WO2022100060A1 - 一种采用随路改变目的地址的报文路由转发方法 - Google Patents

一种采用随路改变目的地址的报文路由转发方法 Download PDF

Info

Publication number
WO2022100060A1
WO2022100060A1 PCT/CN2021/096321 CN2021096321W WO2022100060A1 WO 2022100060 A1 WO2022100060 A1 WO 2022100060A1 CN 2021096321 W CN2021096321 W CN 2021096321W WO 2022100060 A1 WO2022100060 A1 WO 2022100060A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
network
node
address
forwarding
Prior art date
Application number
PCT/CN2021/096321
Other languages
English (en)
French (fr)
Inventor
王劲林
尤佳莉
程钢
孙鹏
Original Assignee
中国科学院声学研究所
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 filed Critical 中国科学院声学研究所
Priority to EP21890591.7A priority Critical patent/EP4231553A4/en
Priority to JP2023528534A priority patent/JP2023549514A/ja
Publication of WO2022100060A1 publication Critical patent/WO2022100060A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/74Address processing for routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/12Shortest path evaluation
    • H04L45/122Shortest path evaluation by minimising distances, e.g. by selecting a route with minimum of number of hops
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00Network arrangements, protocols or services for addressing or naming
    • H04L61/45Network directories; Name-to-address mapping
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/28Flow control; Congestion control in relation to timing considerations

Definitions

  • the invention belongs to the technical field of computer networks, and in particular, relates to a message routing and forwarding method that adopts the method of changing the destination address with the path.
  • the present invention proposes a packet routing and forwarding method that uses a path-dependent change of destination address, which can improve network transmission performance and respond to the perceived network state in time.
  • the present invention provides a packet routing and forwarding method using the path-dependent changing of destination address, the method comprising: in a network using the routing and forwarding method using the path-dependent changing destination address,
  • All network element entities in the network include but are not limited to hosts, devices, data and services.
  • Each network element entity is identified by a unique and immutable identifier ID.
  • the identifier ID of each network element entity corresponds to the corresponding network address.
  • the mapping relationship between them is maintained by a service level-based resolution service system;
  • the service level refers to a differentiated service defined based on measurement, and the definition includes but is not limited to service level defined based on network measurement response delay, service level defined based on geographic distance measurement, semantic correlation measurement and Calculate the service level defined, and the service level defined based on any combination of two or three of the above information;
  • the network routing switching device can rewrite the network address field and network address extension field of the message; Forwards packets and forwards them based on the forwarding table, which can be calculated and maintained by using the location-based routing method;
  • the data message is forwarded in a network layer message format, and the data message includes: a network address field, a network address extension field, an ID identification field that remains unchanged during the session, and a field that carries the application's preference for transmission requirements; wherein , the preference field includes: application-related specific parameters; the application-related specific parameters include but are not limited to service quality requirements and application preferences.
  • the parsing service system includes one or more service levels, which maintains the mapping relationship between the identifier ID and the corresponding network address, and A distributed query service system that can provide resolution services of multiple service levels;
  • the resolution service system is a distributed system with a hierarchical structure based on service levels; each distributed node constituting the distributed system maintains one or more neighbor node lists, and the one or more neighbor node lists in the Nodes cooperatively respond to the same parsing request;
  • mapping relationship between all identifier IDs and corresponding network addresses is registered and managed in the resolution service system
  • the analysis service system includes but is not limited to registration service capability, retrieval service capability, query service capability and management service capability.
  • the distributed system based on the hierarchical structure of service levels, the distributed system is a single-layer structure or a multi-layer structure;
  • Nodes at the same level in the distributed system have the same service level, and nodes at different levels have different service levels;
  • a serviceable node of a user at a certain service level indicating that the user can obtain the resolution query service on the service node with the service performance corresponding to the service level;
  • a service node owns a certain user, which means that the service node can provide the user with a service performance analysis query service corresponding to the service level.
  • a user node has one or more serviceable nodes belonging to the same service level.
  • a serviceable node then the service level has one and only one service node as the master point;
  • all master points of each service level serving the user conform to the nested structure, and all users owned by the user at the master point of service level j must also be users owned by the user higher than the master point of service level j ;
  • ⁇ (ms i, j ) is the set of users served by the master point ms i, j , when service level k > service At level j, then
  • the master point is determined by selecting the function definition or assignment method.
  • the neighbor node list is a local autonomous node list maintained by the node itself, and the node selects some service nodes in the network based on the neighbor determination criterion to construct the neighbor node list ;
  • the neighborhood judgment criteria include but are not limited to:
  • the neighborhood is constructed based on the network delay measurement criterion. If the delay between the node and the candidate neighborhood node is less than a certain threshold, the candidate neighborhood node is added to the neighborhood node list;
  • the neighborhood is constructed based on the geographic location distance criterion. If the physical distance between the node and the candidate neighborhood node is less than the preset distance threshold, the candidate neighborhood node is added to the neighborhood node list;
  • the neighborhood is constructed based on the semantic similarity criterion. If the stored content of the node and the candidate neighborhood node is highly correlated, the candidate neighborhood node will be added to the neighborhood node list; among them, the correlation refers to obtaining two parameters by semantic calculation. The similarity of the content stored by each node;
  • the neighborhood is established based on the structural relationship criterion. For example, all service nodes form a tree structure, and the descendant nodes or descendant nodes in the tree structure form the neighborhood.
  • the nodes in the one or more neighbor node lists cooperate to respond to the same parsing request; the nodes in the neighbor node list maintained by each node can replace the node itself or be parallel to the node Provide services;
  • judgment conditions for alternative or parallel services include but are not limited to:
  • the node itself is overloaded and it is difficult to respond to the request under the service level requirements, and the neighboring nodes that have the same service capability as themselves can provide services that meet the level requirements for the request;
  • the node itself does not store the required content, and the relevant neighboring nodes have the required content and can provide the service level required by the user;
  • multiple neighbor nodes can meet the service requirements at the same time, and can perform request offloading.
  • the information of the network address extension field includes but is not limited to: multiple candidate storage address set information or reference information of the corresponding content identified by the destination ID, multiple information under the same device ID Candidate network address set information or its reference information, and network layer processing indication information; wherein, the reference information includes but is not limited to pointers and offsets.
  • Step 1) according to the ID identification carried by the data message, request the resolution service that meets the on-site response service level requirement from the resolution service system, obtain a new destination address under the ID identification, and rewrite the network of the data message with the new destination address. address field, obtain the rewritten network address field;
  • Step 2) According to the network state perceived by the network routing switching device itself and the network address and network address extension field information that may be carried in the data message, obtain the candidate address through the candidate address set information in the network address extension field or its reference information. gather;
  • Step 3 According to the network state perceived by the network routing switching device itself and the network address extension field information that may be carried in the data packet, based on a pre-policy, calculate part of the field information in the new network address extension field, and rewrite the data packet. Obtain the rewritten NA Expanded field from the corresponding part of the network address extension field in the text.
  • the message of the rewritten network address field and the rewritten network address extension field is reconstructed into a message format supported by the current network, and forwarded according to the forwarding table of the new target address.
  • the method of the invention can optimize and adjust the routing forwarding process in time through the network-perceived state information, and select an appropriate forwarding path for each hop through on-site selection and calculation of the forwarding path, thereby improving the overall routing and forwarding performance of the network.
  • FIG. 1 is a flowchart of a method for routing and forwarding a message by changing the destination address with the path according to the present invention.
  • the present invention provides a message routing and forwarding method that adopts the method of changing the destination address with the path.
  • the method is specifically:
  • All network element entities in the network including but not limited to, hosts, devices, data and services, are identified by a unique and unchanging identifier ID.
  • the identifier ID of each network element entity is associated with the corresponding network address.
  • the mapping relationship between them is maintained by a service level-based parsing service system;
  • the service level refers to the differentiated service defined based on the measurement
  • the definition includes but not limited to the service level defined based on the network measurement response delay, the service level defined based on the geographic distance measurement, and the semantic correlation measurement and calculation definition. , and a service level defined based on any combination of two or three of the above information.
  • the resolution service system includes one or more service levels, and is a distributed query service system that maintains the identifier ID and the corresponding network address mapping relationship and can provide resolution services of multiple service levels;
  • service level refers to the differentiated service defined based on measurement, which includes but is not limited to service level defined based on network measurement response delay, service level defined based on geographic distance, service level defined based on semantic correlation, and various combinations of any two or three of the above service levels;
  • the mapping relationship between all ID identifiers and corresponding network addresses is registered and managed in the resolution service system.
  • the resolution system includes, but is not limited to, registration, retrieval, management and query service capabilities.
  • the parsing service system is a distributed parsing system based on a hierarchical structure of service levels, which can be a single-layer structure or a multi-layer structure;
  • Nodes at the same level have the same level of service, while nodes at different levels have different levels of service.
  • a service node of a certain service level can provide a service to a user, indicating the correspondence between the identifier ID of the user and the network address, which can be queried on the service node with the service performance corresponding to the service level.
  • Each distributed node constituting the system maintains one or more neighbor node lists, and the nodes in the one or more neighbor node lists can cooperatively respond to the same parsing request;
  • the neighbor node list is a local autonomous node list maintained by the node itself, and the node selects some service nodes in the network to construct the neighbor node list based on the neighbor determination criterion;
  • Multiple neighbor node lists can be constructed if multiple criteria are passed.
  • the neighborhood determination criteria include but are not limited to:
  • the neighborhood is constructed based on the network delay measurement criterion. If the delay between the node and the candidate neighborhood node is less than a certain threshold, the candidate neighborhood node is added to the neighborhood node list;
  • the neighborhood is constructed based on the geographic location distance criterion. If the physical distance between the node and the candidate neighborhood node is less than the preset distance threshold, the candidate neighborhood node is added to the neighborhood node list;
  • the neighborhood is constructed based on the semantic similarity criterion. If the stored content of the node and the candidate neighborhood node is highly correlated, the candidate neighborhood node will be added to the neighborhood node list; among them, the correlation refers to obtaining two parameters by semantic calculation. The similarity of the content stored by each node;
  • the neighborhood is established based on the structural relationship criterion. For example, all service nodes form a tree structure, and the descendant nodes or descendant nodes in the tree structure form the neighborhood.
  • the nodes in one or more neighbor node lists cooperate to respond to the same parsing request; the nodes in the neighbor node list maintained by each node can provide services in place of the node itself or in parallel with the node;
  • judgment conditions for alternative or parallel services include but are not limited to:
  • the node itself is overloaded and it is difficult to respond to the request under the service level requirements, and the neighboring nodes that have the same service capabilities as themselves can provide services that meet the level requirements for the request;
  • the node itself does not store the required content, and the relevant neighboring nodes have the required content and can provide the service level required by the user;
  • multiple neighboring nodes can meet the service requirements at the same time, and can perform request offloading.
  • a user node has one or more serviceable nodes belonging to the same service level. If there are multiple serviceable nodes in a certain service level, then the service level has one and only one service node as the master point;
  • ⁇ (ms i, j 0 is the set of users served by the master point ms i, j , when service level k > service At level j,
  • the master point can be determined by selecting the function definition or assignment method.
  • the data message is forwarded in a network layer message format, and the data message includes: a network address field (NA address field), a network address extension field (NA Expanded field); an ID identification field that remains unchanged during the session; and It may carry a preference (Preference) field for the transmission requirement of the application, where the preference field contains specific parameters related to the application, and the specific parameters include but are not limited to service quality requirements and application preferences.
  • NA address field a network address field
  • NA Expanded field network address extension field
  • ID identification field that remains unchanged during the session
  • It may carry a preference (Preference) field for the transmission requirement of the application, where the preference field contains specific parameters related to the application, and the specific parameters include but are not limited to service quality requirements and application preferences.
  • both the NA address field and the NA Expanded field can be dynamically rewritten during the routing and forwarding process.
  • the network routing switching device can rewrite the NA address field and NA Expanded field of the message;
  • the specific process of rewriting the NA address field and the NA Expanded field includes:
  • Step 1) according to the ID identification carried by the data message, request the resolution service that meets the on-site response service level requirement from the resolution service system, obtain the new destination address under the ID identification, and rewrite the NA of the data message with the new destination address. address field, obtain the rewritten NA address field;
  • Step 2) according to the network state perceived by the network routing switching device itself and the network address and network address extension field information that may be carried in the data message, obtain the candidate address collection through the candidate address collection information or reference information of the NA Expanded field;
  • Step 3 According to the network state perceived by the network routing switching device itself and the NA Expanded field information that may be carried in the data message, based on the pre-policy, calculate part of the field information in the new NA Expanded field, and rewrite the data in the data message. The corresponding part in the NA Expanded field of , obtain the rewritten NA Expanded field.
  • the information of the NA Expanded field includes but is not limited to: multiple candidate storage address set information or its reference information of the content corresponding to the destination ID, multiple candidate network address set information under the same device ID or its reference information, and network Layer processing instructions.
  • the reference information includes but is not limited to pointers and offsets.
  • the network routing switching device reconstructs the message to form a forwarding message, and forwards it based on the forwarding table of the new destination address.
  • the routing method of the location is calculated and maintained, eg IP protocol.
  • the message reconstruction is to reconstruct the message with the rewritten NA address field or the NA Expanded field into a message format supported by the current network, and forward it according to the forwarding table of the new target address.
  • All network element entities in the network including: hosts, devices, data and services, are identified by a unique and immutable identifier ID, and IP is used as a network address for routing.
  • the resolution service system maintains the mapping relationship between the ID identifier of the network element entity and the corresponding network address (ie IP address). For different service levels, the parsing service system provides parsing services with parsing delays of d 1 ms, d 2 ms, d 3 ms...different service levels.
  • the network routing switching device is a switch; the steps of address rewriting and forwarding in the transmission process are shown in Figure 1:
  • S(X,Y,Z) is the selection function
  • X is the changed candidate node set CTS'
  • Y is the parameter set R
  • Z is the network state information set NS, such as port load, link bandwidth, etc. ;
  • tip is to finally select the IP address that meets the requirements of the selection function as the forwarding address of the next hop, that is, the new destination address.
  • S(X, Y, Z) is to select the address corresponding to the minimum port load as the next hop address;
  • DEL(X,Y,Z) is the candidate address list clipping function, which can be defined according to service requirements.
  • the nodes that do not meet the performance requirements will be deleted from the changed candidate address set CTS' set, and then the trimmed candidate node list will be used to replace the previous list.
  • DEL(X,Y,Z) is to delete all candidate nodes whose port load is greater than the threshold THL.

Landscapes

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

Abstract

本发明属于计算机网络技术领域,具体地说,涉及一种采用随路改变目的地址的报文路由转发方法,包括:在使用所述采用随路改变目的地址的路由转发方法的网络中,每个网元实体是通过唯一且不变的ID进行标识,每个网元实体的标识ID与对应的网络地址之间的映射关系,是由一个基于服务等级的解析服务系统维护;网络路由交换设备在接收到网络层数据报文后,可以改写该报文的网络地址字段和网络地址扩展字段;依据改写后的报文网络地址字段和网络地址扩展字段,通过报文重构形成转发报文,并基于转发表进行转发,转发表可以采用基于位置的路由方法进行计算和维护。

Description

一种采用随路改变目的地址的报文路由转发方法
相关申请的交叉参考
该申请要求2020年11月12日提交的中国专利申请号为2020112598384的专利申请的优先权,和2021年3月16日提交的中国专利申请号为2021102817089的专利申请的优先权,上述两个专利申请在此被完全引入作为参考。
技术领域
本发明属于计算机网络的技术领域,具体地说,涉及一种采用随路改变目的地址的报文路由转发方法。
背景技术
传统网络的报文路由转发方法中,路由算法将路径计算好后,后续转发过程中无法进行路径的动态修改,然而,网络状态是时刻变化的,前一时刻合适的路径,在下一时刻可能已经发生了拥塞,当前的方法则无法及时对路径进行调整和绕行,从而常常导致显著的服务性能下降。
随着SDN的发展,采用控制面和数据面分离的方式,使得网络状态的感知和控制更加灵活,网络感知的能力也在逐渐提升,同时,移动通信、内容高效分发、多宿主网络传输等场景,带来了报文传输过程中的多地址和变地址处理需求,如何使报文在转发过程中进行动态的地址变化从而改变传输路径,是一个至关重要的问题。
发明内容
为解决现有技术存在的上述缺陷,本发明提出了一种采用随路改变目的地址的报文路由转发方法,能够提高网络传输性能,从而及时对感知到的网络状态进行响应。
本发明提供的一种采用随路改变目的地址的报文路由转发方法,该方法包括:在使用所述采用随路改变目的地址的路由转发方法的网络中,
网络中所有网元实体包括但不限于主机、设备、数据和服务,每个网元实体是通过唯一且不变的标识符ID进行标识,每个网元实体的标识符ID与对应的网络地址之间的映射关系,是由一个基于服务等级的解析服务系统维护;
其中,所述服务等级是指基于测量而定义的差异性服务,该定义包括但不限 于基于网络测量的响应时延定义的服务等级、基于地理距离测量定义的服务等级、基于语义相关性测量和计算定义的服务等级、以及基于上述任意两种或着三种信息组合定义的服务等级;
网络路由交换设备在接收到网络层数据报文后,可以改写该报文的网络地址字段和网络地址扩展字段;依据改写后的报文网络地址字段和网络地址扩展字段,通过报文重构形成转发报文,并基于转发表进行转发,转发表可以采用基于位置的路由方法进行计算和维护;
其中,所述数据报文以网络层报文格式转发,该数据报文包括:网络地址字段,网络地址扩展字段,会话期间保持不变的ID标识字段和携带应用对传输需求的偏好字段;其中,该偏好字段包括:应用相关的特定参数;该应用相关的特定参数包括但不限于服务质量需求和应用偏好。
作为上述技术方案的改进之一,所述基于服务等级的解析服务系统中,该解析服务系统包括一个或多个服务等级,是一个维护标识符ID和对应的网络地址之间的映射关系、并可提供多个服务等级的解析服务的分布式查询服务系统;
该解析服务系统为基于服务等级的层次型结构的分布式系统;构成该分布式系统的每个分布式节点均维护一张或多张邻域节点列表,并且所述一张或多张邻域节点列表中的节点协同响应同一个解析请求;
所有标识符ID标识和对应的网络地址之间的映射关系,在解析服务系统中进行注册和管理;
其中,所述解析服务系统包括但不限于注册服务能力、检索服务能力、查询服务能力和管理服务能力。
作为上述技术方案的改进之一,所述基于服务等级的层次型结构的分布式系统,该分布式系统为单层结构或多层结构;
所述分布式系统中的同一层次的节点,具有相同的服务等级,而不同层的节点具有不同的服务等级;
一个用户在某个服务等级上的可服务节点,表示该用户可以在该服务节点上以该服务等级对应的服务性能获得解析查询服务;
一个服务节点拥有某个用户,表示该服务节点可对该用户提供该服务等级对应的服务性能的解析查询服务。
作为上述技术方案的改进之一,所述基于服务等级的层次型结构的分布式系统中,一个用户节点具有一个或者多个属于同服务等级的可服务节点,如果在某个服务等级有多个可服务节点,则该服务等级有且只有一个服务节点为master点;
其中,服务于该用户的各服务等级所有master点符合嵌套结构,该用户在服务等级j的master点所拥有的所有用户,也必须是该用户高于服务等级j的master点所拥有的用户;
假设用户为u i,ms i,j是为u i在服务等级j所选择的master点,Φ(ms i,j)为master点ms i,j所服务的用户集合,当服务等级k>服务等级j时,则
Figure PCTCN2021096321-appb-000001
其中,master点通过选择函数定义或指派方式确定。
作为上述技术方案的改进之一,所述邻域节点列表是一个由节点自身维护的局部自治的节点列表,该节点基于邻域判定准则,选出部分网络中的服务节点构建成邻域节点列表;
如果通过多个准则,则可以构建多个邻域节点列表;
其中,邻域判定准则包含但不限于:
基于网络时延测量准则构建邻域,如节点与候选邻域节点间时延小于某个阈值,则将该候选邻域节点加入邻域节点列表;
基于地理位置距离准则构建邻域,如节点与候选邻域节点间物理距离小于预先设定的距离阈值,则将该候选邻域节点加入邻域节点列表;
基于语义相近程度准则构建邻域,如节点与候选邻域节点中存储内容相关性较强,则将该候选邻域节点加入邻域节点列表;其中,相关性是指通过语义计算的方式获得两个节点所存储内容的相近程度;
基于结构关系准则建立邻域,如所有服务节点构成一个树形结构,将该树形结构中的子孙节点或者后代节点构成邻域。
作为上述技术方案的改进之一,所述一张或多张邻域节点列表中的节点协同响应同一个解析请求;每个节点所维护的邻域节点列表中的节点,可代替节点自身或与节点并行提供服务;
其中,可替代性或者并行服务的判定条件,包括但不限于:
节点自身负载过重难以在服务等级要求下响应请求,且与自身有相同服务能力的邻域节点可对该请求提供符合等级要求的服务;
节点自身没有存储所需内容,且相关邻域节点具有该所需内容并可提供用户所需服务等级的服务;
高并发请求时,多个邻域节点可同时满足服务要求,可执行请求分流。
作为上述技术方案的改进之一,所述网络地址扩展字段,其信息包括但不限于:目的ID标识对应内容的多个候选存储地址集合信息或其引用信息、同一个设备ID标识下的多个候选网络地址集合信息或其引用信息,和网络层处理指示信息;其中,所述引用信息包括但不限于指针和偏移量。
作为上述技术方案的改进之一,所述数据报文的网络地址字段和网络地址扩展字段改写的具体过程为:
步骤1)根据数据报文携带的ID标识,向解析服务系统请求符合现场响应服务等级要求的解析服务,获取该ID标识下新的目的地址,并用该新的目的地址改写该数据报文的网络地址字段,获得改写后的网络地址字段;
步骤2)根据网络路由交换设备自身感知的网络状态和该数据报文中可能携带的网络地址和网络地址扩展字段信息,通过网络地址扩展字段中的候选地址集合信息或其引用信息,得到候选地址集合;
该候选地址集合与根据ID标识向解析服务系统查询得到的地址构成的集合中,选取部分子集,并重新排序,根据预先策略,选出合适地址作为新的目的地址,将其填入该数据报文的网络地址字段中,并对所选取的部分子集进行筛选,将筛选后的内容写回网络地址扩展字段中的相应位置;
步骤3)根据网络路由交换设备自身感知的网络状态和该数据报文中可能携带的网络地址扩展字段信息,基于预先策略,计算新的网络地址扩展字段中的部分字段信息,并改写该数据报文中的网络地址扩展字段中的相应部分,获得改写后的NA Expanded字段。
作为上述技术方案的改进之一,所述的报文重构的具体过程为:
将改写后的网络地址字段和改写后的网络地址扩展字段的报文重新构建成当前网络所支持的报文格式,并按照新的目标地址的转发表进行转发。
本发明与现有技术相比的有益效果是:
本发明的方法能够通过网络感知的状态信息,对路由转发过程进行及时优化和调整,通过转发路径的现场选择和计算,为每一跳选择合适的转发路径,从而提高网络的整体路由转发性能。
附图说明
图1是本发明的一种采用随路改变目的地址的报文路由转发方法的流程图。
具体实施方式
现结合附图对本发明作进一步的描述。
如图1所示,本发明提供一种采用随路改变目的地址的报文路由转发方法,在使用所述采用随路改变目的地址的报文路由转发方法的网络中,其方法具体为:
(1)网络中所有网元实体包括但不限于,主机、设备、数据和服务,是通过唯一且不变的标识符ID进行标识,每个网元实体的标识符ID与对应的网络地址之间的映射关系,是由一个基于服务等级的解析服务系统维护;
其中,服务等级是指基于测量而定义的差异性服务,该定义包括但不限于基于网络测量的响应时延定义的服务等级、基于地理距离测量定义的服务等级、基于语义相关性测量和计算定义的服务等级、以及基于上述任意两种或三种信息组合定义的服务等级。
(2)对于上述基于服务等级的解析服务系统中,其特点如下:
该解析服务系统包括一个或多个服务等级,是一个维护标识符ID和对应的网络地址映射关系、并可提供多个服务等级的解析服务的分布式查询服务系统;
其中,服务等级是指基于测量而定义的差异性服务,该定义包括但不限于基于网络测量的响应时延定义的服务等级、基于地理距离定义的服务等级、基于语义相关性定义的服务等级、以及上述任意两个或三个服务等级的多种组合;
所有ID标识和对应的网络地址之间的映射关系,在解析服务系统中进行注册和管理。解析系统包括但不限于注册、检索、管理和查询服务能力。
该解析服务系统为基于服务等级的层次型结构的分布式解析系统,可为单层结构,也可为多层结构;
同一层次的节点,具有相同的服务等级,而不同层的节点具有不同的服务等级。
某个服务等级的服务节点可对一个用户提供服务,表示该用户的标识符ID到网络地址的对应关系,可以在该服务节点上以该服务等级对应的服务性能查询到。
构成该系统的每个分布式节点都维护一张或多张邻域节点列表,并且所述一张或多张邻域节点列表中的节点可以协同响应同一个解析请求;
在这里,所述邻域节点列表是一个由节点自身维护的局部自治的节点列表,该节点基于邻域判定准则,选出部分网络中的服务节点构建成邻域节点列表;
如果通过多个准则,则可以构建多个邻域节点列表。
其中邻域判定准则包含但不限于:
基于网络时延测量准则构建邻域,如节点与候选邻域节点间时延小于某个阈值,则将该候选邻域节点加入邻域节点列表;
基于地理位置距离准则构建邻域,如节点与候选邻域节点间物理距离小于预先设定的距离阈值,则将该候选邻域节点加入邻域节点列表;
基于语义相近程度准则构建邻域,如节点与候选邻域节点中存储内容相关性较强,则将该候选邻域节点加入邻域节点列表;其中,相关性是指通过语义计算的方式获得两个节点所存储内容的相近程度;
基于结构关系准则建立邻域,如所有服务节点构成一个树形结构,将该树形结构中的子孙节点或者后代节点构成邻域。
除此之外,一张或多张邻域节点列表中的节点协同响应同一个解析请求;每个节点所维护的邻域节点列表中的节点,可代替节点自身或与节点并行提供服务;
其中,可替代性或者并行服务的判定条件,包括但不限于:
一、节点自身负载过重难以在服务等级要求下响应请求,且与自身有相同服务能力的邻域节点可对该请求提供符合等级要求的服务;
二、节点自身没有存储所需内容,且相关邻域节点具有该所需内容并可提供用户所需服务等级的服务;
三、高并发请求时,多个邻域节点可同时满足服务要求,可执行请求分流。
一个用户节点有一个或者多个属于同服务等级的可服务节点,如果在某个服务等级有多个可服务节点,则该服务等级有且只有一个服务节点为master点;
其中,服务于该用户的各服务等级所有master点需要符合嵌套结构,该用户在服务等级j的master点所拥有的所有用户,也必须是该用户高于服务等级j的master点所拥有的用户;
假设用户为u i,ms i,j是为u i在服务等级j所选择的master点,Φ(ms i,j0为master点ms i,j所服务的用户集合,当服务等级k>服务等级j时,
Figure PCTCN2021096321-appb-000002
其中,master点可以通过选择函数定义或指派方式确定。
所述数据报文以网络层报文格式转发,所述数据报文包括:网络地址字段(NA地址字段),网络地址扩展字段(NA Expanded字段);会话期间保持不变的ID标识字段;还可能携带应用对传输需求的偏好(Preference)字段,该偏好字段包含了应用相关的特定参数,该特定参数包括但不限于服务质量需求、应用偏好。
其中,NA地址字段和NA Expanded字段都可在路由转发过程中被动态改写。
(3)网络路由交换设备在接收到数据报文后,可以改写该报文的NA地址字段和NA Expanded字段;
其中,NA地址字段和NA Expanded字段改写的具体过程包括:
步骤1)根据数据报文携带的ID标识,向解析服务系统请求符合现场响应服务等级要求的解析服务,获取该ID标识下新的目的地址,并用该新的目的地址改写该数据报文的NA地址字段,获得改写后的NA地址字段;
步骤2)根据网络路由交换设备自身感知的网络状态和该数据报文中可能携带的网络地址和网络地址扩展字段信息,通过NA Expanded字段的候选地址集合信息或引用信息,得到候选地址集合;
该候选地址集合与根据ID标识向解析服务系统查询得到的地址构成的集合中,选取部分子集,并重新排序,根据预先策略,选出合适地址作为新的目的地址,将其填入该数据报文中的NA地址字段中,并对所选取的部分子集进行筛选,将筛选后的内容写回NA Expanded字段中的相应位置;
步骤3)根据网络路由交换设备自身感知的网络状态和该数据报文中可能携带的NA Expanded字段信息,基于预先策略,计算新的NA Expanded字段中的部分字段信息,并改写该数据报文中的NA Expanded字段中的相应部分,获得改写后的NA Expanded字段。
其中,NA Expanded字段,其信息包括但不限于:目的ID对应内容的多个候选存储地址集合信息或其引用信息、同一个设备ID下的多个候选网络地址集合信息或其引用信息,和网络层处理指示信息。其中,所述引用信息包括但不限于指针和偏移量。
(4)网络路由交换设备依据改写后的报文的NA字段和NA Expanded字段,通过报文重构形成转发报文,并基于新的目的地址的转发表进行转发,该转发表可以是采用基于位置的路由方法计算和维护的,例如IP协议。
具体地,报文重构是将改写后的NA地址字段或NA Expanded字段的报文重新构建成当前网络所支持的报文格式,并按照新的目标地址的转发表进行转发。
实施例1.
网络中所有网元实体,包括:主机、设备、数据和服务,是通过唯一且不变的标识符ID进行标识,用IP作为网络地址进行路由。
解析服务系统维护了网元实体的ID标识与对应的网络地址(即IP地址)之间的映射关系;其中,解析服务系统提供时延差异化的服务,通过网络测量得到的不同时延范围为不同的服务等级,该解析服务系统提供解析时延在d 1ms,d 2ms,d 3ms......不同服务等级的解析服务。
数据报文以网络层报文格式转发,所述数据报文包括:源地址s、目的地址ts,ID标识字段id,以及候选地址集合信息CTS={ts 1,ts 2….},其中,初始时,CTS为空。数据报文还携带应用需求信息,包含应用相关的特定参数R={r 1,r 2…..},该相关的特定参数包括但不限于服务质量需求、应用偏好。其中,网络路由交换设备为交换机;其传输过程中的地址改写和转发的步骤,如图1所示:
1)在每一次转发时,首先看当前节点地址是否是报文转发的目的地址,如果是,则停止转发,如果不是,则交换机通过解析服务系统对数据报文中携带的id进行解析查询,根据对交换机接收到的报文解析服务等级l的要求,交换机向解析服务系统中对应l服务等级的服务节点发出解析查询请求,解析服务节点需要 在l服务等级服务性能要求范围内返回结果,则该返回结果可能为返回空(即CIP=null),也可能返回一个IP地址集合,即候选地址集合CIP={cip1,cip2….cipn};其中,n>=1。
2)交换机接收到返回的候选地址集合CIP后,将CIP加入到候选地址字段CTS中,得到更改后的候选地址集合CTS’,即CTS’=CTS∪CIP,并根据选择函数,对其中地址进行选择,
tip=S(CTS’,R,NS)
其中,S(X,Y,Z)为选择函数,X值为更改后的候选节点集合CTS’,Y值为参数集合R,Z值为网络状态信息集合NS,如端口负载、链路带宽等;
tip为最终选择出符合选择函数要求的ip地址,作为下一跳的转发地址,即新的目的地址。其中,S(X,Y,Z)为选择最小端口负载所对应的地址作为下一跳地址;
计算候选节点列表CTS1=DEL(CTS,R,NS),
其中,DEL(X,Y,Z)为候选地址列表裁剪函数,该裁剪函数可根据服务需求定义,
如端口负载超出阈值的对应候选地址需裁剪,则将不符合性能要求的节点从更改后的候选地址集合CTS’集合中删除,之后,用裁剪后的候选节点列表代替之前列表。
其中,DEL(X,Y,Z)为删除端口负载大于阈值THL的所有候选节点。
3)通过新选出的目的地址,以及之前的源地址s和ID标识id,以及更改后的候选地址集合CTS’,重构具有当前网络所支持的报文格式的新的IP数据报文,并通过IP路由转发机制,基于IP路由协议,按照新的目的地址的转发表进行报文转发。
4)上述过程不断执行,直至到达目的地址ts。
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 (9)

  1. 一种采用随路改变目的地址的报文路由转发方法,其特征在于,该方法包括:在使用所述采用随路改变目的地址的路由转发方法的网络中,
    网络中所有网元实体包括但不限于主机、设备、数据和服务,每个网元实体是通过唯一且不变的标识符ID进行标识,每个网元实体的标识符ID与对应的网络地址之间的映射关系,是由一个基于服务等级的解析服务系统维护;
    其中,所述服务等级是指基于测量而定义的差异性服务,该定义包括但不限于基于网络测量的响应时延定义的服务等级、基于地理距离测量定义的服务等级、基于语义相关性测量和计算定义的服务等级、以及基于上述任意两种或着三种信息组合定义的服务等级;
    网络路由交换设备在接收到网络层数据报文后,可以改写该报文的网络地址字段和网络地址扩展字段;依据改写后的报文网络地址字段和网络地址扩展字段,通过报文重构形成转发报文,并基于转发表进行转发,转发表可以采用基于位置的路由方法进行计算和维护;
    其中,所述数据报文以网络层报文格式转发,该数据报文包括:网络地址字段,网络地址扩展字段,会话期间保持不变的ID标识字段和携带应用对传输需求的偏好字段;其中,该偏好字段包括:应用相关的特定参数;该应用相关的特定参数包括但不限于服务质量需求和应用偏好。
  2. 根据权利要求1所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述基于服务等级的解析服务系统中,该解析服务系统包括一个或多个服务等级,是一个维护标识符ID和对应的网络地址之间的映射关系、并可提供多个服务等级的解析服务的分布式查询服务系统;
    该解析服务系统为基于服务等级的层次型结构的分布式系统;构成该分布式系统的每个分布式节点均维护一张或多张邻域节点列表,并且所述一张或多张邻域节点列表中的节点协同响应同一个解析请求;
    所有标识符ID标识和对应的网络地址之间的映射关系,在解析服务系统中进行注册和管理;
    其中,所述解析服务系统包括但不限于注册服务能力、检索服务能力、查询服务能力和管理服务能力。
  3. 根据权利要求2所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述基于服务等级的层次型结构的分布式系统,该分布式系统为单层结构或多层结构;
    所述分布式系统中的同一层次的节点,具有相同的服务等级,而不同层的节点具有不同的服务等级;
    一个用户在某个服务等级上的可服务节点,表示该用户可以在该服务节点上以该服务等级对应的服务性能获得解析查询服务;
    一个服务节点拥有某个用户,表示该服务节点可对该用户提供该服务等级对应的服务性能的解析查询服务。
  4. 根据权利要求2或3所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述基于服务等级的层次型结构的分布式系统中,一个用户节点具有一个或者多个属于同服务等级的可服务节点,如果在某个服务等级有多个可服务节点,则该服务等级有且只有一个服务节点为master点;
    其中,服务于该用户的各服务等级所有master点符合嵌套结构,该用户在服务等级j的master点所拥有的所有用户,也必须是该用户高于服务等级j的master点所拥有的用户;
    假设用户为u i,ms i,j是为u i在服务等级j所选择的master点,Φ(ms i,j)为master点ms i,j所服务的用户集合,当服务等级k>服务等级j时,则
    Figure PCTCN2021096321-appb-100001
    其中,master点通过选择函数定义或指派方式确定。
  5. 根据权利要求2所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述邻域节点列表是一个由节点自身维护的局部自治的节点列表,该节点基于邻域判定准则,选出部分网络中的服务节点构建成邻域节点列表;
    如果通过多个准则,则可以构建多个邻域节点列表;
    其中,邻域判定准则包含但不限于:
    基于网络时延测量准则构建邻域,如节点与候选邻域节点间时延小于某个阈值,则将该候选邻域节点加入邻域节点列表;
    基于地理位置距离准则构建邻域,如节点与候选邻域节点间物理距离小于预 先设定的距离阈值,则将该候选邻域节点加入邻域节点列表;
    基于语义相近程度准则构建邻域,如节点与候选邻域节点中存储内容相关性较强,则将该候选邻域节点加入邻域节点列表;其中,相关性是指通过语义计算的方式获得两个节点所存储内容的相近程度;
    基于结构关系准则建立邻域,如所有服务节点构成一个树形结构,将该树形结构中的子孙节点或者后代节点构成邻域。
  6. 根据权利要求2所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述一张或多张邻域节点列表中的节点协同响应同一个解析请求;每个节点所维护的邻域节点列表中的节点,可代替节点自身或与节点并行提供服务;
    其中,可替代性或者并行服务的判定条件,包括但不限于:
    节点自身负载过重难以在服务等级要求下响应请求,且与自身有相同服务能力的邻域节点可对该请求提供符合等级要求的服务;
    节点自身没有存储所需内容,且相关邻域节点具有该所需内容并可提供用户所需服务等级的服务;
    高并发请求时,多个邻域节点可同时满足服务要求,可执行请求分流。
  7. 根据权利要求1所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述网络地址扩展字段,其信息包括但不限于:目的ID标识对应内容的多个候选存储地址集合信息或其引用信息、同一个设备ID标识下的多个候选网络地址集合信息或其引用信息,和网络层处理指示信息;其中,所述引用信息包括但不限于指针和偏移量。
  8. 根据权利要求1所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述数据报文的网络地址字段和网络地址扩展字段改写的具体过程为:
    步骤1)根据数据报文携带的ID标识,向解析服务系统请求符合现场响应服务等级要求的解析服务,获取该ID标识下新的目的地址,并用该新的目的地址改写该数据报文的网络地址字段,获得改写后的网络地址字段;
    步骤2)根据网络路由交换设备自身感知的网络状态和该数据报文中可能携带的网络地址和网络地址扩展字段信息,通过网络地址扩展字段中的候选地址集 合信息或其引用信息,得到候选地址集合;
    该候选地址集合与根据ID标识向解析服务系统查询得到的地址构成的集合中,选取部分子集,并重新排序,根据预先策略,选出合适地址作为新的目的地址,将其填入该数据报文的网络地址字段中,并对所选取的部分子集进行筛选,将筛选后的内容写回网络地址扩展字段中的相应位置;
    步骤3)根据网络路由交换设备自身感知的网络状态和该数据报文中可能携带的网络地址扩展字段信息,基于预先策略,计算新的网络地址扩展字段中的部分字段信息,并改写该数据报文中的网络地址扩展字段中的相应部分,获得改写后的NA Expanded字段。
  9. 根据权利要求1所述的采用随路改变目的地址的报文路由转发方法,其特征在于,所述的报文重构的具体过程为:
    将改写后的网络地址字段和改写后的网络地址扩展字段的报文重新构建成当前网络所支持的报文格式,并按照新的目标地址的转发表进行转发。
PCT/CN2021/096321 2020-11-12 2021-05-27 一种采用随路改变目的地址的报文路由转发方法 WO2022100060A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21890591.7A EP4231553A4 (en) 2020-11-12 2021-05-27 MESSAGE ROUTING AND FORWARDING METHOD WITH CHANGE OF DESTINATION ADDRESS USING AN ASSOCIATED CHANNEL
JP2023528534A JP2023549514A (ja) 2020-11-12 2021-05-27 経路に従って宛先アドレスを変更するメッセージのルーティング及び転送方法

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202011259838.4 2020-11-12
CN202011259838 2020-11-12
CN202110281708.9 2021-03-16
CN202110281708.9A CN114500374B (zh) 2020-11-12 2021-03-16 一种采用随路改变目的地址的报文路由转发方法

Publications (1)

Publication Number Publication Date
WO2022100060A1 true WO2022100060A1 (zh) 2022-05-19

Family

ID=81491521

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/096321 WO2022100060A1 (zh) 2020-11-12 2021-05-27 一种采用随路改变目的地址的报文路由转发方法

Country Status (4)

Country Link
EP (1) EP4231553A4 (zh)
JP (1) JP2023549514A (zh)
CN (1) CN114500374B (zh)
WO (1) WO2022100060A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115022242A (zh) * 2022-08-08 2022-09-06 深圳高灯计算机科技有限公司 网络层识别方法、装置、计算机设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105471747A (zh) * 2015-11-25 2016-04-06 武汉烽火网络有限责任公司 一种智能路由器选路方法和装置
WO2018183553A1 (en) * 2017-03-29 2018-10-04 Fungible, Inc. Non-blocking any-to-any data center network having multiplexed packet spraying within access node groups
CN110636001A (zh) * 2018-06-25 2019-12-31 华为技术有限公司 发送网络性能参数、计算网络性能的方法和网络节点
CN111555982A (zh) * 2020-07-10 2020-08-18 武汉绿色网络信息服务有限责任公司 一种基于IPv6扩展头的报文智能选路的方法和系统

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1622551A (zh) * 2004-12-15 2005-06-01 中国科学院计算机网络信息中心 层次式交换网络内部服务系统及其管理控制方法
CN102546356B (zh) * 2010-12-09 2015-01-28 中兴通讯股份有限公司 一种保证标识网中移动节点服务质量的方法及系统
CN104052667B (zh) * 2013-03-15 2017-05-31 华为技术有限公司 报文处理方法及设备
WO2015149351A1 (zh) * 2014-04-04 2015-10-08 华为技术有限公司 监控服务器、解析服务器、请求设备及节点选择方法
CN105704069B (zh) * 2014-11-26 2020-02-14 中兴通讯股份有限公司 一种节省网元内处理资源的方法及装置
CN109729514B (zh) * 2017-10-30 2020-07-17 中国科学院声学研究所 一种移动网络实体的动态位置信息的快速查询方法
CN109842574B (zh) * 2017-11-28 2020-07-17 中国科学院声学研究所 一种基于可编程网络技术的多宿主网络路由转发方法
JP7142106B2 (ja) * 2018-05-28 2022-09-26 中国科学院声学研究所 Icnメッセージの転送方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105471747A (zh) * 2015-11-25 2016-04-06 武汉烽火网络有限责任公司 一种智能路由器选路方法和装置
WO2018183553A1 (en) * 2017-03-29 2018-10-04 Fungible, Inc. Non-blocking any-to-any data center network having multiplexed packet spraying within access node groups
CN110636001A (zh) * 2018-06-25 2019-12-31 华为技术有限公司 发送网络性能参数、计算网络性能的方法和网络节点
CN111555982A (zh) * 2020-07-10 2020-08-18 武汉绿色网络信息服务有限责任公司 一种基于IPv6扩展头的报文智能选路的方法和系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4231553A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115022242A (zh) * 2022-08-08 2022-09-06 深圳高灯计算机科技有限公司 网络层识别方法、装置、计算机设备和存储介质
CN115022242B (zh) * 2022-08-08 2022-11-18 深圳高灯计算机科技有限公司 网络层识别方法、装置、计算机设备和存储介质

Also Published As

Publication number Publication date
EP4231553A4 (en) 2024-02-28
CN114500374B (zh) 2023-07-14
CN114500374A (zh) 2022-05-13
JP2023549514A (ja) 2023-11-27
EP4231553A1 (en) 2023-08-23

Similar Documents

Publication Publication Date Title
USRE49943E1 (en) System and method for a context layer switch
US11601522B2 (en) Data management in an edge network
US10298713B2 (en) Distributed content discovery for in-network caching
US9838333B2 (en) Software-defined information centric network (ICN)
US20130041982A1 (en) Method and node for acquiring content and content network
US20030200307A1 (en) System and method for information object routing in computer networks
WO2010118638A1 (zh) 一种基于地理位置信息编码的cdn网络系统及数据分发方法
WO2009155802A1 (zh) 选择服务提供实体的方法、系统、服务选择实体、服务管理实体
WO2012106918A1 (zh) 内容处理方法、装置和系统
US9407687B2 (en) Method, apparatus, and network system for acquiring content
CN108293023B (zh) 支持信息为中心的网络中的上下文感知的内容请求的系统和方法
US20160255047A1 (en) Methods and systems for determining domain names and organization names associated with participants involved in secured sessions
WO2009100637A1 (zh) 一种建立路由的方法和装置
WO2022100060A1 (zh) 一种采用随路改变目的地址的报文路由转发方法
US20150006571A1 (en) Method And Apparatus For Enabling Queries In An Information-Centric Network
CN109561355B (zh) 一种ccn/ndn内容注册、内容位置解析和内容路由的系统及方法
US11218841B2 (en) Fast query method for dynamic location information of mobile network entity
US20150098468A1 (en) Method of controlling routing of identifier/locator mapping service
WO2003023607A1 (en) System and method for information object routing in computer networks
CN111262785B (zh) 一种命名数据网络中的多属性概率缓存方法
CN114650296B (zh) 一种信息中心网络副本选择方法
CN108449281B (zh) 网络流量协同卸载方法及协同卸载控制器
CN113660162B (zh) 一种临近缓存感知的半集中式路由方法及其系统
CN108810169B (zh) 一种基于社会关系的内容缓存方法
WO2023097856A1 (zh) 一种icn网络中基于热度的随路协同缓存方法

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 2023528534

Country of ref document: JP

ENP Entry into the national phase

Ref document number: 2021890591

Country of ref document: EP

Effective date: 20230516

NENP Non-entry into the national phase

Ref country code: DE