口的匹配，根据所匹配的端口对地址进行归类，使得每个端口都对应一个目的地址组，
    基于链路信息表中的信息和端口选择函数，选择合适的端口和相应的目的地址组，裁剪掉其余端口对应的目的地址组；
    利用地址过滤函数以及数据携带的前跳信息对剩下的目的地址组进行进一步裁剪；
    对于目的地址组中剩余的地址，如果采用单一地址选取方法能够确定单一目的地址，则将其作为目的地址，采用通用路由转发方法将报文转发至目的地址，否则封装目的地址组，根据其对应的端口将数据包转发，采用单一地址选取方法逐跳确定单一目的地址，直至到达目的地址。
  2. 根据权利要求1所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述可编程网络设备为在SDN架构下具有处理自定义报文和协议无关转发能力的网络设备；所述可编程网络设备通过对Openflow协议扩展，在转发层实现转发流表对自定义报文的处理和控制；所述可编程网络设备为支持P4，POF或PIF技术的路由器或交换机。
  3. 根据权利要求1-2之一所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述转发流表为报文转发所使用的流表，记录数据报文到目的地的转发信息；所述控制器通过路由协议获得转发流表并下发到可编程网络设备；所述链路信息表记录可编程网络设备中的每个转发端口port k和对应相关链路的状态信息值status k,1，status k,2……，所述状态信息值包括下一跳或多跳的链路状态和可能的路径的链路状态，所述链路状态包括下一跳或者多跳的端口负载、链路带宽和延迟。
  4. 根据权利要求1-2之一所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述控制器是网络策略管控中心，管理着网络的路由和转发策略，由SDN控制器经扩展得到；所述控制器包括转发流表控制模块和链 路信息表控制模块；
    所述转发流表控制模块，用于根据路由规则向可编程网络设备下发对转发流表的表项进行创建、修改和删除的命令；
    所述链路信息表控制模块，用于通过对网络状态信息进行获取、计算和学习后，向可编程网络设备下发对链路信息表的表项进行创建、修改和删除的命令。
  5. 根据权利要求4所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述单一地址选择方法为：如果只有一个地址或者只有一个地址满足目的地址匹配条件，则选择该地址为目的地址；如果当前有多个地址达到了目的地址匹配条件，则依据选择策略选择一个满足目的地址匹配条件的地址作为目的地址。
  6. 根据权利要求5所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述方法具体包括：
    步骤1)每个数据包携带用户多个地址以及路由到当前地址的前k跳信息，若用户有n个宿主，则对应的多个地址的集合为{addr 1,addr 2,…,addr n}；
    步骤2)对数据包进行路由转发时，根据转发流表地址匹配方法进行地址到端口的匹配，根据所匹配的端口对地址进行归类，使得每个端口都对应一个目的地址组；
    步骤3)查询所有端口在链路信息表中的信息，并通过端口选择函数选择状态好的端口作为下一跳的转发端口，裁剪掉没被选中的端口所对应的目的地址组；
    步骤4)通过地址过滤函数及数据携带的前跳信息对剩下的目的地址组进行进一步裁剪；如果得到单一目的地址，转入步骤6)；否则，转入步骤5)；所述单一目的地址是指只有一个地址或者多个地址中只有一个地址满足目的地址匹配条件；
    步骤5)如果有多个地址满足目的地址匹配条件，则依据选择策略选择一个满足目的地址匹配条件的地址作为的单一目的地址，转入步骤6)；否则，封装地址组，根据对应端口将数据包转发，转入步骤3)；
    步骤6)将获取的单一目的地址作为目的地址，采用通用路由转发方法转发报文，直至到达目的地。
  7. 根据权利要求6所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述步骤2)具体包括：
    步骤201)将包含多个地址数据包进行转发流表匹配，每个地址根据转发流表地址匹配方法匹配对应的转发端口，对有n个地址{addr 1,addr 2,…,addr n}的数据包和m个端口的可编程网络设备{port 1,port 2,…,port m}进行匹配，对于每个地址有映射关系Mapping(addr i)＝port j，其中，1≤i≤n,1≤j≤m，且port m为默认端口；则每个地址都会根据转发流表匹配到一个端口；
    步骤202)对地址根据匹配到的转发端口进行归类；每个端口所对应的地址归类到一个地址组，其映射关系为{port 1:[addr 11,addr 12,…],port 2:[addr 21,addr 22,…],…,port m:[addr m1,addr m2,…]}。
  8. 根据权利要求1或7所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述转发流表地址匹配方法为：根据数据包的地址或标识匹配出对应的转发端口，不同路由机制有不同的匹配方法，包括：IP地址匹配和ICN中的名字匹配。
  9. 根据权利要求6或7所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述步骤3)的链路信息表的端口选择函数为port x＝F(Z,pack)＝F 2(F 1(Z,Pack))，其中，Z为链路信息表的端口与链路信息表之间的映射关系Z＝{port 1:[status 1,1,status 1,2……],port 2:[status 2,1,status 2,2……],…,port m:[status m,1,status m,2……]}，Pack为报文内容，F(x)为基于状态信息的端口选择函数；
    所述端口选择函数F(x)，包含状态计算函数F 1(x)和选择函数F 2(x)；所述状态计算F 1(x)通过链路信息表中的某一项的值，或者根据多项的值计算，从而为每个候选端口计算状态值；所述选择函数F 2(x)则是根据计算得到的状态值选择下一跳端口，选择状态值最高的端口作为下一跳转发端口。
  10. 根据权利要求9所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述步骤4)的地址过滤函数用于对端口选择函数所选出的端口port i对应的地址组ADDR i进行过滤，实现过程为：根据前k跳地址过滤信息，如果在k跳之内都没有地址删除，则以概率p删除u个ADDR i中的地址信息，在该过程中，保证ADDR i中至少要有一个地址。
  11. 根据权利要求5或6所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述目的地址匹配条件为：计算当前路由机制下地址匹配打分函数为G(x)，其中x为地址，G(x)根据不同路由机制定义的不同匹配方法得到的得分，表示与目的地址的匹配程度；当G(x)>w时，w为预先设定或者动态 设定的阈值，表明该地址x达到了目的地址匹配条件。
  12. 根据权利要求5或6所述的基于可编程网络技术的多宿主网络的路由转发方法，其特征在于，所述步骤5)的选择策略为：随机选择、根据链路预测信息选择或子网负载选择。
PCT/CN2017/117700 2017-11-28 2017-12-21 一种基于可编程网络技术的多宿主网络路由转发方法 WO2019104788A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
US16/760,737 US11522784B2 (en) 2017-11-28 2017-12-21 Routing and forwarding method for multi-homed network based on programmable network technology
JP2020524190A JP6921322B2 (ja) 2017-11-28 2017-12-21 プログラマブルネットワーク技術に基づくマルチホストネットワークのルーティング転送方法
AU2017441267A AU2017441267B2 (en) 2017-11-28 2017-12-21 Routing and forwarding method for multi-homed network based on programmable network technology
EP17933683.9A EP3709583B1 (en) 2017-11-28 2017-12-21 Programmable network technology-based multi-homed network routing and forwarding method
KR1020207015402A KR102342723B1 (ko) 2017-11-28 2017-12-21 프로그래머블 네트워크 기술 기반의 멀티 홈드 네트워크 라우팅 및 포워딩 방법

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711213654.2A CN109842574B (zh) 2017-11-28 2017-11-28 一种基于可编程网络技术的多宿主网络路由转发方法
CN201711213654.2 2017-11-28

Publications (1)

Publication Number Publication Date
WO2019104788A1 true WO2019104788A1 (zh) 2019-06-06

Family

ID=66664281

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/117700 WO2019104788A1 (zh) 2017-11-28 2017-12-21 一种基于可编程网络技术的多宿主网络路由转发方法

Country Status (7)

Country Link
US (1) US11522784B2 (zh)
EP (1) EP3709583B1 (zh)
JP (1) JP6921322B2 (zh)
KR (1) KR102342723B1 (zh)
CN (1) CN109842574B (zh)
AU (1) AU2017441267B2 (zh)
WO (1) WO2019104788A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114553773A (zh) * 2020-11-24 2022-05-27 中国科学院声学研究所 一种层次结构的标识寻址方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110602230A (zh) * 2019-09-20 2019-12-20 重庆洲雄科技有限责任公司 一种报案者案情信息的转发方法及相关设备
US11240153B1 (en) * 2020-07-31 2022-02-01 Cisco Technology, Inc. Scoring policies for predictive routing suggestions
CN114500374B (zh) * 2020-11-12 2023-07-14 中国科学院声学研究所 一种采用随路改变目的地址的报文路由转发方法
US11949604B2 (en) * 2021-10-06 2024-04-02 Inernational Business Machines Corporation Integrated network switch operation
CN114047881B (zh) * 2021-11-18 2023-11-17 江苏省未来网络创新研究院 一种基于用户策略的网络数据包存储装置及方法
CN115150324B (zh) * 2022-06-09 2024-05-10 南京邮电大学 一种基于可编程数据平面实现可变重路由阈值的方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103220255A (zh) * 2012-01-18 2013-07-24 中兴通讯股份有限公司 一种实现单播反向路径转发urpf检查的方法及装置
CN104243317A (zh) * 2014-09-26 2014-12-24 杭州华三通信技术有限公司 一种实现ip路由转发的方法和装置
CN105337857A (zh) * 2015-11-23 2016-02-17 北京邮电大学 一种基于软件定义网络的多路径传输方法
EP3125472A1 (en) * 2015-07-30 2017-02-01 Alcatel Lucent Telecommunication system, method and computer readable medium to control how a transmission of packets of a data flow is realized
CN104184664B (zh) * 2014-08-05 2017-07-04 新华三技术有限公司 路由转发表项生成方法及装置

Family Cites Families (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7164653B1 (en) * 2002-06-07 2007-01-16 At&T Corp. Shared bandwidth reservation in PL, ATM, FR and IP/MPLS networks
US7284068B1 (en) * 2002-06-20 2007-10-16 Cisco Technology, Inc. Technique for providing multi-homing capability in data networks
JP4356557B2 (ja) * 2003-09-02 2009-11-04 Ｋｄｄｉ株式会社 インターネットにおけるネットワークの障害位置を特定する方法及びプログラム
US8260922B1 (en) * 2005-09-16 2012-09-04 Cisco Technology, Inc. Technique for using OER with an ECT solution for multi-homed sites
KR100859712B1 (ko) * 2006-12-08 2008-09-23 한국전자통신연구원 위조된 멀티캐스트 패킷을 막는 장치 및 그 방법
US20080320116A1 (en) * 2007-06-21 2008-12-25 Christopher Briggs Identification of endpoint devices operably coupled to a network through a network address translation router
JP5520928B2 (ja) * 2008-03-31 2014-06-11 オランジュ ネットワークにおけるデータパケットのルーティング方法および関連デバイス
JP5638063B2 (ja) * 2010-03-23 2014-12-10 キヤノン株式会社 通信装置、通信装置の制御方法、プログラム
US9160707B2 (en) * 2010-10-22 2015-10-13 Telefonaktiebolaget L M Ericsson (Publ) Differentiated handling of network traffic using network address translation
CN102158497B (zh) * 2011-05-11 2014-01-22 中国人民解放军国防科学技术大学 Ip地址过滤方法及装置
JP2014068258A (ja) * 2012-09-26 2014-04-17 Oki Electric Ind Co Ltd 制御装置
CN103118083B (zh) * 2013-01-21 2015-12-02 北京交通大学 一种业务报文转发的方法和装置
US9900252B2 (en) * 2013-03-08 2018-02-20 A10 Networks, Inc. Application delivery controller and global server load balancer
WO2014147197A1 (en) * 2013-03-20 2014-09-25 Wolting Holding B.V. Compiler for and method for software defined networks
JP5954236B2 (ja) * 2013-03-28 2016-07-20 日立金属株式会社 ネットワーク中継装置
US9178816B1 (en) * 2013-09-27 2015-11-03 Juniper Networks, Inc. Control plane messaging in all-active multi-homed ethernet virtual private networks
US9485180B2 (en) * 2014-02-04 2016-11-01 Juniper Networks, Inc. Loop free alternate selection for multi-homed networks
US9537751B2 (en) 2014-03-31 2017-01-03 Kulcloud Divided hierarchical network system based on software-defined networks
US11057342B2 (en) * 2014-12-09 2021-07-06 Telefonaktiebolaget L M Ericsson (Publ) Network address translation
US9787573B2 (en) * 2014-12-31 2017-10-10 Juniper Networks, Inc. Fast convergence on link failure in multi-homed Ethernet virtual private networks
US10469379B2 (en) * 2017-02-17 2019-11-05 Cisco Technology, Inc. System and method to facilitate content delivery to multiple recipients in a network environment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103220255A (zh) * 2012-01-18 2013-07-24 中兴通讯股份有限公司 一种实现单播反向路径转发urpf检查的方法及装置
CN104184664B (zh) * 2014-08-05 2017-07-04 新华三技术有限公司 路由转发表项生成方法及装置
CN104243317A (zh) * 2014-09-26 2014-12-24 杭州华三通信技术有限公司 一种实现ip路由转发的方法和装置
EP3125472A1 (en) * 2015-07-30 2017-02-01 Alcatel Lucent Telecommunication system, method and computer readable medium to control how a transmission of packets of a data flow is realized
CN105337857A (zh) * 2015-11-23 2016-02-17 北京邮电大学 一种基于软件定义网络的多路径传输方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3709583A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114553773A (zh) * 2020-11-24 2022-05-27 中国科学院声学研究所 一种层次结构的标识寻址方法
CN114553773B (zh) * 2020-11-24 2022-11-22 中国科学院声学研究所 一种层次结构的标识寻址方法

Also Published As

Publication number Publication date
EP3709583B1 (en) 2023-02-15
AU2017441267B2 (en) 2021-04-29
CN109842574A (zh) 2019-06-04
US11522784B2 (en) 2022-12-06
JP2021501527A (ja) 2021-01-14
CN109842574B (zh) 2020-07-17
EP3709583A4 (en) 2021-01-20
KR20200071767A (ko) 2020-06-19
AU2017441267A1 (en) 2020-05-14
US20200358692A1 (en) 2020-11-12
EP3709583A1 (en) 2020-09-16
JP6921322B2 (ja) 2021-08-18
KR102342723B1 (ko) 2021-12-23

Similar Documents

Publication Publication Date Title
WO2019104788A1 (zh) 一种基于可编程网络技术的多宿主网络路由转发方法
US11134012B2 (en) Communication system, communication device, controller, and method and program for controlling forwarding path of packet flow
EP3468117B1 (en) Service function chaining (sfc)-based packet forwarding method, device and system
AU2002238410B2 (en) Addressing and routing in wireless mesh networks
US10587494B2 (en) Network control method and apparatus
EP3711264A1 (en) Dynamic link state routing protocol
JP5488979B2 (ja) コンピュータシステム、コントローラ、スイッチ、及び通信方法
KR101457317B1 (ko) 라우팅 정보 업데이트의 우선 순위 지정
CN106656793B (zh) 一种sdn网络与ip网络的数据交互方法
WO2020108587A1 (zh) 数据处理方法、控制器和转发设备
JP6011619B2 (ja) 移動通信端末、通信方法、通信システムおよび制御装置
CN112615749A (zh) 一种基于VXLAN和OpenFlow的流量调度系统和方法
KR101786623B1 (ko) 소프트웨어 정의 네트워크에서 브로드캐스트를 처리하는 방법, 장치 및 컴퓨터 프로그램
Yu et al. Dynamic load balancing multipathing in data center ethernet
KR101786616B1 (ko) 소프트웨어 정의 네트워크에서 서브넷을 지원하는 방법, 장치 및 컴퓨터 프로그램
Khang et al. Performance Evaluation of Wireless Routing Protocols: RIP vs OSPF
CN110401733A (zh) 一种sdn网络的arp协议实现方法、系统及控制器
JP5464360B2 (ja) 移動体通信における改良された端末情報管理方式及び通信方式を実現するメッシュ型ネットワーク及び基地局
KR20080060882A (ko) 적응형 패킷 크기 기반 라우팅 방법 및 라우팅 장치

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: 17933683

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020524190

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2017441267

Country of ref document: AU

Date of ref document: 20171221

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20207015402

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2017933683

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2017933683

Country of ref document: EP

Effective date: 20200611