CN1725761A - Method of implementing link aggregation of network equipment - Google Patents

Method of implementing link aggregation of network equipment Download PDF

Info

Publication number
CN1725761A
CN1725761A CN 200510076709 CN200510076709A CN1725761A CN 1725761 A CN1725761 A CN 1725761A CN 200510076709 CN200510076709 CN 200510076709 CN 200510076709 A CN200510076709 A CN 200510076709A CN 1725761 A CN1725761 A CN 1725761A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
port
link aggregation
physical
network layer
group
Prior art date
Application number
CN 200510076709
Other languages
Chinese (zh)
Other versions
CN100479465C (en )
Inventor
汪洪远
胡光
Original Assignee
杭州华为三康技术有限公司
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

Links

Abstract

A method for aggregating links in network device includes confirming n numbers of physical ports ( PP ) of link aggregation port ( AP ) set for carrying out the same configuration of network layer to generate n numbers of equivalent routes ( ER ) for specific route using link AP as outlet , setting outlet interface information to be n numbers of PP , distributing message to be retransmitted by network layer to n numbers of ER by network device uniformly for revising ER corresponding to interrupted PP to be normal operation PP when connection between partial PP in said port set and opposite end is interrupted . The method for aggregating links in link layer and between network switches is also disclosed.

Description

网络设备实现链路聚合的方法 A method of network device link aggregation

技术领域 FIELD

本发明涉及具有网络层和/或链路层转发功能的网络设备,尤其涉及上述网络设备中链路聚合的实现方法。 The present invention relates to a network device having a network layer and / or the link layer forwarding, and particularly to a method to achieve the above-described network device link aggregation.

背景技术 Background technique

随着互联网技术的发展和网络应用的普及,在网络上传输的数据越来越多,对传输时间和可靠性的要求也越来越高。 With the popularity of network applications and the development of Internet technology, data transmission over the network more and more demands on the transmission time and reliability are also increasing. 减小传输时间的一个方法是增加传输链路的带宽,近年来带宽已由10M(兆)增加到100M、1000M、10G(千兆)等等,但是带宽的增加并不能提高链路的可靠性。 One way to reduce the transmission time is to increase the bandwidth of the transmission link, the bandwidth in recent years has been 10M (megabytes) increased 100M, 1000M, 10G (gigabit), etc., but does not increase the bandwidth and improve the reliability of the link .

如果能够把多条链路当作一条链路使用,不但能够增加网络带宽,而且能够起到链路备份的作用。 If you can put multiple links as a link to use, not only can increase network bandwidth, but also can play the role of a backup link. 为此,IEEE(Institute of Electrical and ElectronicsEngineers,电气和电子工程师协会)专门制定了802.3ad标准,用以规范链路聚合(Aggregated Link)的实现。 For this reason, IEEE (Institute of and ElectronicsEngineers, Electrical Institute of Electrical and Electronics Engineers) standard 802.3ad specifically formulated for link aggregation specification (Aggregated Link) a.

所谓链路聚合,是将多条物理链路捆绑在一起形成一条逻辑上的链路,实现负载在各成员物理链路上的分担,同时也提供了更高的可靠性。 The so-called link aggregation, a plurality of physical links is tied to a logical link is formed, to achieve load sharing on each physical link members, while also providing greater reliability. 图1所示为链路聚合的示意图。 Figure 1 is a schematic diagram of link aggregation. 网络设备100的物理端口110、物理端口120和物理端口130分别与网络设备200的物理端口210、物理端口220和物理端口230相连接,形成3条物理链路。 A physical network device 100, port 110, physical port 120 and physical port 130, respectively 210, physical port 220 and physical port 230 is connected to the physical port of the network device 200, three physical links are formed. 在网络设备100上将物理端口110、120和130设置为链路聚合端口组,同时在网络设备200上将物理端口210、220和230也设置为链路聚合端口组,则上述3条物理链路组合成一条逻辑链路。 In the network device 100 on the physical port 110, 120 and 130 is set to the link aggregation group, while the physical port 200 on the network devices 210, 220 and 230 are also provided to the link aggregation group, the above three physical links Road combined into one logical link.

链路聚合的出现,使得网络组网更加灵活方便,在增加网络带宽的同时提高了网络的安全性。 Link aggregation occurs, so that more flexible networking is, an increase in network bandwidth while improving network security. 当链路聚合端口组中的某条物理链路出现问题,数据流量能够自动切换到其他物理链路,实现逻辑链路的保护功能。 When a physical link in the link aggregation group problems, data traffic can be automatically switched to a different physical link, logical link to achieve protection.

在现有技术中,链路聚合通常通过硬件芯片实现,这种实现方式需要将实现聚合算法的代码写入芯片的内存,占用大量的芯片内存空间,这些内存空间被这些代码固定占有,在运行的时候不会释放出来,因此增加了网络设备的成本。 In the prior art, typically by hardware link aggregation chip, such an implementation will need to write code to achieve aggregation algorithm memory chip, memory chips take up a lot of space, the memory space occupied these codes are fixed, running when not released, thereby increasing the cost of network equipment. 从性价比角度考虑,网络设备供应商倾向于提供数量有限的链路聚合端口组数目,用户无法根据实际组网需求任意设置链路聚合端口组。 From the perspective of cost, network equipment suppliers tend to provide a limited number of link aggregation group port number, the user can not arbitrarily set according to the link aggregation group networking requirements.

发明内容 SUMMARY

本发明要解决的技术问题是现有技术中链路聚合的实现方式成本过高的问题。 The present invention is to solve the technical problem of the prior art is the link aggregation high implementation cost.

本发明所述网络设备在网络层实现链路聚合的方法包括以下步骤:a)确定链路聚合端口组包括的n个物理端口,n大于等于2,将上述n个物理端口在网络层进行相同配置;b)为以所述链路聚合端口组为出口的某个路由生成n条等价路由,所述n条等价路由的出接口信息分别设置为所述链路聚合端口组的n个物理端口;c)网络设备将需要网络层转发的报文均匀分配到所述n个等价路由上,相应地以上述n个物理端口为报文出接口;d)当链路聚合端口组的部分物理端口与对端的连接中断后,将连接中断的物理端口所对应的等价路由的出接口信息修改为链路聚合端口组中正常工作的物理端口。 The present invention is in the process of the network device the network layer link aggregation comprises the steps of: a) determining the n-th link aggregation group comprising physical ports of ports, n is greater than or equal to 2, the n physical ports will be in the same network layer configuration; b) of the link aggregation group to generate n pieces of routes as a route outlet, the n equivalent routes the outbound interface information are provided to the link aggregation port group of n physical port; c) the network device required to forward the network layer packets to the uniform distribution of the n equivalent routes, corresponding to the n physical port is an outbound interface; D) when the link aggregation port group after part of the physical connection to the peer port interrupt, routes the connection interrupted equivalent physical port interface information corresponding to the link aggregation modify physical ports in the port group working.

优选地,所述方法还包括:将链路聚合端口组的每个物理端口设置为不转发其他物理端口接收的组播和广播报文。 Preferably, the method further comprising: a link aggregation group each physical port is a port not to forward the other physical port to receive multicast and broadcast packets.

优选地,所述步骤d)之后还包括:e)当链路聚合端口组连接中断的某个物理端口恢复正常工作后,查找所述n条等价路由中具有相同出接口信息的等价路由,将其中一条等价路由的出接口信息修改为所述恢复正常工作的物理端口。 Preferably, after said step d) further comprising: e) When a physical link aggregation group interruption of the connection port after the port to resume normal operation, the equivalent route lookup n equal-cost routes that have the same interface information , in which an equivalent route outbound interface information is modified to the physical port to restore normal operation.

优选地,所述步骤d)中对连接中断的物理端口所对应的等价路由的出接口信息的修改满足以下条件:任意两个正常工作的物理端口所对应的等价路由数目的差值小于等于1。 Preferably, said step d) to modify the interface information for routing connections interrupted equivalent physical port corresponding to the following conditions are met: the equivalent number of routes to work any two physical ports corresponding to the difference is smaller than equal to 1.

优选地,所述步骤a)与步骤b)之间包括下述至少一项:通过动态路由协议学习以所述网络层地址为出口的间接路由;配置以所述网络层地址为出口静态路由;通过链路聚合端口组收到的地址解析协议ARP报文,学习直接路由。 Preferably comprised between said step a) and step b) at least one of the following: to study the network layer address of an indirect route to the outlet by a dynamic routing protocol; configured to the network layer address is a static route outlet; by ARP link aggregation port group received ARP packet, learning direct route.

优选地,所述步骤a)之后还包括:a1)链路聚合端口组从收到的报文中学习以其为出口的媒介接入控制MAC地址;a2)将学习到的MAC地址的出接口信息设置为链路聚合端口组正常工作的物理端口;a3)当链路聚合端口组的部分物理端口与对端的连接中断后,删除以连接中断的物理端口为出接口的MAC地址,转步骤a1);a4)当链路聚合端口组连接中断的物理端口恢复正常工作后,删除以链路聚合端口组为出口的所有MAC地址,转步骤a1)。 Preferably, said step a) further comprises: a1) a link aggregation group to study its exit from the media access control (MAC) address of the received packet; A2) an interface to the learned MAC address link aggregation information to physical ports of the port group work; A3) when the rear portion of the physical port connection to the peer port interrupt link aggregation group, to delete a physical port connected to an interrupt interface MAC address, proceed to step a1 ); A4) when the link aggregation group of physical port connection is broken resume normal work, to remove the link aggregation group to export all MAC addresses, proceed to step a1).

优选地,所述方法还包括:设置链路聚合端口组的每个物理端口不转发其他物理端口接收的组播和广播报文。 Preferably, the method further comprising: setting each physical port link aggregation port group does not forward the other physical port to receive multicast and broadcast packets.

优选地,所述步骤a2)具体为:设定哈希算法,使其哈希值与正常工作的物理端口具有一一对应关系;用所述哈希算法计算所述MAC地址的哈希值;将所述MAC地址的出接口设置为计算所得的哈希值对应的物理端口。 Preferably, said step a2) is specifically: setting hash algorithm, the hash value so that a physical port and having a working one relationship; calculating a hash value of the MAC address with the hash algorithm; the outgoing interface is set to the MAC address corresponding to a hash value calculated physical port.

本发明还提供了一种在链路层实现链路聚合的方法,包括以下步骤: The present invention further provides a method implemented in the link layer link aggregation, comprising the steps of:

A.确定链路聚合端口组包括的物理端口;B.将链路聚合端口组的每个物理端口设置为不转发其他物理端口接收的组播和广播报文;C.链路聚合端口组从收到的报文中学习以其为出口的媒介接入控制MAC地址;D.将学习到的MAC地址的出接口信息设置为链路聚合端口组正常工作的物理端口;E.当链路聚合端口组的部分物理端口与对端的连接中断后,删除以连接中断的物理端口为出接口的MAC地址,转步骤C。 A. determining link aggregation group comprising physical ports of ports;. B each physical link aggregation port not to forward port group other physical port to receive multicast and broadcast packets;. C from the link aggregation group learning to its outlet a medium access control packet received MAC address;. D interface information provided to the learned MAC address to a physical port link aggregation port group of normal operation; E when the link aggregation. a rear portion connected to the remote physical port and port group interrupt, to delete a physical port connected to an interrupt interface MAC address, proceed to step C.

优选地,所述步骤E之后还包括:当链路聚合端口组连接中断的物理端口恢复正常工作后,删除以链路聚合端口组为出口的所有MAC地址,转步骤C;所述步骤D具体为:设定哈希算法,使其哈希值与正常工作的物理端口具有一一对应关系;用所述哈希算法计算所述MAC地址的哈希值;将所述MAC地址的出接口设置为计算所得的哈希值对应的物理端口。 Preferably, after said step E further comprises: when the link aggregation group the physical port connection is broken after the resumption of normal operation, the link aggregation group to remove all the MAC address port, proceed to step C; D DETAILED step of: hashing algorithm is set so that the hash value of the physical port and having a working one relationship; calculating a hash value of the MAC address with the hash algorithm; the outgoing interface of the MAC address provided is a physical port corresponding to the hash value is calculated.

本发明还提供了一种在网络层交换机之间实现链路聚合的方法,包括以下步骤:(1)在第一交换机上选取预定数量的物理端口,并对这些物理端口进行相目同的网络层配置;(2)在第二交换机上相应选取所述预定数量的物理端口,并对这些物理端口进行相同的网络层配置;将选取的第一和第二交换机的物理端口一一相连;(3)在路由表中将第一交换机需要转发到第二交换机的网络层报文的出接口按选取的物理端口进行均匀分配;(4)若发现第一交换机选取的物理端口中的某一物理端口无法正常发出报文,则将路由表中的该物理端口修改为选取的物理端口中其他正常工作的物理端口。 The present invention also provides a method for switch between the network layer link aggregation, comprising the steps of: (1) selecting a predetermined number of physical ports on a first switch, and the phase with the purpose of the physical network ports layer is disposed; (2) a second switch on the appropriate selection of the predetermined number of physical ports, and the same physical network layer configure these ports; physical ports to the selected first and second switches connected one by one; ( 3) a first switch in the route table needs to be forwarded to the network layer packet a second switch according to a uniform distribution interface selected physical port; (4) a first switch selecting a physical port in the physical if found port can not send normal messages, the physical port routing table will be modified to select the physical ports in the physical port other normal work.

优选地,第一交换机或第二交换机从选取的物理端口接收的报文中学习MAC地址,并将其出接口均匀分配在选取的物理端口上。 Preferably, the first switch or the second switch learns the MAC address from the packet received by the physical port selected in the interface and the uniform distribution on the selected physical port.

优选地,在步骤(4)中,第一交换机还将出接口为该物理端口的MAC地址删除优选地,在步骤(3)中,第一和第二交换机从选取的物理端口接收的ARP报文中学习直接路由,并将其出接口在选取的物理端口上均匀分配;第一或第二交换机进行报文转发之前先判断报文是否具有网络层报头,如果是,进行网络层转发;如果否,进行链路层转发。 Preferably, in step (4), the first switch will preferably remove the interface for a physical port MAC address, in step (3), the first and second switch receives the selected physical port ARP packets routing packets directly to learn, and the uniform distribution of the interface on the selected physical port; a first switch is determined before the first or the second packet forwarding packets having a network layer header whether, and if so, forwarding the network layer; if No, a link-layer forwarding.

本发明通过将组成链路聚合端口组的物理端口设置为等价路由的出接口,以软件实现了网络层的链路聚合;通过在组成链路聚合端口组的物理端口上动态分配以链路聚合端口组为出口的MAC地址,以软件实现了链路层的链路聚合;降低了链路聚合的实现成本,同时使得用户可以根据组网需求随意设置链路聚合。 The composition of the present invention, by a physical link aggregation port to the port group interfaces equivalent routes to the link aggregation network software layer; dynamically allocated by polymerizing the physical ports in the port group consisting link to link aggregation port group MAC address port, implemented in software link layer link aggregation; reduces the implementation costs for the link aggregation, and so that the user can freely set the link aggregation according to the network requirements.

附图说明 BRIEF DESCRIPTION

图1所示为链路聚合的示意图;图2所示为本发明所述网络层链路聚合实现方法的流程图;图3所示为本发明所述链路层链路聚合实现方法的流程图;图4所示为应用本发明网络层和链路层链路聚合方法的网络设备的报文处理流程图。 FIG 1 is a schematic illustration of link aggregation; FIG. 2 of the present invention, a method flowchart of an implementation of the network layer link aggregation; FIG. 3 of the present invention, the flow link layer link aggregation implementation of the method shown in FIG. FIG; message processing shown in a flowchart of a network layer network device link and a link-layer polymerization process of the present invention is applied FIG.

具体实施方式 detailed description

在802.3ad标准中,链路聚合应具有如下功能:负载分担:通过链路聚合的数据报文应当均匀地分担在组成链路聚合的多条链路上。 In standard 802.3ad link aggregation should have the following functions: load balancing: The link aggregation data packets should be uniformly shared on multiple links form a link aggregation.

冗余备份:当链路聚合中某条链路发生故障后,数据报文能够快速从其他链路转发。 Redundancy: When a link failure occurs in the link aggregation, the data packet can be quickly transferred from the other link.

在一些网络设备上,组成链路聚合端口组的物理端口能够对数据报文进行网络层路由,则其链路聚合可以工作在网络层,也可以工作在网络层和链路层;而在另一些网络设备上组成链路聚合端口组的物理端口可能只具有链路层转发功能,其链路聚合则只能工作在链路层。 In some network equipment, consisting of the physical port link aggregation port group can be performed on the network layer routing data packets, link aggregation it may operate at the network layer, also work in the network layer and link layer; and in another Some components port link aggregation group of physical ports may only have link-layer forwarding device on the network, which link aggregation can only work at the link layer. 在本发明中,提供了以软件实现网络层和链路层的链路聚合的方法,两者可以分别使用,也可以结合使用。 In the present invention, there is provided a method implemented in software link aggregation network layer and link layer, both may be used separately, it may be used in combination.

图2所示为本发明在网络层实现链路聚合的方法的流程图。 Flowchart of a method of network layer link aggregation 2 of the present invention shown in FIG.

在步骤S210,确定链路聚合端口组包括的n个物理端口,n大于等于2,将上述n个物理端口在网络层进行相同配置。 In step S210, n is determined link aggregation group comprising physical ports of ports, n is two or more, and the n physical ports disposed in the same network layer. 组成链路聚合端口组的所有物理端口应当具有路由功能。 Composition all physical link aggregation port group should have a routing function.

从链路聚合端口组外部看来,这n个物理端口组成一个逻辑端口,具有相同的网络层地址和其他网络层配置。 From the external port link aggregation group opinion, the n physical ports a logical port, with the same network layer address and other network-layer.

在步骤S220,将链路聚合端口组的每个物理端口设置为不转发其他物理端口接收的组播和广播报文。 In step S220, the link aggregation group each physical port of the port not to forward the other physical port to receive multicast and broadcast packets. 如果从链路聚合端口组的某个物理端口进入的组播和广播报文从其他物理端口转发出去,则会形成报文的回环,影响网络的性能。 If a physical link from port aggregation port group incoming multicast and broadcast packets are forwarded from another physical port, packets will loop formation and performance of the network.

本步骤可以采用现有的网络设备上提供的设置实现,设置方法与具体采用的硬件和该硬件的设置命令相关,此处不再赘述。 This step can be provided disposed on the existing network device to implement, provided the hardware and methods associated with the hardware specific command set employed, is not repeated here.

在步骤S230,通过动态路由协议学习以链路聚合端口组的网络层地址为出口的间接路由。 In step S230, the dynamic routing protocol to study the network layer address of the link aggregation group to an outlet port of an indirect route. 配置了动态路由协议的网络设备会与其他网络层设备交换路由信息,其中包括以链路聚合端口组的网络层地址为出口的间接路由信息。 Configuring the dynamic routing protocol network devices to exchange routing information with other network-layer device, which comprises a network layer address to link aggregation group to an outlet port of an indirect route information.

在步骤S240,通过链路聚合端口组收到的ARP(Address ResolutionProtocol,地址解析协议)报文,学习直接路由。 In step S240, the link aggregation by ARP (Address ResolutionProtocol, Address Resolution Protocol) packets received by the port group, learning direct route. ARP报文在子网或VLAN(Virtual Local Area Network,虚拟局域网)内传输,用来将网络节点的网络层地址解析为MAC(Media Access Control,媒介接入控制)地址,网络设备可以通过链路聚合端口组收到的ARP报文学习到哪些网络层地址为直接路由的目的地址。 ARP packets transmitted in the subnet or VLAN (Virtual Local Area Network, VLAN), the network layer address for the network node to resolve MAC (Media Access Control, media access control) address, a network device may link aggregation port group received ARP packet learn which network layer address as the destination address of the direct route.

在步骤S250,配置以链路聚合端口组的网络层地址为出口静态路由。 In step S250, the network layer is configured to link aggregation group address port for export static route.

步骤S230至S250是现有技术中获得路由信息的方法,根据实际组网和网络设备的功能可以采用其中一种,也可以同时采用其中几种。 Step S230 to Step S250 is a process for obtaining routing information in the prior art, according to the actual networking function and the network device may employ one, several of which may be simultaneously employed. 在本发明中不需对这些获得路由信息的方法作改动,可以直接应用。 In the present invention, no need to make changes to the routing information obtained by these methods can be applied directly.

在步骤S260,为以链路聚合端口组的网络层地址为出口的路由生成n条等价路由,这n条等价路由的出接口信息分别设置为链路聚合端口组的n个物理端口。 In step S260, the link aggregation is the network layer address to a port group for the export route generating n equal-cost routes that n equal-cost routes outbound interface information are set to n physical port link aggregation port group.

如果网络设备中到某个目的地址的路径有多条,而且这些路径具有相同的优先级,则这些路径形成了等价路由。 If the network device to a destination address in a plurality of paths, and these paths have the same priority, these routing paths formed equivalence.

在网络设备的路由表中,每一条表项都包括出接口信息,即能够转发至目的地址的本设备上的物理端口。 In the routing table of the network device, each entry includes the interface information, i.e., can be a physical port on the forwarding device according to the destination address. 本发明中,将等价路由的出接口分别配置为组成链路聚合端口组的各个物理端口,这样可以利用等价路由具有的负载分担功能,实现链路聚合的负载均衡。 In the present invention, the equivalent routes egress interface configured as an integral link aggregation group each physical interface ports, so that the route can be used having an equivalent load sharing, load balancing link aggregation.

在步骤S270,网络设备将需要网络层转发的报文均匀分配到所述n个等价路由上,相应地以上述n个物理端口为报文出接口。 In step S270, the network device requires network layer packets to be forwarded to the uniform distribution of the n equivalent routes, corresponding to the n physical port on the outbound interface.

等价路由对通过其转发数据报文具有负载分担的作用,可以使数据报文大致均匀地根据等价路由的多条路径进行转发。 Equivalent load balancing has the effect of routing packets through its forwarding data, the data packets can be forwarded substantially uniformly according to the multiple paths of equal cost routes.

常用的网络层网络设备都具有等价路由功能,对其实现不再赘述。 Common network layer network devices having equivalent routing function, to its realization omitted.

在步骤S280,当链路聚合端口组的部分物理端口与对端的连接中断后,将连接中断的物理端口所对应的等价路由的出接口信息修改为链路聚合端口组中正常工作的物理端口。 In step S280, the physical port and the rear portion when the connection to the peer port interrupt link aggregation group, equivalent routes connecting interruption physical port corresponding to the port interface information to modify the physical link aggregation group works correctly .

当物理端口与对端的连接中断时,网络设备能够检测到该物理端口的状态变化。 When a physical port connected to the interrupt terminal, the network device can detect the change in state of the physical port. 因此,可以根据此状态变化信息得知连接中断的物理端口,在路由表中查找到连接中断的物理端口对应的等价路由,将其出接口信息修改为链路聚合端口组中其他正常工作的物理端口。 Thus, the information that can vary the physical port connection is broken, find the corresponding physical port connection interruption equivalent routes in the routing table, which was modified to the interface information in the other link aggregation group to work according to this state physical port.

当有多个物理端口的连接中断时,最好能够将这些物理端口的流量分散在不同的正常工作的端口上。 When there are a plurality of physical ports connected to the interrupt, the flow rate is preferably capable of dispersing the physical ports on a port different from normal operation. 在修改连接中断的物理端口所对应的等价路由的出接口信息时,如果能够满足任意两个正常工作的物理端口所对应的等价路由数目的差值小于等于1,则可以持续负载基本均衡的状态。 When the interface information equivalent routes modified connection interruption physical port corresponding to the equivalent number of routes if normal operation can satisfy any two physical ports corresponding to the difference is less than or equal to 1, it may be substantially continuous load balancing status.

在步骤S290,当链路聚合端口组连接中断的某个物理端口恢复正常工作后,查找所述n条等价路由中具有相同出接口信息的等价路由,将其中一条等价路由的出接口信息修改为所述恢复正常工作的物理端口。 In step S290, when a physical link aggregation group interruption of the connection port after the port to resume normal operation, the look n equal-cost routes that have the same interface to an equivalent routing information, in which the output interface by an equivalent modifying the physical port information for the resumption of normal operation.

同样,当某个物理端口的连接恢复时,可以由网络设备检测到的状态变化信息得知恢复连接的物理端口。 Similarly, when the physical port connected to a recovery can be detected by the network device that the state change information to restore the physical port. 对具有相同出接口信息的等价路由的其中之一进行修改,可以更好地实现负载均衡。 Modify one of which has the same interface information equivalent routes, better load balancing.

可见,本发明把等价路由融合在链路聚合的实现中,使网络层报文通过路由来进行负载分担。 Be seen, the present invention is equivalent to the link aggregation fusion routes in the network layer packets to be routed through load balancing. 一条链路承载一条路由,多条链路承载多条路由,这些路由都指向同一网络层目的地址,且优先级都相同,因而能够达到链路聚合的效果,在网络层实现了负载分担和冗余备份功能。 A link carries a route, a plurality of link carries multiple routes, the routing point to which the destination address in the same network layer, and have the same priority, it is possible to achieve the effect of the link aggregation, the network layer implements load balancing and redundancy More than backup. 同时,本发明还可以防止报文回环。 Meanwhile, the present invention can also prevent loopback packets.

图3所示为本发明在链路层实现链路聚合的方法的流程图。 Figure 3 is a flowchart of the link layer in the invention is a method of link aggregation.

在步骤S310,确定链路聚合端口组包括的物理端口,链路聚合端口组包括的物理端口的数目大于等于2。 In step S310, the determined number of physical ports of the link aggregation group comprising physical ports ports, the link aggregation group comprising of 2 or more.

在步骤S320,将链路聚合端口组的每个物理端口设置为不转发其他物理端口接收的组播和广播报文。 In step S320, the link aggregation group each physical port is a port not to forward the other physical port to receive multicast and broadcast packets.

如前所述,本步骤能够防止形成链路层的广播报文回环,避免影响网络的性能。 As described above, this step prevents the formation of a broadcast link layer loopback packets, to avoid the performance impact of the network. 同样,本步骤也可以采用现有的网络设备上提供的设置实现。 Similarly, the present step may be employed to achieve set available on existing network devices.

在步骤S330,链路聚合端口组从收到的报文中学习以其为出口的媒介接入控制MAC地址。 In step S330, the link aggregation group to learn from the message received with its outlet to a medium access control (MAC) address.

网络设备的端口(包括逻辑端口和物理端口)收到的报文中的源MAC地址,即为与该端口直接连接的网络节点的MAC地址。 MAC address of the port of the network device (including logical ports and physical ports) MAC source address of the packet received, the network node that is directly connected to the port. 从收到的报文中学习MAC地址是几乎所有具有链路层转发功能的网络设备必备的功能,此处不再赘述。 MAC address learning from packets received almost all network devices having the necessary link layer forwarding functions will not be repeated here.

在步骤S340,设定哈希算法,使其哈希值与正常工作的物理端口具有一一对应关系。 In step S340, the hash algorithm is set so as to work with the hash value of physical ports having one relationship.

哈希算法用来将任意长度的输入报文或数据映射为固定长度的输出,此输出称为哈希值。 Hashing algorithm used to map the input message of any length or fixed length data output, this output is called the hash value. 哈希值以数值为表示形式,数据唯一且非常紧凑。 The hash value is a numerical representation of the data is unique and very compact. 即使输入报文或数据的内容只更改一个字母或数字,都将产生不同的哈希值。 Even typing messages or data change only one letter or number, they will produce different hash values.

哈希算法在现有技术中有多种实现形式。 Hashing algorithm implemented in many forms in the prior art. 只要某种哈希算法产生的哈希值范围能够根据设定变化,该算法就可以在本发明中应用。 As long as a range of the hash value generated by hash algorithm can be the algorithm can be applied in accordance with the present invention, the setting change.

在步骤S350,用哈希算法计算学习到的MAC地址的哈希值。 In step S350, the learned MAC address calculated by the hash value hash algorithm.

在步骤S360,将该MAC地址的出接口设置为计算所得的哈希值对应的物理端口。 In step S360, the MAC address of the outgoing interface is provided to the physical port corresponding to the calculated hash value of.

举一个简单的例子,可以采用如下的哈希算法:将MAC地址的6个字节相加,将所得的和数以正常工作的物理端口数量取余,当正常工作的物理端口数量为4时,可能的余数为0、1、2和3,令其分别对应与4个正常工作的物理端口,这样可以根据哈希值对MAC地址的出接口进行设置。 As a simple example, the following hash algorithms may be employed: the six-byte MAC address by adding, to the resultant sum of the number of physical ports in the normal operation modulo, the number of physical ports when the normal operation is 4 possible remainders 0, 1 and 3, so that it corresponds to normal operation of the four physical ports, respectively, so that the interface can be set according to the MAC address hash value.

在本发明中,也可以采用其他方法设置MAC地址的出接口,替代步骤S340至S360,只要能够较为均匀地将学习到的MAC地址分配在正常工作的物理端口上即可,以实现链路聚合的负载分担。 In the present invention, other methods may be used provided that the MAC address of the interface, instead of step S340 to S360, as long as more uniformly learned MAC address allocation on a physical port to normal operation, in order to link aggregation load sharing.

在步骤S370,当链路聚合端口组的部分物理端口与对端的连接中断后,删除以连接中断的物理端口为出接口的MAC地址,转步骤S330。 In step S370, when some of the physical port link aggregation group and the connection to the peer interruption, to delete a physical port connected to an interrupt interface MAC address, proceed to step S330.

当链路聚合端口组中有物理端口的连接中断时,该物理端口上的流量最好分散在其他正常工作的物理端口上。 When the link aggregation group in connection physical port interrupt traffic on the physical port is preferably dispersed on a physical port of the other work. 因此,删除以连接中断的物理端口为出接口的MAC地址,重新启动这些MAC地址的学习和在物理端口上的分配过程,可以较好地实现负载均衡。 Therefore, in order to delete the connection is lost to a physical port MAC address of the interface, restart the learning process and assign these MAC addresses on a physical port, you can achieve better load balancing.

在步骤S380,当链路聚合端口组连接中断的物理端口恢复正常工作后,删除以链路聚合端口组为出口的所有MAC地址,转步骤S330。 In step S380, when the link aggregation group of physical port connection is interrupted to resume normal work, deletes all MAC addresses at the link aggregation group to an outlet, proceed to step S330.

当链路聚合端口组中有中断的物理端口恢复连接时,为了实现负载均衡,只能删除以链路聚合端口组为出口的所有MAC地址,重新启动这些MAC地址的学习和分配过程。 When the link aggregation port group interrupt physical port connection is restored, in order to achieve load balancing, only to delete the link aggregation port group for all the MAC address of the exporter, restart the learning process and assign these MAC addresses.

本发明在链路层也实现了负载分担和冗余备份功能,同时也可以防止报文回环。 The present invention is also achieved at the link layer load balancing and redundancy backup, also can prevent loopback packets.

单独在网络层或链路层实施本发明后,通过链路聚合端口组发送和接收数据包的方法不会有变化。 The method of the present invention embodiments separately after the network layer or data link layer, the link aggregation group by sending and receiving data packets will not change. 在结合使用本发明两种方法时,对经链路聚合转发的TCP/IP(Transfer Control Protocol/Intemet Protocol,传输控制协议/互联网协议)报文推荐采用图4所示的处理流程,此时在网络层实现中必须执行步骤S240,使链路聚合端口组可以通过直接路由转发目的IP地址为位于同一子网或VLAN的节点的TCP/IP报文。 When using two methods in conjunction with the present invention, a link aggregation forwarded via TCP / IP (Transfer Control Protocol / Intemet Protocol, Transmission Control Protocol / Internet Protocol) packets is recommended in the processing flow shown in FIG. 4, in this case network layer must be performed in step S240, the link aggregation group so that the destination IP address may be forwarded to TCP / IP packet is in the same subnet or VLAN node by the direct route.

请参阅图4,在步骤S410,判断该报文是否有网络层报头,如果有,执行步骤S420;如果没有,执行步骤S450。 Please refer to FIG. 4, at step S410, the packet is determined whether a network layer header, and if so, executing step S420; if not, to step S450.

在步骤S420,判断在路由表中有该报文的目的IP地址,如果有,执行步骤S440,根据等价路由对该报文进行网络层转发,报文处理结束;如果没有,执行步骤S430,根据缺省路由对该报文进行网络层转发,报文处理结束。 In step S420, the routing table is determined with a destination IP address of the packet, if yes, perform step S440, the network layer packet forwarding according to the equal-cost routes the packet process ends; if not, performing step S430, the the default route to the network layer packet forwarding, packet processing ends.

在步骤S450,判断该报文的目的MAC地址是否是组播或广播地址,如果是,执行步骤S480;如果否,执行步骤S460。 In step S450, the packet is determined whether the destination MAC address is a multicast or broadcast address, if yes, performs step S480; if NO, to step S460.

在步骤S460,判断该报文的目的MAC地址是否在MAC地址表中,如果是,执行步骤S470,根据MAC地址表对该报文进行转发,报文处理结束;如果否,执行步骤S480。 In step S460, the object is determined whether the MAC address of the packet in the MAC address table, if yes, step S470, forwards the message according to a MAC address table, the packet processing ends; if not, to step S480.

在步骤S480,任选一个链路汇聚组的物理端口对该报文进行转发,报文处理结束。 In step S480, optionally, a physical port link aggregation group to forward the message, the message processing is terminated.

请参见图1,假设网络设备100和网络设备200都是具有网络层转发功能的交换机,在网络层交换机100和200之间,可以通过下述方法实现链路聚合:在网络层交换机100上选取预定数量的物理端口,并对这些物理端口进行相同的网络层配置;在网络层交换机200上相应选取相同数量的物理端口,并对这些物理端口进行相同的网络层配置;将网络层交换机100和网络层交换机200上选取的物理端口一一相连;在网络层交换机100上,将路由表中将下一跳为网络层交换机200的路由的出接口在选取的物理端口上进行均匀分配;同样,在网络层交换机200上,将路由表中将下一跳为网络层交换机100的路由的出接口在选取的物理端口上也进行均匀分配;如果网络层交换机100发现选取的物理端口中的某一个无法发送和/或接收报文,则将其路由表中的该物理端口修改为选取的物理 Referring to Figure 1, assume that the network device 100 and network device 200 is a network layer having a forwarding function switch, the switch 100 and between the network layer 200, link aggregation can be realized by the following method: selecting switch 100 in the network layer a predetermined number of physical ports, the physical ports and the same network-layer; at the network level selection switch 200 corresponding to the same physical port number, and the same physical network layer configure these ports; the switch 100 and the network layer network layer 200 on the selected switch port physically connected to eleven; switch 100 at the network layer, the hop routing table in the network layer interface routes the switch 200 is uniformly distributed over selected physical port; Similarly, at the network level switch 200, the routing table in the routing of the interface to the next hop for the network layer 100 of switch also uniformly distributed over selected physical port; a network layer if a switch 100 to select a physical port found in sending and / or receiving packets, then the physical port routing table which is selected to modify the physical 口中其他正常工作的物理端口;同时,网络层交换机200也会发现与该物理端口连接的本机上的物理端口对网络层报文转发出现故障,也将以同样的方式修改其路由表;当网络层交换机100发现出现故障的物理端口恢复正常工作时,修改其路由表中下一跳为网络层交换机200的路由中出接口的值重复的路由,将其出接口修改为恢复工作的物理端口;同时,网络层交换机200也会发现与该物理端口连接的物理端口对网络层报文的转发恢复正常,则其也以同样方式修改其路由表。 Other physical port mouth normal operation; at the same time, the network switch 200 will find the physical layer ports on the machine is connected to the physical port of the network layer packet forwarding fails, the same will also modify its routing table; when the physical network layer 100 found faulty switch port resume normal operation, to modify its routing table the next-hop interface value is repeated routing network layer routing switch 200, which is an interface to modify the physical port recovery ; Meanwhile, the network layer is also found that the physical port switch 200 is connected to the physical port of the network layer packet forwarding back to normal, it is also in the same manner as modify its routing table.

在实现了网络层链路聚合后,还可以同时实现链路层的链路聚合:网络层交换机100和网络交换机200分别从其选取的物理端口上接收的报文中学习MAC地址,并将学习的MAC地址的出接口均匀分配到其选取的物理端口上;如果网络层交换机100或网络层交换机200发现选取的物理端口中的某一个无法发送和/或接收报文,则删除MAC地址转发表中包括该物理端口的表项。 After the polymerization to achieve a network layer link, link aggregation can simultaneously link layer: The network layer packet and the network switch 100 receives the switch 200 to select a physical port from which the MAC address learning, learning and an interface to uniformly MAC address assigned to its chosen physical port; if the network layer switch 100 or the network layer switch 200 found in the selected physical port in either one can not send and / or receive messages, delete the MAC address forwarding table entry includes the physical port.

如果在网络层交换机100和网络层交换机200上应用图4所示的流程,则应当令网络层交换机100和网络层交换机200从选取的物理端口接收的ARP报文中学习直接路由,并将其出接口在选取的物理端口上均匀分配。 If the network layer and the flow shown in the switch 100 switches the network layer 200 of FIG. 4 the application, should the switch 100 so that the network layer and the network layer switch 200 learning from the ARP packet received by the physical port selected in the direct route, and a uniform distribution of the interface on the selected physical port. 这样,网络层交换机100和网络层交换机200可以在进行报文转发前先判断报文是否具有网络层报头,如果是,进行网络层转发;如果否,进行链路层转发。 Thus, the network layer and the network layer 100 switches switch 200 may be performed before forwarding the packet to judge whether the packet has a network layer header, and if so, forwarding the network layer; if not, the link-layer forwarding.

本发明非常适用于以下情形:在树状网络中处于汇聚层或者核心层的网络设备,与下级网络设备进行网络层和/或链路层互联;在环状网络中处于汇聚层或者核心层的网络设备,和同级别的网络设备进行网络层和/或链路层互联。 The present invention is applicable to the following scenario: the convergence layer device in the network or the core layer in a tree network, the network layer and / or link layer interconnection and the subordinate network device; convergence layer or the core layer is in the ring network network equipment, network equipment and the same level of the network layer and / or link layer interconnect.

以上所述本发明的实施方式,并不构成对本发明保护范围的限定。 The above described embodiment of the present invention, not to limit the scope of the present invention. 任何在本发明的精神和原则之内所作的任何修改、等同替换和改进等,均应包含在本发明的权利要求保护范围之内。 Any Any modifications within the spirit and principle of the present invention, equivalent substitutions and improvements should be included within the protection scope claimed in the claims of the present invention.

Claims (14)

  1. 1.一种网络设备在网络层实现链路聚合的方法,其特征在于,包括以下步骤:a)确定链路聚合端口组包括的n个物理端口,n大于等于2,将上述n个物理端口在网络层进行相同配置;b)为以所述链路聚合端口组为出口的某个路由生成n条等价路由,所述n条等价路由的出接口信息分别设置为所述链路聚合端口组的n个物理端口;c)网络设备将需要网络层转发的报文均匀分配到所述n个等价路由上,相应地以上述n个物理端口为报文出接口;d)当链路聚合端口组的部分物理端口与对端的连接中断后,将连接中断的物理端口所对应的等价路由的出接口信息修改为链路聚合端口组中正常工作的物理端口。 1. A network link aggregation device to implement the method at the network layer, characterized by comprising the steps of: a) determining the n-th link aggregation group comprising physical ports of ports, n is two or more, the physical ports of the n disposed in the same network layer; b) the polymerization is a link of the link aggregation group to generate n pieces of routes as a route outlet, the n equivalent routes the outbound interface information are provided n physical ports port group; c) a network layer network device needs to forward packets to the uniform distribution of the n equivalent routes, corresponding to the n ports a physical interface packets; D) when the chain after passage portion aggregation port group connected to the remote physical port interrupt, routes the connection interrupted equivalent physical port interface information corresponding to the link aggregation modify physical ports in the port group working.
  2. 2.按照权利要求1所述在网络层实现链路聚合的方法,其特征在于,所述方法还包括:将链路聚合端口组的每个物理端口设置为不转发其他物理端口接收的组播和广播报文。 2. The method implemented in the network layer link aggregation according to claim 1, characterized in that, said method further comprising: a link aggregation group each physical port is not the port forwards the received multicast physical ports and broadcast packets.
  3. 3.按照权利要求2所述在网络层实现链路聚合的方法,其特征在于,所述步骤d)之后还包括:e)当链路聚合端口组连接中断的某个物理端口恢复正常工作后,查找所述n条等价路由中具有相同出接口信息的等价路由,将其中一条等价路由的出接口信息修改为所述恢复正常工作的物理端口。 3. The method implemented in the network layer link aggregation claim 2, wherein, after said step d) further comprising: e) When a physical link aggregation port group connected to an interrupt port to resume normal work after searching the n equal-cost routes with the same equivalent route outbound interface information, in which an equivalent route outbound interface information is modified to the physical port resumes normal operation.
  4. 4.按照权利要求3所述在网络层实现链路聚合的方法,其特征在于,所述步骤d)中对连接中断的物理端口所对应的等价路由的出接口信息的修改满足以下条件:任意两个正常工作的物理端口所对应的等价路由数目的差值小于等于1。 4. The method at the network layer link aggregation of claim 3, wherein said step d) to modify the interface information for routing connections interrupted equivalent physical port corresponding to satisfy the following condition: the equivalent number of routes difference between any two normal working physical port corresponding to 1 or less.
  5. 5.按照权利要求4所述在网络层实现链路聚合的方法,其特征在于,所述步骤a)与步骤b)之间包括下述至少一项:通过动态路由协议学习以所述网络层地址为出口的间接路由;配置以所述网络层地址为出口静态路由;通过链路聚合端口组收到的地址解析协议ARP报文,学习直接路由。 5. The method at the network layer link aggregation to claim 4, characterized by comprising between said step a) and step b) at least one of the following: to study the dynamic routing protocol network layer address is an indirect route outlet; configured to the network layer address is a static route outlet; ARP through link aggregation port group received ARP packet, learning direct route.
  6. 6.按照权利要求1至5任意一项所述在网络层实现链路聚合的方法,其特征在于,所述步骤a)之后还包括:a1)链路聚合端口组从收到的报文中学习以其为出口的媒介接入控制MAC地址;a2)将学习到的MAC地址的出接口信息设置为链路聚合端口组正常工作的物理端口;a3)当链路聚合端口组的部分物理端口与对端的连接中断后,删除以连接中断的物理端口为出接口的MAC地址,转步骤a1);a4)当链路聚合端口组连接中断的物理端口恢复正常工作后,删除以链路聚合端口组为出口的所有MAC地址,转步骤a1)。 1-5 6. The method of any one of the link aggregation implemented in a network layer, wherein said step a) further including the claims: a1) a link aggregation group from the received packet learning its media access control (MAC) address of the outlet; A2) to the set interface information learned MAC addresses of physical ports as a link aggregation port group of normal operation; A3) as part of the physical link aggregation group port port a connection with the peer interruption, deletion interruption of the connection to the physical port interface MAC address, proceed to step a1); a4) when the link aggregation group the physical port connection is broken after the resumption of normal operation, in order to remove the link aggregation port group export all MAC addresses, proceed to step a1).
  7. 7.按照权利要求6所述在网络层实现链路聚合的方法,其特征在于,所述方法还包括:设置链路聚合端口组的每个物理端口不转发其他物理端口接收的组播和广播报文。 7. The method at the network layer link aggregation of claim 6, wherein said method further comprises: setting each physical port link aggregation port group does not forward the other physical port to receive broadcast and multicast message.
  8. 8.按照权利要求7所述在网络层实现链路聚合的方法,其特征在于,所述步骤a2)具体为:设定哈希算法,使其哈希值与正常工作的物理端口具有一一对应关系;用所述哈希算法计算所述MAC地址的哈希值;将所述MAC地址的出接口设置为计算所得的哈希值对应的物理端口。 Setting port has eleven hash algorithm, and hash value so that physical work is: 8. The method of network layer link aggregation according to the claim 7, wherein said step a2) is particularly corresponding relation; calculating a hash value of the MAC address with the hash algorithm; physical port that the interface is provided for the MAC address hash value corresponding to the calculated.
  9. 9.一种网络设备在链路层实现链路聚合的方法,其特征在于,包括以下步骤:A.确定链路聚合端口组包括的物理端口;B.将链路聚合端口组的每个物理端口设置为不转发其他物理端口接收的组播和广播报文;C.链路聚合端口组从收到的报文中学习以其为出口的媒介接入控制MAC地址;D.将学习到的MAC地址的出接口信息设置为链路聚合端口组正常工作的物理端口;E.当链路聚合端口组的部分物理端口与对端的连接中断后,删除以连接中断的物理端口为出接口的MAC地址,转步骤C。 A network device link aggregation method implemented in the link layer, characterized by comprising the steps of:. A link aggregation determining physical port comprises a port group; B each physical port link aggregation group. port not to forward the other physical port to receive multicast and broadcast packets;. C link aggregation group learning to its exit from the media access control (MAC) address of the received packet;. D will be learned the interface information provided to the MAC address of the physical port link aggregation port group of normal operation;. E port when the portion of the physical link aggregation port group connected to the end of interrupt, the interrupt connection to delete a physical port of interface MAC address, proceed to step C.
  10. 10.按照权利要求9所述在网络层实现链路聚合的方法,其特征在于,所述步骤E之后还包括:当链路聚合端口组连接中断的物理端口恢复正常工作后,删除以链路聚合端口组为出口的所有MAC地址,转步骤C;所述步骤D具体为:设定哈希算法,使其哈希值与正常工作的物理端口具有一一对应关系;用所述哈希算法计算所述MAC地址的哈希值;将所述MAC地址的出接口设置为计算所得的哈希值对应的物理端口。 10. The method in the network layer of the link aggregation according to claim 9, wherein, after said step E further comprises: When the link aggregation group of physical port connection is interrupted to resume normal operation, to remove the link aggregation port group for all the MAC address port, proceed to step C; D step specifically comprises: hashing algorithm is set so as to work with the hash value of physical ports having one relationship; hashing algorithm with the calculates a hash value of the MAC address; an interface to a physical port of the set MAC address hash value corresponding to the calculated.
  11. 11.一种在网络层交换机之间实现链路聚合的方法,其特征在于,包括以下步骤:(1)在第一交换机上选取预定数量的物理端口,并对这些物理端口进行相同的网络层配置;(2)在第二交换机上相应选取所述预定数量的物理端口,并对这些物理端口进行相同的网络层配置;将选取的第一和第二交换机的物理端口一一相连;(3)在路由表中将第一交换机需要转发到第二交换机的网络层报文的出接口按选取的物理端口进行均匀分配;(4)若发现第一交换机选取的物理端口中的某一物理端口无法正常发出报文,则将路由表中的该物理端口修改为选取的物理端口中其他正常工作的物理端口。 A switch implemented between the network layer link aggregation method, characterized by comprising the steps of: (1) selecting a predetermined number of physical ports on a first switch, and the network layer of the same physical port configuration; (2) on a respective second switch selecting said predetermined number of physical ports, and the same physical network layer configure these ports; physical ports to the selected first and second switches connected one by one; (3 ) a first switch in the route table needs to be forwarded to the network layer packet a second switch according to a uniform distribution interface selected physical port; (4) a certain physical port of the first switch of the selected physical port if found can not send normal messages, the physical port routing table will be modified to select the physical ports in the physical port other normal work.
  12. 12.按照权利要求11所述在网络层交换机之间实现链路聚合的方法,其特征在于,第一交换机或第二交换机从选取的物理端口接收的报文中学习MAC地址,并将其出接口均匀分配在选取的物理端口上。 12. The method switches between the network layer of the link aggregation according to claim 11, wherein the first switch or the second switch learns the MAC address from the physical port selected packets received, and its Interface uniformly distributed over selected physical port.
  13. 13.按照权利要求12所述在网络层交换机之间实现链路聚合的方法,其特征在于,在步骤(4)中,第一交换机还将出接口为该物理端口的MAC地址删除。 13. The method switches between a network layer link aggregation according to claim 12, wherein, wherein, in step (4), the first switch will be a physical port interface MAC address for the deletion.
  14. 14.按照权利要求11所述在网络层交换机之间实现链路聚合的方法,其特征在于,在步骤(3)中,第一和第二交换机从选取的物理端口接收的ARP报文中学习直接路由,并将其出接口在选取的物理端口上均匀分配;第一或第二交换机进行报文转发之前先判断报文是否具有网络层报头,如果是,进行网络层转发;如果否,进行链路层转发。 14. The method switches between the network layer of the link aggregation according to claim 11, wherein, in step (3), the first and second switches learning from the ARP packet received by the physical port selected in direct route, and a uniform distribution on the interface selected physical port; a first or a second packet to the switch determines whether the packet having a network layer header, if, before forwarding the network layer forwarding; if not, be link layer forwarding.
CN 200510076709 2005-06-10 2005-06-10 Method of implementing link aggregation of network equipment CN100479465C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200510076709 CN100479465C (en) 2005-06-10 2005-06-10 Method of implementing link aggregation of network equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200510076709 CN100479465C (en) 2005-06-10 2005-06-10 Method of implementing link aggregation of network equipment

Publications (2)

Publication Number Publication Date
CN1725761A true true CN1725761A (en) 2006-01-25
CN100479465C CN100479465C (en) 2009-04-15

Family

ID=35925000

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200510076709 CN100479465C (en) 2005-06-10 2005-06-10 Method of implementing link aggregation of network equipment

Country Status (1)

Country Link
CN (1) CN100479465C (en)

Cited By (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008009160A1 (en) * 2006-06-30 2008-01-24 Zte Corporation An access configuration method for the link aggregation and a link aggregation system
WO2008077313A1 (en) * 2006-12-22 2008-07-03 Huawei Technologies Co., Ltd. A method and a device for aggregating ports
WO2008089667A1 (en) * 2007-01-22 2008-07-31 Huawei Technologies Co., Ltd. Method and apparatus for reducing service loss in a link aggregation group
CN101252521A (en) * 2008-03-19 2008-08-27 中兴通讯股份有限公司 Method for forwarding message and distributed three layer network eqiupment
CN100492970C (en) 2006-02-17 2009-05-27 阿拉克斯拉网络株式会社 Network relay apparatus and method for transmitting data packet
CN100589408C (en) 2007-06-05 2010-02-10 中兴通讯股份有限公司 A communication network connection method and its device
CN101827038A (en) * 2010-05-24 2010-09-08 杭州华三通信技术有限公司 Distributed equipment and method for forwarding message in distributed equipment
CN101877677A (en) * 2010-06-25 2010-11-03 中兴通讯股份有限公司 Tunnel switching method and system for multi-protocol label switching services
CN101286899B (en) 2007-04-12 2011-06-08 株式会社日立制作所 Network system and node apparatus
CN102158398A (en) * 2011-02-25 2011-08-17 杭州华三通信技术有限公司 Method and device for forwarding messages
CN101136899B (en) 2006-08-31 2011-09-07 杭州华三通信技术有限公司 Ethernet physical layer transmission protecting method, and device and equipment for implementing the same
CN101170723B (en) 2006-10-26 2012-01-04 三星电子株式会社 Method for managing mobile information of mobile phone between communication systems
WO2012051888A1 (en) * 2010-10-20 2012-04-26 中兴通讯股份有限公司 Method and system for realizing cross-node aggregated links in a packet switching network
CN102457540A (en) * 2010-10-25 2012-05-16 成都谛听科技有限公司 Low-cost network-attached storage (NAS) integration equipment bandwidth multiplication method
CN101510837B (en) 2009-04-01 2012-10-10 福建星网锐捷网络有限公司 Ethernet bridge equipment, method and equipment for migrating polymerization mouth address
CN101626342B (en) 2008-07-07 2013-06-26 华为技术有限公司 Method, device and system for realizing link aggregation
CN103236975A (en) * 2013-05-09 2013-08-07 杭州华三通信技术有限公司 Message forwarding method and message forwarding device
CN103370910A (en) * 2010-11-19 2013-10-23 极进网络公司 Methods, systems, and computer readable media for next hop scaling with link aggregation
CN103401707A (en) * 2013-07-26 2013-11-20 杭州华三通信技术有限公司 Link aggregation method and access equipment
CN103501250A (en) * 2013-09-30 2014-01-08 杭州华三通信技术有限公司 Data flow processing method and device in distribution type link aggregation network
WO2014059873A1 (en) * 2012-10-19 2014-04-24 中兴通讯股份有限公司 Method and device for flow path negotiation in link aggregation group
CN103986658A (en) * 2014-05-14 2014-08-13 北京锐安科技有限公司 Message output method and device
CN104144129A (en) * 2014-06-27 2014-11-12 杭州华三通信技术有限公司 Method and device for transmitting Fast-Hello messages
CN104486124A (en) * 2014-12-19 2015-04-01 盛科网络(苏州)有限公司 Device and method for realizing MLAG (multi-system link aggregation) by logical ports
WO2016033785A1 (en) * 2014-09-04 2016-03-10 华为技术有限公司 Method for synchronizing forwarding tables, network device, and system
CN105515998A (en) * 2015-11-26 2016-04-20 北京邮电大学 Method for communicating third-level domain and second-level domain in software-defined packet transport network (SPTN) domain and system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1172488C (en) 2002-04-01 2004-10-20 港湾网络有限公司 Dividing method for bond ports of switch and switch chip
CN1571354A (en) 2003-07-12 2005-01-26 华为技术有限公司 A method for implementing link aggregation

Cited By (39)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8081637B2 (en) 2006-02-17 2011-12-20 Alaxala Networks Corporation Network apparatus and method for forwarding packet
CN100492970C (en) 2006-02-17 2009-05-27 阿拉克斯拉网络株式会社 Network relay apparatus and method for transmitting data packet
US7715407B2 (en) 2006-02-17 2010-05-11 Alaxala Networks Corporation Network apparatus and method for forwarding packet
WO2008009160A1 (en) * 2006-06-30 2008-01-24 Zte Corporation An access configuration method for the link aggregation and a link aggregation system
CN101136899B (en) 2006-08-31 2011-09-07 杭州华三通信技术有限公司 Ethernet physical layer transmission protecting method, and device and equipment for implementing the same
CN101170723B (en) 2006-10-26 2012-01-04 三星电子株式会社 Method for managing mobile information of mobile phone between communication systems
WO2008077313A1 (en) * 2006-12-22 2008-07-03 Huawei Technologies Co., Ltd. A method and a device for aggregating ports
WO2008089667A1 (en) * 2007-01-22 2008-07-31 Huawei Technologies Co., Ltd. Method and apparatus for reducing service loss in a link aggregation group
US8107506B2 (en) 2007-01-22 2012-01-31 Huawei Technologies Co., Ltd. Method and apparatus for reducing service impairment in link aggregation group
CN101286899B (en) 2007-04-12 2011-06-08 株式会社日立制作所 Network system and node apparatus
CN100589408C (en) 2007-06-05 2010-02-10 中兴通讯股份有限公司 A communication network connection method and its device
CN101252521A (en) * 2008-03-19 2008-08-27 中兴通讯股份有限公司 Method for forwarding message and distributed three layer network eqiupment
CN101626342B (en) 2008-07-07 2013-06-26 华为技术有限公司 Method, device and system for realizing link aggregation
CN101510837B (en) 2009-04-01 2012-10-10 福建星网锐捷网络有限公司 Ethernet bridge equipment, method and equipment for migrating polymerization mouth address
CN101827038A (en) * 2010-05-24 2010-09-08 杭州华三通信技术有限公司 Distributed equipment and method for forwarding message in distributed equipment
RU2528149C1 (en) * 2010-06-25 2014-09-10 ЗетТиИ Корпорейшн Tunnel switching method and system for multi-protocol label switching services
CN101877677A (en) * 2010-06-25 2010-11-03 中兴通讯股份有限公司 Tunnel switching method and system for multi-protocol label switching services
CN101877677B (en) 2010-06-25 2014-08-13 中兴通讯股份有限公司 Tunnel switching method and system for multi-protocol label switching services
WO2011160517A1 (en) * 2010-06-25 2011-12-29 中兴通讯股份有限公司 Tunnel switching method and system for multi-protocol label switching services
WO2012051888A1 (en) * 2010-10-20 2012-04-26 中兴通讯股份有限公司 Method and system for realizing cross-node aggregated links in a packet switching network
CN102457540A (en) * 2010-10-25 2012-05-16 成都谛听科技有限公司 Low-cost network-attached storage (NAS) integration equipment bandwidth multiplication method
CN103370910A (en) * 2010-11-19 2013-10-23 极进网络公司 Methods, systems, and computer readable media for next hop scaling with link aggregation
CN103370910B (en) * 2010-11-19 2016-05-18 极进网络公司 Methods for the next hop link aggregation scaling, systems, and computer-readable medium
CN102158398A (en) * 2011-02-25 2011-08-17 杭州华三通信技术有限公司 Method and device for forwarding messages
US9736060B2 (en) 2012-10-19 2017-08-15 Zte Corporation Method and device for negotiating traffic path in link aggregation group
WO2014059873A1 (en) * 2012-10-19 2014-04-24 中兴通讯股份有限公司 Method and device for flow path negotiation in link aggregation group
CN103236975A (en) * 2013-05-09 2013-08-07 杭州华三通信技术有限公司 Message forwarding method and message forwarding device
CN103236975B (en) * 2013-05-09 2017-02-08 杭州华三通信技术有限公司 Message forwarding method and apparatus
CN103401707A (en) * 2013-07-26 2013-11-20 杭州华三通信技术有限公司 Link aggregation method and access equipment
CN103401707B (en) * 2013-07-26 2016-09-28 杭州华三通信技术有限公司 Link aggregation method and an access device
CN103501250A (en) * 2013-09-30 2014-01-08 杭州华三通信技术有限公司 Data flow processing method and device in distribution type link aggregation network
CN103986658B (en) * 2014-05-14 2017-06-06 北京锐安科技有限公司 OSPF packets output method and apparatus
CN103986658A (en) * 2014-05-14 2014-08-13 北京锐安科技有限公司 Message output method and device
CN104144129B (en) * 2014-06-27 2017-11-28 新华三技术有限公司 One kind of Fast-Hello packet transmission method and apparatus
CN104144129A (en) * 2014-06-27 2014-11-12 杭州华三通信技术有限公司 Method and device for transmitting Fast-Hello messages
US10079781B2 (en) 2014-09-04 2018-09-18 Huawei Technologies Co., Ltd. Forwarding table synchronization method, network device, and system
WO2016033785A1 (en) * 2014-09-04 2016-03-10 华为技术有限公司 Method for synchronizing forwarding tables, network device, and system
CN104486124A (en) * 2014-12-19 2015-04-01 盛科网络(苏州)有限公司 Device and method for realizing MLAG (multi-system link aggregation) by logical ports
CN105515998A (en) * 2015-11-26 2016-04-20 北京邮电大学 Method for communicating third-level domain and second-level domain in software-defined packet transport network (SPTN) domain and system

Also Published As

Publication number Publication date Type
CN100479465C (en) 2009-04-15 grant

Similar Documents

Publication Publication Date Title
Yu et al. Scalable flow-based networking with DIFANE
US8369335B2 (en) Method and system for extending routing domain to non-routing end stations
US6208649B1 (en) Derived VLAN mapping technique
US6014380A (en) Mechanism for packet field replacement in a multi-layer distributed network element
US5920566A (en) Routing in a multi-layer distributed network element
US8953590B1 (en) Layer two virtual private network having control plane address learning supporting multi-homed customer networks
US7269135B2 (en) Methods and systems for providing redundant connectivity across a network using a tunneling protocol
US6807176B1 (en) Arrangement for switching data packets in a network switch based on subnet identifier
US20080112323A1 (en) Hash-based multi-homing
US20060039385A1 (en) Method and system for router aggregation
US7428219B2 (en) System and method to exchange information between a control element and forwarding elements in a network element architecture
US6952397B2 (en) Communication in a bidirectional ring network with single-direction receiving
US20120033665A1 (en) System and method for multi-chassis link aggregation
US20120033668A1 (en) IP Multicast Snooping and Routing with Multi-Chassis Link Aggregation
US8750820B2 (en) Method and apparatus for selecting between multiple equal cost paths
US20120163164A1 (en) Method and system for remote load balancing in high-availability networks
US20110228767A1 (en) Virtual link aggregation of network traffic in an aggregation switch
US9350680B2 (en) Protection switching over a virtual link aggregation
US20140071987A1 (en) Systems and methods providing reverse path forwarding compliance for a multihoming virtual routing bridge
US6910149B2 (en) Multi-device link aggregation
US20120110393A1 (en) Method and apparatus providing failover for a point to point tunnel for wireless local area network split-plane environments
US9019814B1 (en) Fast failover in multi-homed ethernet virtual private networks
US20080186968A1 (en) Triple-tier anycast addressing
US20130094357A1 (en) Fhrp optimizations for n-way gateway load balancing in fabric path switching networks
US20100329147A1 (en) Inter-node link aggregation system and method

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model
CP03