CN100407671C - Network communication method for carrying out load division - Google Patents
Network communication method for carrying out load division Download PDFInfo
- Publication number
- CN100407671C CN100407671C CN031197582A CN03119758A CN100407671C CN 100407671 C CN100407671 C CN 100407671C CN 031197582 A CN031197582 A CN 031197582A CN 03119758 A CN03119758 A CN 03119758A CN 100407671 C CN100407671 C CN 100407671C
- Authority
- CN
- China
- Prior art keywords
- gateway
- area network
- main
- main frame
- local area
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Lifetime
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
Abstract
The present invention provides a network communication method for realizing a network load sharing function and a redundant backup function in a communication network. The communication network comprises host computers of a wide area network and a local area network, and a plurality of gateways for connecting the host computers of the wide area network and the local area network. The gateways comprise main gateways and standby gateways, and gateway addresses configured by the host computers in the local area network are virtual IP addresses of the gateways. The method comprises the following steps that when messages are transmitted between the host computer and the wide area network, the host computer can transmit an ARP request message according to the configured gateway addresses to obtain MAC addresses of the IP addresses of the gateways, the main gateways are determined into the gateways for supplying serve for message transmission according to the states of the gateways, the virtual MAC address of the appointed gateway is used for responding the ARP request, and then, the appointed gateway can supply the message transmitting service for the host computer. The host computers in the local area network have no need to be especially configured by the present invention, and can sufficiently enjoy benefit brought by load share and redundant backup.
Description
Invention field
The present invention relates generally to the network communications technology, be particularly related to a kind of network communicating system and method thereof that offered load is shared function that have, can transmit by the data between a plurality of gateways and the wide area network for the main frame of local area network (LAN) load balancing and redundancy backup function are provided.
Background technology
At present, along with the development based on the network communications technology of TCP/IP (transmission control protocol and Internet protocol), people also improve day by day to the requirement of network communicating system.Fig. 1 is the composition frame chart of a typical network communicating system.As shown in Figure 1, this system is by the local area network (LAN) that comprises a plurality of main frames, wide area network and the gateway that is used to connect local area network (LAN) and wide area network.Main frame in the local area network (LAN) can be by a plurality of gateway access wide area networks.But, when the employed gateway of the main frame in the local area network (LAN) breaks down, just need and can realize and the communicating by letter of wide area network by other gateway, thus the load balancing of gateway and the requirement of backup functionality have been proposed communication system.
In the prior art, adopt VRRP (Virtual Router Redundancy Protocol) technology to provide backup functionality usually for gateway.This technology is by RFC (Request for Comment) 2338 definition of IETF (internet engineering task group), and it has proposed the redundancy backup scheme at the local area network (LAN) main frame by multiple gateway visit wide area network.This scheme is to the effect that: for main frame is provided with the virtual ip address that gateway address is a gateway, these outlet gateways can be divided into main with gateway and standby gateway in local area network (LAN), mainly are responsible for gateway that main frame provides message to transmit service in the local area network (LAN).Main frame is in by the main process that sends datagram with gateway, if main with the gateway inefficacy, main frame can send data by standby gateway.The master has nothing to do with the switching of gateway and standby gateway and the operation of main frame, and main frame can not know to have a plurality of gateways to exist, needn't know whether the switching that gateway has taken place yet.For main frame, as long as there is gateway to provide correct service for it.
Fig. 2 utilizes the VRRP technology to carry out the principle schematic of transfer of data and gateway switching in the prior art.As shown in Figure 2, the gateway address of each main frame all points to the virtual ip address of VRRP, and main frame carries out address resolution by ARP (address resolution protocol), promptly obtains corresponding between IP address and MAC (media interviews control) address.When main frame when wide area network sends data, at first need to resolve the MAC Address of gateway IP, this gateway IP is exactly the virtual IP address of VRRP backup group, so when main frame sends the ARP request, mainly will reply with a virtual mac address with gateway.Afterwards, main frame just sends data to main with gateway by this virtual mac address.Main is the data and the forwarding of its virtual mac address with gateway reception purpose MAC.When gateway when main frame sends data, if the MAC Address of this main frame not also then mainly can send the ARP request to this main frame with gateway, obtaining the MAC Address of this main frame, and then data are sent to this main frame.
In addition, the main gateway of using also regularly sends the VRRP protocol massages to standby gateway in operate as normal, so that the main state with gateway of standby gateway monitoring.When standby gateway lost continuously specify number behind the main VRRP protocol massages that sends with gateway, standby gateway judges and mainly breaks down with gateway, and takes over automatically and mainly provide service with gateway, so just finish mainly use, the switching between the standby gateway.
In such scheme shown in Figure 2, exist a problem, promptly, because the master is being connected the wide area network link with gateway with standby gateway, but the gateway MAC Address of each main frame all is to point to the MAC Address of same gateway (the main gateway of using) in the local area network (LAN), only utilized at ordinary times to connect the main gateway wide area network link transmission data of using, the link of other connection standby gateway then is idle.Only when the master was lost efficacy with gateway, just transmit message data by standby gateway.Therefore this technology does not have the load balancing function, and the unbalanced situation that causes load too much to concentrate on certain some link easily takes place.In order to address this problem, need to adopt the load balancing technology.The way of VRRP provides a plurality of VRRP backup groups.The main gateway difference of using of each VRRP backup group, as shown in Figure 3, be equipped with two backup groups in the network communicating system, be respectively VRRP backup group 1 (wherein gateway 1 be main with, gateway 2 for standby) and VRRP backup group 2 (wherein gateway 2 be mainly use, gateway 1 is standby).The virtual ip address of each backup group is different with virtual mac address.The gateway of a part of main frame in the local area network (LAN) is set at the virtual ip address of VRRP backup group 1, and the gateway of a part of main frame is set at the virtual ip address of VRRP backup group 2 in addition.Like this, above-mentioned two parts main frame uses the virtual ip address of different VRRP backup groups as gateway ip address respectively, has realized function is shared in the statistical average of load.
Though above-mentioned prior art has solved the load balancing problem, it still exists a problem, promptly, in this technology, if will realize sharing of load, just must set different gateways respectively, and that this is the user is unacceptable usually to the main frame in the same local area network (LAN).In addition, in above-mentioned prior art, after All hosts was all specified gateway, the discharge model of gateway was also with regard to basic fixed.If desired the discharge model of gateway is adjusted again, its difficulty is relatively large.
Summary of the invention
Therefore, at the above-mentioned problems in the prior art, purpose of the present invention just provides a kind of like this system and method, and it can either realize that the load balancing of gateway and redundancy backup can solve the problem that main frame need dispose a plurality of gateway address again.
To achieve these goals, the invention provides a kind of offered load of in communication network, realizing and share the network communication method of function and redundancy backup function, described communication network comprises wide area network, local area network (LAN) main frame and is used to be connected a plurality of gateways of described wide area network and described local area network (LAN) main frame, described gateway comprises main with gateway and standby gateway, it is characterized in that, described gateway has identical virtual ip address and different virtual mac address, the gateway address of each host configuration of described local area network (LAN) is the virtual ip address of described gateway, and described method comprises:
(1) described main frame is when wide area network sends message, to the described ARP of LAN broadcast request;
(2) the main state of using gateway according to each described gateway is defined as the gateway that described main frame provides service, with the described virtual mac address that is confirmed as gateway described ARP request is made and being replied;
(3) after described main frame obtains the MAC Address of the described gateway that is determined, with described virtual mac address of replying the transmission message is carried out Ethernet and encapsulates, and be sent to described be determined service is provided gateway to transmit.
In above-mentioned implementation method of the present invention, described main gateway and the described standby gateway used come mutual monitor state by sending the HELLO message at interval mutually according to preset time, when failing to receive continuously, determine that then this gateway lost efficacy from HELLO message that certain gateway sends.
In above-mentioned implementation method, described monitoring step further comprises:
(4-1) when described standby gateway lost efficacy, determine that with gateway described gateway substitutes the standby gateway of described inefficacy by main;
(4-2) when described master was lost efficacy with gateway, determine the new main gateway of using by described standby gateway;
(4-3) described new master to determine that with gateway described gateway is taken over described inefficacy main with gateway handled main frame flow.
In the method for the invention, also can comprise and detect the step that the inefficacy gateway recovers, comprise:
(5-1) gateway to described inefficacy is provided with detection time, and described detection time is greater than the time interval of the network HELLO message transmission of described inefficacy gateway;
(5-2) recover just often when described inefficacy gateway, described inefficacy gateway detected, receive continuously in according to this detection time specify a HELLO message that sends from this inefficacy gateway after, just judge this inefficacy gateway recovery;
(5-3) when losing efficacy once more in the described inefficacy gateway recovery process, prolong detection time;
(5-4) after described inefficacy gateway has recovered the ability of service is provided, will return to detection time detection time of initial setting up.
In said method of the present invention, to determining by described master with gateway or undertaken of gateway that service is provided by the control unit that is provided with in the described local area network (LAN).
In addition, in said method of the present invention, described replying by described main with gateway, the described gateway that is determined or the described control unit execution in described local area network (LAN).
In addition, in the method for the invention, also can further comprise when the priority of described gateway changes, lead step with the switching of gateway and standby gateway.This master can adopt conversion mode or non-conversion mode to carry out with the switching of gateway and standby gateway.
In sum, the invention provides a kind of load balancing and redundancy backup function of simple multiple gateway outlet, and the realization of these functions is transparent to the main frame in the local area network (LAN).The interior main frame of local area network (LAN) does not need to do any particular arrangement and just can enjoy the benefit that load balancing and redundancy backup bring to the full.The present invention has made full use of the bandwidth of wide area network link and the disposal ability of outlet gateway.
Description of drawings
By detailed explanatory note and in conjunction with the following drawings, above-mentioned purpose of the present invention, feature and the advantage easy to understand more that will become, wherein:
Fig. 1 is the composition frame chart of a typical network communicating system;
Fig. 2 utilizes the VRRP technology to carry out the schematic diagram of transfer of data and gateway switching in the prior art;
Fig. 3 utilizes the VRRP technology to carry out the principle schematic that offered load is shared in the prior art;
The schematic diagram of Fig. 4 is used to help to illustrate to have the network communication method that offered load is shared function according to embodiments of the present invention;
Fig. 5 is the schematic diagram that each gateway sends the HELLO message in the embodiment of the present invention.
Embodiment
Describe the example of embodiment of the present invention in detail with reference to figure 4.Be noted that at first that in the present invention main notion with gateway and standby gateway is different with the notion of prior art in fact.This be because, mainly in the prior art provide service for main frame in the local area network (LAN) with gateway, standby gateway then is in idle state.And in the present invention, main difference with gateway and standby gateway only is mainly have controlled function and standby gateway does not have with gateway, is still mainly all providing service for main frame with gateway and standby gateway.This point obtains embodying in the description to embodiment of the present invention below.
The present embodiment be local area network (LAN) host configuration identical virtual gateway address, this point is obviously different with above-mentioned prior art.As shown in Figure 4, the gateway address of main frame is the virtual ip address of gateway in the present embodiment.Use gateway if gateway 1 is the master, gateway 2 is a standby gateway, and then the gateway address of main frame is 10.1.1.3.In this example, main frame is finished the forwarding of data message by these two gateways.It is the message of virtual mac address A:00.00.00.00.00.03 that gateway 1 receives all target MAC (Media Access Control) address, and it is the message of virtual mac address B:00.00.00.00.00.04 that gateway 2 is responsible for receiving all target MAC (Media Access Control) address of processing.
When host B will at first will send the ARP request to gateway when wide area network sends datagram.This request is for broadcasting packet, so gateway 1 and gateway 2 can both receive this request.Because the gateway address of host B is the virtual ip address of gateway in this example, therefore, after the master receives this request with gateway 1, state according to present each gateway, decision is the host B service by which gateway, and method is to use this appointment gateway corresponding virtual MAC Address that the ARP request of host B is made and replys.Make reply can be main can be other gateways also with gateway, be preferably and use the gateway that is designated as the host B service to make to reply.So just can main control system B the outlet gateway of visit wide area network, also just can reach the purpose of load statistics equilibrium.
Main make the condition of above-mentioned decision institute foundation with gateway can be very flexible, for example can be according to following at least a condition: according to outlet bandwidth assign host machine or one group of main frame (as server) with same characteristic features is assigned to identical gateway service or mean allocation main frame to each gateway, also can carry out flexible allocation according to the strategies such as bandwidth load of go out interface bandwidth load or lan interfaces recently.The data that so just can solve the visit wide area network of local area network (LAN) main frame are shared problem.On the other hand, this technology is transparent to main frame, and can be which host services according to which gateway of condition flexible dispatching.
Suppose that the master should provide service with gateway 1 by gateway 2 according to the condition judgment host B, so just use virtual mac address B:00.00.00.00.00.04 to reply, replying gateway can be that the master replys with gateway notification gateway 2 with gateway or master, even also can notify other gateways to reply with gateway by main, host B will all be encapsulated as virtual mac address B:00.00.00.00.00.04 transmission to the purpose MAC of message like this.Gateway 2 receives this message and transmits.On the other hand, if host A wants to send data to wide area network, to send the ARP request equally earlier.At this moment, if the master should provide service for host A with gateway 1 by oneself according to conditional decision, then will be with virtual mac address A:00.00.00.00.00.03 to this arp reply, main frame 1 will all be encapsulated as virtual mac address A:00.00.00.00.00.03 to the purpose MAC of message and send like this, and the master just receives this message with gateway 1 and transmits like this.
By the way, make the main frame of local area network (LAN) have identical gateway ip address, but the data of actual access wide area network may be passed through different gateways.Its basic reason is, according to the solution of the present invention, the identical virtual gateway IP address that is disposed for the local area network (LAN) main frame, different ethernet hardware addresses can be provided, MAC Address just, and these MAC Address based on this, have just formed of the present invention to the transparent load balancing scheme of local host machine corresponding to different gateways.
When the data that send to the host A of local area network (LAN) when wide area network arrived gateway, no matter gateway was main with gateway or standby gateway, if this gateway does not have the MAC Address of host A, this gateway can send ARP and asks in local area network (LAN) so.Wherein the source IP of the ARP request message IP address gateway 1 for example that uses this gateway oneself promptly carries in the source IP field in the ARP request message: 10.1.1.1.Receive the ARP request message when host A, will use oneself MAC Address that this request is made and reply that this gateway just can carry out the Ethernet encapsulation to the MAC Address that the above-mentioned message of going to host A is replied with host A like this, sends to host A then.
In the foregoing description of the inventive method, the master bears forwarding work simultaneously with gateway and standby gateway.Lead with gateway by finishing flow scheduling for the different gateway of host assignment.After guaranteeing that gateway lost efficacy, by the main frame of inefficacy gateway service still can be normally and external network carry out data communication, just need monitor each gateway.In the present invention, comprise that the master is mutual with the monitoring of each gateway of gateway and standby gateway.This monitoring realizes by sending HELLO message (monitoring message) mutually.The HELLO message is a kind of communication message of equipment room, and this message uses multicast to send usually, can receive processing to guarantee each gateway, and its major function is to make other gateways monitoring HELLO message sender's operating state (whether operate as normal); Can make other gateways understand current HELLO message sender's correlation behavior information (taking situation) such as link.
In one embodiment of the invention, can comprise in the HELLO message: priority facility, virtual ip address, gateway device ID, gateway device ID and this gateway processes virtual mac address, outbound available bandwidth, link take situation, auth type, authentication word, the HELLO message time interval, message checking and etc.This HELLO message uses multicast to send usually, can receive processing to guarantee each gateway.
Wherein, which gateway is priority be used to determine as the main gateway of using.When the master was lost efficacy with gateway, each standby gateway can determine who is the main gateway of using by this priority.Virtual ip address is for the address of gateway is provided in the local area network (LAN).Virtual mac address and which gateway of device id illustrated together should be handled which or which virtual mac address.Link takies situation and the outbound available bandwidth is the main foundation that host schedules is provided with gateway.Auth type and authentication word are for security consideration, are used to guarantee that each gateway is legal gateway.
This HELLO message of gateway can be according to the own time interval information that carries in this HELLO message, with this time interval is the cycle outwards to send the HELLO message, and other gateways are judged whether operate as normal of other gateway according to this HELLO message that whether can receive other gateway.When other gateways can't then be judged gateway fails when certain gateway receives the HELLO message of specifying number continuously.Fig. 5 has schematically shown the situation that sends this HELLO message between the gateway.As shown in Figure 5, each gateway in the local area network (LAN) respectively other gateway in this local area network (LAN) send the HELLO message.For example gateway 1 sends the HELLO message to gateway 2 and 3, and gateway 2 sends the HELLO message to gateway 1 and 3, or the like.
Like this, when standby gateway lost efficacy, which gateway the virtual mac address of the main standby gateway that can lose efficacy according to certain conditional decision with gateway should be taken over by, and just the gateway by this inefficacy provides the main frame of service need adjust to other and can provide service on the gateway of operate as normal originally.This condition can comprise, for example, and the link operating position of each present gateway, the main frame number of the service of each gateway, user's specified configuration etc.
Specifically, lost efficacy if detect standby gateway, then other standby gateway are not done any action, the main standby gateway work of then according to certain which gateway of strategy decision taking over inefficacy with gateway.Its method is to specify a kind of type of message in the HELLO message, and such message includes following field:
HELLO message version number, HELLO type of message, send gateway device ID, priority facility, virtual IP address, gateway device ID and the virtual MAC table of this gateway processes, outbound available bandwidth, link take situation, auth type, authentication word, the HELLO message time interval, message checking and etc. field.
The main virtual mac address table of handling by gateway device ID and this gateway device of appointment with gateway is notified the gateway of taking over failing gateway, to take over the service of inefficacy gateway.That is to say when gateway discovery oneself needs to handle the message of certain inefficacy gateway, it finishes the host service switching by the mode that sends such ARP request message so, sender's protocol address in this ARP request message can be the IP address of this gateway, but can not be virtual ip address, target protocol address in this ARP request message also is the IP address of this gateway, source MAC in the Ethernet encapsulation of this ARP request message is the virtual mac address of inefficacy gateway, and the message that this appointed gateway is opened simultaneously to the virtual mac address of inefficacy gateway receives processing.
When the master was lost efficacy with gateway, at first detect main standby gateway main signal with gateway is re-elected in transmission with the gateway inefficacy, this can pass through to send one type HELLO message realization.After new master determined with gateway, new master will issue the gateway that lost efficacy with gateway was somebody's turn to do the instruction of being taken over job by which gateway.Thereafter operation can be identical with above-mentioned standby gateway crash handling method.More than the switch operating of fault gateway has just been finished in operation.
Specifically, main with the gateway inefficacy if a certain standby gateway detects at first, this standby gateway should be sent the HELLO message immediately.The type of this message is the main type of message with gateway of election, can comprise following field:
HELLO message version number, HELLO type of message, send gateway device ID, priority facility, virtual IP address, gateway device ID and the virtual MAC table of this gateway processes, outbound available bandwidth, link take situation, auth type, authentication word, the HELLO message time interval, message checking and etc. field.
Other standby gateway are received the main HELLO message with gateway type of this election, promptly send the main message with gateway type of election and use gateway to re-elect the master.For example, the gateway among Fig. 4 can become the main gateway of using 2 this moments.Then, whom the master determines take over original main flow of using those main frames of gateway processes with gateway according to strategy.
The following describes main selected process with gateway:
In one embodiment of the invention, main is to determine according to the priority of gateway device with gateway.For example, supposing has three gateways in certain local area network (LAN), and its priority facility is respectively 100,80 and 60, and priority facility numerical value is big more, this priority facility high more (certainly opposite setting also can).While supposes that in this example the main condition with gateway of election is: with the highest gateway of priority facility serves as that the master uses gateway, compares gateway device ID if priority facility is identical, and the master that just becomes that ID is big uses gateway.Work as gateway like this by multicast form transmission HELLO message separately, this message promptly is labeled as election HELLO message.This HELLO message that each gateway device all receives according to oneself compares, to determine the main gateway of using.In this example, gateway 1 will become the main gateway of using.This election type HELLO message of equipment room can comprise following field:
HELLO message version number, HELLO type of message, send gateway device ID, priority facility, virtual IP address, gateway device ID and the virtual MAC table of this gateway processes, outbound available bandwidth, link take situation, auth type, authentication word, the HELLO message time interval, message checking and etc. field.
Explanation is main with the selection of gateway to the gateway of main frame in more detail below.
In embodiments of the invention, main is which virtual MAC decision uses reply for the ARP request of each main frame according to certain strategy with gateway.This strategy can comprise: 1) the static particular host of specifying is through the particular gateway service; 2) average wheel flow distribution (each gateway provides the service of N/M main frame, and N is a main frame number in the network, and M is the gateway number); 3) exporting total available bandwidth ratio according to static state dispatches; 4) taking situation according to outlet bandwidth dynamically adjusts; 5) dispatch according to Ethernet port bandwidth occupancy situation.
In addition, after main frame rolled off the production line, gateway can detect this situation by the aging mechanism of ARP.Gateway can be by the information about the main frame number handled of carrying in the HELLO message, makes other gateways can understand each gateway and be responsible for the main frame number handled at present.So that the main distribution of adjusting main frame with gateway according to this situation.
That can select main have two kinds with, standby gateway switching mode, promptly converts formula and non-conversion formula.When using the conversion formula, when the priority of gateway device in the communication system changes, may cause main pressure to be switched with gateway.Forcing to switch the condition that takes place is as long as the current main priority height with gateway of the priority ratio that equipment is arranged in the gateway will be re-elected with gateway main.Take place condition that gateway priority changes have usually under the priority facility of the gateway situation relevant with the outlet available bandwidth, under the situation of the manual priority of assignment gateway device again or the master of losing efficacy recover with gateway etc. under the situation, just may cause the master to use the switching of gateway.At this moment, need will send as new main gateway and once re-elect main HELLO message, so that each gateway begins again the selected main gateway of using with gateway type with gateway.
For the switching of non-conversion formula, though priority facility is arranged in the gateway, do not lead conversion with gateway than main priority height with gateway yet, take place with the gateway election up to main next time.At this moment, it is stable that the stand-by state of each gateway keeps, and do not send because of the variation of priority to re-elect main re-electing with the HELLO message of gateway type and mainly use gateway, uses gateway but re-elect the master by the time next time.
For example, after the inefficacy gateway recovers, can use the hand-off process of conversion or non-conversion.No matter adopt above-mentioned which kind of mode, main all can give back this inefficacy gateway processes after the recovery to the virtual MAC that before is responsible for handling by this inefficacy gateway with gateway.In one embodiment of the present invention, the inefficacy gateway is provided with a detection time.Be meant this detection time normally detect continuously specify the monitoring message of number after, just can think the time that this inefficacy gateway recovers.This detection time is greater than the time interval of the network HELLO message transmission of inefficacy gateway.After gateway lost efficacy, will be to lengthening detection time of this gateway that lost efficacy.To detection time of inefficacy gateway, the unstable situation of the gateway of can avoiding once losing efficacy exerts an influence to system communication by lengthening.Each when unsettled situation generation, prolong, make the instability of inefficacy gateway the web influence minimum with regard to tackling detection time mutually.
In embodiments of the invention, also consider the situation of new adding gateway.In the case, initiate gateway at first sends the HELLO message, and other gateways in network are stated the adding of this new gateway.Other gateway correspondingly upgrades its relevant state.Simultaneously, this initiate gateway also receives the HELLO message of other gateways, upgrades its correlation behavior.The master, after new adding gateway receives the HELLO message of specifying number continuously from this, begins it is carried out service dispatching with gateway after detecting this initiate gateway.The main detection with gateway newly adds the process that gateway whether can dispatching services and detects the process basically identical that the inefficacy gateway recovers.
Although describe the present invention above, should be appreciated that these illustrate the example of just having enumerated concrete enforcement more of the present invention, its objective is in order to make the public understand spirit of the present invention and essence better.Therefore, those skilled in the art obviously can carry out various modifications and adopt the various means that are equal to each details of the invention process.Therefore, scope of the present invention is only limited by claims.
For example, in above-mentioned these embodiments, also can in local area network (LAN), dispose control unit (for example on server) separately.Take over the master by this control unit and be chosen as the gateway that main frame provides service with gateway.In addition, obviously also can come operations such as main monitoring, switching in the control area net by control unit with gateway and standby gateway.In this case, can reply the ARP request of main frame by the equipment in the control unit designated local area network with ARP function with the virtual mac address of appointed gateway.And be not limited only to reply by appointed gateway.
Claims (11)
1. realize that in communication network offered load shares the functional network communication means for one kind, described communication network comprises wide area network, local area network (LAN) main frame and is used to be connected a plurality of gateways of described wide area network and described local area network (LAN) main frame, described gateway comprises main with gateway and standby gateway, it is characterized in that, described gateway has identical virtual ip address and different virtual mac address, the gateway address of each host configuration of described local area network (LAN) is the virtual ip address of described gateway, and described method comprises:
(1) described main frame is when wide area network sends message, to the described ARP of LAN broadcast request;
(2) mainly be defined as the gateway that described main frame provides service according to the state of each described gateway, and described ARP asked to make to reply with the described virtual mac address that is confirmed as gateway with gateway;
(3) after described main frame obtains the MAC Address of the described gateway that is determined, with described virtual mac address of replying the transmission message is carried out Ethernet and encapsulates, and be sent to described be determined service is provided gateway to transmit.
2. method according to claim 1, it is characterized in that, further comprise: described main gateway and the described standby gateway used come mutual monitor state by sending the HELLO message at interval mutually according to preset time, when fail to receive continuously specify number from HELLO message that certain gateway sends the time, determine that then this gateway lost efficacy.
3. method according to claim 2 is characterized in that, further comprises:
(4-1) when described standby gateway lost efficacy, determine that with gateway described gateway substitutes the standby gateway of described inefficacy by main;
(4-2) when described master was lost efficacy with gateway, determine the new main gateway of using by described standby gateway;
(4-3) described new master determines to take over the main gateway with the handled main frame flow of gateway of described inefficacy with gateway.
4. method according to claim 2 is characterized in that, further comprises the steps:
(5-1) gateway to described inefficacy is provided with detection time, and described detection time is greater than the time interval of the network HELLO message transmission of described inefficacy gateway;
(5-2) recover just often when described inefficacy gateway, described inefficacy gateway detected, receive continuously in according to above-mentioned detection time specify a HELLO message that sends from this inefficacy gateway after, just determine this inefficacy gateway recovery;
(5-3) when losing efficacy once more in the described inefficacy gateway recovery process, prolong set detection time;
(5-4) after described inefficacy gateway has recovered the ability of service is provided, will return to detection time detection time of initial setting up.
5. method according to claim 1 is characterized in that, further comprises:
(7-1) comprise that the master realizes mutual monitoring with the monitoring of each gateway of gateway and standby gateway by mutual transmission HELLO message, after the main frame in the described local area network (LAN) rolls off the production line, by the information about the main frame number handled of in the HELLO of described gateway message, carrying, notify other gateways; With
(7-2) master according to the situation that described main frame rolls off the production line, adjusts the distribution of each gateway host with gateway.
6. method according to claim 1 is characterized in that, comprises that the master realizes mutual monitoring with the monitoring of each gateway of gateway and standby gateway by mutual transmission HELLO message, further comprises the step of new adding gateway, comprising:
(8-1) gateway of described adding other gateway in local area network (LAN) sends the HELLO message;
(8-2) after described other gateways correctly detect the gateway of this adding, described other gateway correspondingly upgrades its relevant state;
(8-3) described master, begins it is carried out service dispatching with gateway after new adding gateway receives the HELLO message of specifying number from this.
7. require each described method according to aforesaid right, it is characterized in that, gateway that service is provided determined to be carried out with the control unit that is provided with in gateway or the described local area network (LAN) by described master.
8. method according to claim 7 is characterized in that, described replying by described main with gateway, the described gateway that is determined, other described gateway in described local area network (LAN) or the described control unit execution in described local area network (LAN).
9. method according to claim 7 is characterized in that, is defined as the gateway that described message transmits provides service according to following at least a condition:
The outlet bandwidth ratio of gateway;
The main frame that will have same characteristic features is returned identical gateway service;
The go out bandwidth load of interface bandwidth load or lan interfaces;
The mode of each gateway mean allocation main frame;
The specified gateway of main frame of the appointment by static configuration.
10. method according to claim 7 is characterized in that, described main be that priority facility and device id according to gateway decides with gateway and standby gateway.
11. method according to claim 7 is characterized in that, further comprises: when the priority of described gateway changes, lead the switching with gateway and standby gateway, described switching mode adopts any in the following dual mode:
The conversion mode, when the satisfied master of generation selected the situation of condition again with gateway, promptly selected again master used gateway;
Non-conversion mode, after take place satisfying main situation with gateway selected condition again, wait until always main with gateway lost efficacy or artificial mandatory requirement selected again main when use gateway, just selected again master uses gateway.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN031197582A CN100407671C (en) | 2003-03-11 | 2003-03-11 | Network communication method for carrying out load division |
| PCT/CN2004/000192 WO2004082222A1 (en) | 2003-03-11 | 2004-03-11 | Communication method having the function of partaking the network load |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN031197582A CN100407671C (en) | 2003-03-11 | 2003-03-11 | Network communication method for carrying out load division |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1531262A CN1531262A (en) | 2004-09-22 |
| CN100407671C true CN100407671C (en) | 2008-07-30 |
Family
ID=32968473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN031197582A Expired - Lifetime CN100407671C (en) | 2003-03-11 | 2003-03-11 | Network communication method for carrying out load division |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN100407671C (en) |
| WO (1) | WO2004082222A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2993838A4 (en) * | 2013-05-29 | 2016-09-21 | Huawei Tech Co Ltd | Method for setting identity of gateway device and management gateway device |
Families Citing this family (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100459601C (en) * | 2005-06-23 | 2009-02-04 | 华为技术有限公司 | Method for realizing active/standby gateway apparatus in network |
| EP1897339B1 (en) * | 2005-06-30 | 2013-08-07 | Telefonaktiebolaget LM Ericsson (publ) | Mapping an original mac address of a terminal to a unique locally administrated virtual mac address |
| CN1859285A (en) * | 2005-11-05 | 2006-11-08 | 华为技术有限公司 | Method for equalizing load between gateways of WiMAX access service network |
| CN101009640B (en) * | 2005-11-05 | 2011-04-06 | 华为技术有限公司 | Service forwarding method between the gateways in the service network and its service gateway |
| JP4652285B2 (en) * | 2006-06-12 | 2011-03-16 | 株式会社日立製作所 | Packet transfer device with gateway selection function |
| US8699327B2 (en) * | 2007-01-31 | 2014-04-15 | Alcatel Lucent | Multipath virtual router redundancy |
| CN101052047B (en) * | 2007-05-22 | 2010-06-09 | 网御神州科技（北京）有限公司 | Load equalizing method and device for multiple fire-proof wall |
| CN101159701B (en) * | 2007-12-04 | 2010-06-09 | 杭州华三通信技术有限公司 | VRRP based router dynamic bandwidth assignment method and system |
| CN101217448B (en) * | 2008-01-18 | 2013-01-30 | 福建星网锐捷网络有限公司 | Method and system to realize gateway dynamic load sharing |
| JP5074290B2 (en) * | 2008-05-13 | 2012-11-14 | 株式会社日立国際電気 | Redundancy switching system, redundancy management device and application processing device |
| CN101588325B (en) * | 2008-05-20 | 2013-08-07 | 华为技术有限公司 | Disaster recovery method, device and system based on wireless packet gateways |
| CN102196529A (en) * | 2010-03-19 | 2011-09-21 | 中兴通讯股份有限公司 | Access method and system for wireless sensor network |
| CN102209064B (en) * | 2010-03-29 | 2014-10-22 | 杭州华三通信技术有限公司 | Method of using VRRP to provide backup for access equipment and VRRP gateway equipment |
| CN102143046B (en) * | 2010-08-25 | 2015-03-11 | 华为技术有限公司 | Load balancing method, equipment and system |
| CN102387514B (en) * | 2010-08-27 | 2016-06-29 | 中兴通讯股份有限公司 | The backup method of wireless sensor network gateway and system |
| US8559299B2 (en) * | 2010-11-30 | 2013-10-15 | Telefonaktiebolaget L M Ericsson (Publ) | Mobile gateways in pool for session resilience |
| WO2012126208A1 (en) * | 2011-03-18 | 2012-09-27 | 中兴通讯股份有限公司 | Method and system for wireless sensor network connecting to telecommunication network |
| CN102223365B (en) * | 2011-06-03 | 2014-02-12 | 杭州华三通信技术有限公司 | User access method and device based on SSL (Secure Socket Layer) VPN (Virtual Private Network) gateway cluster |
| CN102333027B (en) * | 2011-06-17 | 2014-05-07 | 杭州华三通信技术有限公司 | Traffic load sharing realization method based on virtual router redundancy protocol extend (VRRPE) backup group and realization apparatus thereof |
| CN102957755B (en) * | 2011-08-22 | 2018-06-19 | 中兴通讯股份有限公司 | A kind of address resolution method, device and information transferring method |
| CN102624929B (en) * | 2012-04-20 | 2015-04-08 | 武汉邮电科学研究院 | Wireless sensor network gateway redundancy method based on virtual gateway |
| CN102647360B (en) * | 2012-04-26 | 2015-02-18 | 杭州华三通信技术有限公司 | Method and equipment for transmitting messages in VRRPE (virtual router redundancy protocol equilibrium) |
| CN102801623B (en) * | 2012-08-15 | 2015-07-22 | 杭州华三通信技术有限公司 | Multi-access data transmitting method and device |
| CN102932500A (en) * | 2012-11-07 | 2013-02-13 | 曙光信息产业股份有限公司 | Method and system for taking over fault interface node |
| CN103888994B (en) * | 2012-12-21 | 2017-11-21 | 中国科学院计算技术研究所 | A kind of multiple gateway processing method and system with hot calamity for ability |
| CN103259740B (en) * | 2013-04-27 | 2017-03-15 | 杭州华三通信技术有限公司 | A kind of load balance process method and device |
| CN103607293B (en) * | 2013-10-30 | 2017-08-22 | 新华三技术有限公司 | A kind of flow rate protecting method and equipment |
| CN104811375A (en) * | 2015-05-08 | 2015-07-29 | 三星电子（中国）研发中心 | Smart home gateway and networking method thereof |
| CN105357674B (en) * | 2015-10-26 | 2023-05-16 | 青岛海尔智能家电科技有限公司 | Data distribution method, device and system |
| CN107948086A (en) * | 2016-10-12 | 2018-04-20 | 北京金山云网络技术有限公司 | A kind of data packet sending method, device and mixed cloud network system |
| CN108551679B (en) * | 2018-04-18 | 2021-05-18 | 深圳市银河风云网络系统股份有限公司 | Wireless device communication method and system based on local area network Mesh technology |
| CN108259283B (en) * | 2018-04-19 | 2021-01-05 | 深圳市银河风云网络系统股份有限公司 | Wireless device communication method and system based on wide area network Mesh technology |
| CN110611590A (en) * | 2019-09-16 | 2019-12-24 | 南京国电南自电网自动化有限公司 | Method and system for internet of things gateway communication backup |
| CN111786840A (en) * | 2020-07-27 | 2020-10-16 | 平安证券股份有限公司 | Gateway switching method and device based on network access platform |
| CN113285874B (en) * | 2021-05-20 | 2022-03-25 | 烽火通信科技股份有限公司 | Gateway backup method and device based on VRRP single instance |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5949753A (en) * | 1997-04-11 | 1999-09-07 | International Business Machines Corporation | Redundant internet protocol gateways using local area network emulation |
| US6091732A (en) * | 1997-11-20 | 2000-07-18 | Cisco Systems, Inc. | Method for configuring distributed internet protocol gateways with lan emulation |
| CN1357997A (en) * | 2000-12-15 | 2002-07-10 | 华为技术有限公司 | Virtual local area network access method in Ethernet access network |
| JP2002368776A (en) * | 2001-06-07 | 2002-12-20 | Sony Corp | Information processing unit and method, information processing system, recording medium, and program |
-
2003
- 2003-03-11 CN CN031197582A patent/CN100407671C/en not_active Expired - Lifetime
-
2004
- 2004-03-11 WO PCT/CN2004/000192 patent/WO2004082222A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5949753A (en) * | 1997-04-11 | 1999-09-07 | International Business Machines Corporation | Redundant internet protocol gateways using local area network emulation |
| US6091732A (en) * | 1997-11-20 | 2000-07-18 | Cisco Systems, Inc. | Method for configuring distributed internet protocol gateways with lan emulation |
| CN1357997A (en) * | 2000-12-15 | 2002-07-10 | 华为技术有限公司 | Virtual local area network access method in Ethernet access network |
| JP2002368776A (en) * | 2001-06-07 | 2002-12-20 | Sony Corp | Information processing unit and method, information processing system, recording medium, and program |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2993838A4 (en) * | 2013-05-29 | 2016-09-21 | Huawei Tech Co Ltd | Method for setting identity of gateway device and management gateway device |
| US10419531B2 (en) | 2013-05-29 | 2019-09-17 | Huawei Technologies Co., Ltd. | Method for setting gateway device identity, and management gateway device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2004082222A1 (en) | 2004-09-23 |
| CN1531262A (en) | 2004-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100407671C (en) | Network communication method for carrying out load division | |
| CN102315975B (en) | Fault processing method based on intelligent resilient framework (IRF) system and equipment thereof | |
| CN101404621B (en) | Method and routing device for implementing VRRP load balance | |
| AU770985B2 (en) | Fault-tolerant networking | |
| CN101160836B (en) | Redundancy protection method for bridge mode resilient packet ring | |
| CN101316236B (en) | Vrrp backup group load sharing method and router | |
| KR101591102B1 (en) | Method for router of virtual router redundancy protocol and communication system therefor | |
| CN100493086C (en) | Method for realizing virtual IP protocol | |
| US20050265230A1 (en) | Apparatus and method for performing state transition of backup router in router redundancy system | |
| CN1973499A (en) | Method for switching network connecting device in network system and network system | |
| US20060206611A1 (en) | Method and system for managing programs with network address | |
| CN101902360A (en) | Method for synchronizing information between master equipment and standby equipment and server | |
| US20110231508A1 (en) | Cluster control system, cluster control method, and program | |
| CN104104570A (en) | Aggregation processing method in IRF (Intelligent Resilient Framework) system and device | |
| CN100388703C (en) | A method and system for Ethernet interface node backup | |
| CN101827039B (en) | Method and equipment for load sharing | |
| CN102447615A (en) | Switching method and router | |
| CN100492984C (en) | Method for realizing data transfer backup through address interpretation protocol messages | |
| CN102209035B (en) | Traffic forwarding method and devices | |
| CN107294845A (en) | VRRP load-balancing methods and device | |
| CN102238092B (en) | Method for performing load sharing on encoder and encoder | |
| US7738362B2 (en) | System and method for address notification in a network | |
| CN1330137C (en) | Method for sending address analytic protocol request message into specific object in local area network | |
| JP2009003491A (en) | Server switching method in cluster system | |
| JPH05316116A (en) | Unified management equipment for server standby system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term | ||
| CX01 | Expiry of patent term |
Granted publication date: 20080730 |