CN101237413A - Method for realizing high-availability of control part under forward and control separated network architecture - Google Patents
Method for realizing high-availability of control part under forward and control separated network architecture Download PDFInfo
- Publication number
- CN101237413A CN101237413A CNA2008100597962A CN200810059796A CN101237413A CN 101237413 A CN101237413 A CN 101237413A CN A2008100597962 A CNA2008100597962 A CN A2008100597962A CN 200810059796 A CN200810059796 A CN 200810059796A CN 101237413 A CN101237413 A CN 101237413A
- Authority
- CN
- China
- Prior art keywords
- control piece
- control
- forwarding element
- available
- forwarding
- 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.)
- Granted
Links
Images
Abstract
The invention discloses a method for realizing the high availability of control elements under the transmitting and the controlling of separating a networkware framework, which comprises the following steps: 1) an available control element list is determined through a control element management unit and informs a forwarding element through a forwarding element management unit; 2) each control element monitors the change of the states of other control elements and maintains the available control element list and the load condition; 3) after the forwarding element acquires the available element list, the forwarding element asks the control elements for the available control element lists, the load condition and a load balancing method one by one and selects the control element actively; 4) more than two control elements bear the network service function facing to a plurality of forwarding elements and trigger the forwarding elements to switch the control elements according to the load condition of the control elements; 5) data is backup among a plurality of control elements to realize the consistency of related data. The invention adopts a plurality of control elements to increase the availability of the control elements and improves the overall performance of the networkware through the load balance.
Description
Technical field
The invention belongs to the communications field, be specifically related to a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework.
Background technology
In the immanent society of this information, the most critical issue that computing system and communication system are faced is how to set up and keep the stability of system and the continuation of operation.Concerning the network equipment, mainly be to utilize computer system that information and data processing reliably are provided in time, computer hardware and software all can break down inevitably, and these faults might be brought great loss to the network equipment, even termination, the paralysis of the service of whole network.For those network equipments that is used for critical data communication, any shut-down all will produce serious property loss.Have influence on the prestige of enterprise or company in addition owing to network paralysis, cause the client that enterprise is lost confidence, such loss will be fatal.The high availability HA of system (HighAvailability) seems extremely important for this reason, high-availability arrangement utilization redundant component still less, by the software detection fault, in case taking place to isolate immediately, fault damages parts, by being provided, fault recovery realizes the availability of using.The HA system designs the HA characteristic that improves system with hardware platform and two aspects of software at present.The implementation method of hardware platform mainly comprises: the high availability structure of the monitoring of hardware redundancy backup, hardware fault and isolation, enhancing etc.Method of software realization mainly comprises: the running status of monitoring local task or thread, when backup (hardware redundancy) task or thread to the opposite end are monitored, and find software fault and report and handle (isolate, recover) and the result that handles to opposite end (during the hardware redundancy backup) report etc.
Along with the rapid expansion in applications of computer network fields such as the Internet, new features, new demand constantly occur and constantly change, and this generation network equipment of will looking for novelty should have enough flexibilities, can make response to new business, new demand apace; Require new generation network equipment should have enough openings, make the user can be according to the resource flexible combination of being opened, so that different network demand services to be provided; Require new generation network equipment should have enough modular natures, and carry out standardization by standardization body, make each module to develop by the different manufacturers independent studies, can become an integral body by the cordwood system type organic assembling again, this cordwood development model can greatly be accelerated the Products Development process, also help benign competition simultaneously, reduce cost, the development of personal computer (PC) has fully proved the validity of this pattern.The networkware that forwarding separates with control has obtained broad research and application at present just because of the above requirement of having satisfied new generation network.Carry out about transmitting mainly to formulate to organize under the IETF guiding at present in internet standard with the research of control separated network spare, IETF has set up in special ForCES (Forwarding and Control ElementSeparation transmits and separates with the control piece) working group.
The typical structure of forwarding and control separated network spare as shown in Figure 1.The networkware that forwarding separates with control is divided into forwarding element and control piece two parts, stipulated among the Internet protocol RFC3654 to transmit and the demand (Requirements) of controlling separated network spare, Internet protocol RFC3746 has stipulated the framework (Framework) of forwarding with control separated network spare.Information exchange between them uses the Information Exchange Agreement of standardized being called as " ForCES agreement " (to see Avri Doria, Wang Weiming etc.: " ForCES Protocol Specification ", http://www.ietf.org/internet-drafts/draft-ietf-ForCES-protocol-11.txt), and then transmit with controlling separated network spare and be also referred to as the ForCES networkware, standardization makes that forwarding element and control piece can be by different manufacturers productions.Redundancy is the main path that obtains high availability, even use the assembly that may break down, redundancy also can make the speed of recovering from chance failure with inside the plan upgrading faster.A ForCES networkware can be made up of a plurality of forwarding elements and a plurality of control piece, and wherein the use of a plurality of control pieces is exactly for the system redundancy purpose, thereby realizes the high availability of control piece.
The standardized method of a plurality of control piece information exchanges is one of core contents of ForCES networkware technical research.It comprises two aspect research contents at least: the formal definition method of the information exchange of a plurality of control pieces and a plurality of control piece are realized the mechanism of system redundancy.ForCES group research of exploration to this at present, research contents is called the redundant and recovery policy (CE redundancy and failover) of control piece, can abbreviate CERFPolicy as.The present redundancy that has only proposed to realize control piece by the active and standby pattern of employing, and to after the inefficacy of forwarding element discovery control piece, proposed two kinds of implementation strategies: 1, forwarding element stops forwarding, forwarding element rebulids with standby control piece and is connected then; 2, forwarding element can continue to transmit packet, initiatively standby control piece is switched to main control element then.Mechanism and the information interaction ForCES agreement concrete about CERFPolicy also do not illustrate clear fully.So, a kind ofly can be used for carrying out between a plurality of control pieces the information exchange mechanism of redundancy backup and the major issue that agreement becomes the current needs solution of ForCES networkware technical research by standardized.
Summary of the invention
The purpose of this invention is to provide a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework.
Comprise the steps:
Determine available control piece tabulation by the control piece administrative unit when 1) initial, and inform forwarding element by the forwarding element administrative unit;
2) each control piece is monitored the variation of other control piece states, all tabulation of available control piece and its loading conditions in the maintaining network spare;
3) the available control piece tabulation of forwarding element by obtaining from the forwarding element administrative unit, all available control piece tabulation, loading condition and load-balancing methods in the networkware that control piece request control piece is safeguarded are one by one initiatively selected control piece then;
4) two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, and many control pieces are not distinguished main or backup, and equity fully each other triggers forwarding element switching controls part according to the loading condition of control piece, realizes load balancing;
5) between many control pieces, carry out data backup, realize the consistency of related data.
Describedly determine available control piece tabulation by the control piece administrative unit when initial, and inform forwarding element: determine available control piece tabulation by the control piece administrative unit when initial by the forwarding element administrative unit, be to use configuration file or order to inform the IP address of available control piece in the control piece administrative unit system in advance by administrative staff, the forwarding element administrative unit is obtained its available control piece tabulation by carrying out exchanges data with the control piece administrative unit afterwards,, when the forwarding element administrative unit is inquired, available control piece tabulation is informed at forwarding element.
Described each control piece is monitored the variation of other control piece states, all tabulation of available control piece and its loading conditions in the maintaining network spare: each control piece is by the monitoring of regular packet exchange realization to other control piece state variation, and wherein the control piece state comprises whether cpu busy percentage, internal memory operating position and the control piece of control piece can normally send heartbeat signal; The controlled product that can normally send heartbeat signal is defined as available control piece, and the relevant information by all available control pieces in the form record networkware, comprises IP address and its state information.
The available control piece tabulation of described forwarding element by obtaining from the forwarding element administrative unit, all available control piece tabulations in the networkware that control piece request control piece is safeguarded one by one then, loading condition and load-balancing method, initiatively select control piece: forwarding element is by inquiring about to the forwarding element administrative unit, obtain available control piece tabulation, the available control piece of safeguarding to its control piece of available control piece request is one by one tabulated and state information then, by relatively the cpu busy percentage of all control pieces and the operating position of internal memory select the control piece of load minimum as the own control piece that will be connected.
Described two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, many control pieces are not distinguished main or backup, reciprocity fully each other, loading condition according to control piece triggers forwarding element switching controls part, realize load balancing: two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, many control pieces are not distinguished main or backup, and are reciprocity fully each other; The threshold value that gap between the control piece of the control piece of load maximum and load minimum sets greater than the keeper, to trigger forwarding element switching controls part, first forwarding element that the control piece of load maximum is controlled is controlled by the control piece of load minimum, realizes load balancing.
Describedly between many control pieces, carry out data backup, realize the consistency of related data: between many control pieces, carry out data backup, each control piece will backup to other one or more control pieces with forwarding element of being controlled and the required data of operational network service of oneself, by timing data synchronously and bursty data upgrade the consistency of dual mode realization Backup Data.
The present invention solves the major issue of in realization of ForCES technology and the technology standardization process CERFPolicy strategy being badly in need of solution, be a kind of effectively enforceable many control pieces redundancy backup mechanism, and defined the method for carrying out information exchange between forwarding element and control piece.Work from IETF ForCES working group, the present invention is that an at present unique cover has total solution, can carry out standardized many control pieces redundancy backup method to the application of IETF ForCES working group, yet there are no other any effective solution and delivers.Remarkable advantage of the present invention has: the demand that 1) has satisfied many control pieces redundancy backup; 2) by load-balancing mechanism, given full play to the performance of each control piece, help improving the service performance that networkware externally provides; 3) the present invention has been used by real system and has proved practicable; 4) can be made the present invention have broad application prospects and practical effect prospect widely by standardized prospect.
Description of drawings
Fig. 1 transmits and the structural representation of controlling separated network spare;
Fig. 2 is the system configuration schematic diagram of embodiments of the invention;
Fig. 3 is the workflow diagram of the control piece of Fig. 2;
Fig. 4 is the workflow diagram of forwarding element among Fig. 2.
Embodiment
Provide a kind of method that under forwarding and control separated network spare framework, realizes the control piece high availability,, improve systematic function by load balancing simultaneously by the redundancy backup measure of control piece is guaranteed that the network equipment provides continual service.
Transmitting and controlling the method that realizes the control piece high availability under the separated network spare framework.
Comprise the steps:
Determine available control piece tabulation by the control piece administrative unit when 1) initial, and inform forwarding element by the forwarding element administrative unit;
2) each control piece is monitored the variation of other control piece states, all tabulation of available control piece and its loading conditions in the maintaining network spare;
3) the available control piece tabulation of forwarding element by obtaining from the forwarding element administrative unit, all available control piece tabulation, loading condition and load-balancing methods in the networkware that control piece request control piece is safeguarded are one by one initiatively selected control piece then;
4) two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, and many control pieces are not distinguished main or backup, and equity fully each other triggers forwarding element switching controls part according to the loading condition of control piece, realizes load balancing;
5) between many control pieces, carry out data backup, realize the consistency of related data.
Describedly determine available control piece tabulation by the control piece administrative unit when initial, and inform forwarding element: determine available control piece tabulation by the control piece administrative unit when initial by the forwarding element administrative unit, be to use configuration file or order to inform the IP address of available control piece in the control piece administrative unit system in advance by administrative staff, the forwarding element administrative unit is obtained its available control piece tabulation by carrying out exchanges data with the control piece administrative unit afterwards,, when the forwarding element administrative unit is inquired, available control piece tabulation is informed at forwarding element.
Control piece comprises high availability module and ForCES protocol module, and the high availability module is used to monitor all control piece state variation, and according to load condition variation between many control pieces, carrying out load allocating and data backup; The ForCES protocol module then disposes the flow chart of data processing of forwarding element according to the required network type service of user.
Forwarding element mainly comprises high availability module and ForCES protocol module, and the ForCES protocol module is used for according to the control command that control piece issues packet being transmitted processing; The high availability module interactive information of high availability module and control piece according to the suitable control piece of load balancing strategy Dynamic Selection, or connects in lost efficacy back and standby control element of control piece.
The high availability module of control piece comprises: heartbeat module, load allocating module and data backup module, wherein:
Heartbeat module is used for sending and receives heartbeat signal with monitoring control piece state variation;
Load allocating function mould certainly carries out load balancing according to control piece and forwarding element state variation;
The data backup module backups to a plurality of control pieces according to backup policy with the information and the network information on services of forwarding element that control piece is controlled;
Described each control piece is monitored the variation of other control piece states, all tabulation of available control piece and its loading conditions in the maintaining network spare: each control piece is by the monitoring of regular packet exchange realization to other control piece state variation, and wherein the control piece state comprises whether cpu busy percentage, internal memory operating position and the control piece of control piece can normally send heartbeat signal; The controlled product that can normally send heartbeat signal is defined as available control piece, and the relevant information by all available control pieces in the form record networkware, comprises IP address and its state information.
The control piece monitor state changes and the process of carrying out load balancing may further comprise the steps:
A. the high availability module of control piece is determined the control piece tabulation that the networkware system can use according to the control piece state variation, and the control piece state comprises performance, the load condition and whether normal of control piece, and is handed down to the forwarding element node of being controlled.
B. the high availability module of control piece is coordinated the task whether the needs adjustment is born according to the state information of control piece, and the relevant forwarding element of notice.
Above-mentioned steps A also need comprise:
Judge that the networkware state variation is the adding of forwarding element or control piece or withdraws from, if the control piece node adds or withdraws from, then in the control piece node listing, add or delete this node, and upgrade the control piece node listing of each networkware node.If control piece withdraws from the relevant forwarding element of triggering is switched to standby control element, redistribute load afterwards, and upgrade backup policy; Only need to redistribute load if control piece adds, and upgrade backup policy.
If the forwarding element node adds or withdraws from, if add, the Search Control node listing finds the control piece of load minimum, and connects with it, accepts its control, and its control piece also need be selected backup policy; If withdraw from, only need redistribute load, and upgrade backup policy.
The available control piece tabulation of described forwarding element by obtaining from the forwarding element administrative unit, all available control piece tabulations in the networkware that control piece request control piece is safeguarded one by one then, loading condition and load-balancing method, initiatively select control piece: forwarding element is by inquiring about to the forwarding element administrative unit, obtain available control piece tabulation, the available control piece of safeguarding to its control piece of available control piece request is one by one tabulated and state information then, by relatively the cpu busy percentage of all control pieces and the operating position of internal memory select the control piece of load minimum as the own control piece that will be connected.
Initiate forwarding element is initiatively selected the control piece that connects by the control piece node listing, may further comprise the steps:
A, the available control piece node IP tabulation that initialization obtains according to this node, one by one to the up-to-date control piece node state tabulation of each control piece request, if failure is then to next control piece request, up to obtaining current available control piece node status information tabulation;
B, forwarding element are selected suitable control piece, and are connected with it according to the control piece node status information tabulation of receiving.
Described two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, many control pieces are not distinguished main or backup, reciprocity fully each other, loading condition according to control piece triggers forwarding element switching controls part, realize load balancing: two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, many control pieces are not distinguished main or backup, and are reciprocity fully each other; The threshold value that gap between the control piece of the control piece of load maximum and load minimum sets greater than the keeper, to trigger forwarding element switching controls part, first forwarding element that the control piece of load maximum is controlled is controlled by the control piece of load minimum, realizes load balancing.
The control piece node carries out work in ad-hoc mode, controls all forwarding elements, realizes redundancy backup and load balancing, comprising:
Each control piece node is reciprocity mutually in the networkware, but the handling property of each control piece node can be inconsistent, need by quantizing controlled joint behavior index, the load of again the control piece node being born quantizes, and realizes the internodal load balancing of control piece according to above two indexs.
Describedly between many control pieces, carry out data backup, realize the consistency of related data: between many control pieces, carry out data backup, each control piece will backup to other one or more control pieces with forwarding element of being controlled and the required data of operational network service of oneself, by timing data synchronously and bursty data upgrade the consistency of dual mode realization Backup Data.
Need between the control piece node content of this node is carried out multi-computer back-up, will backup to one or more other control piece node at least, may further comprise the steps:
A, according to the control piece node listing, select a plurality of suitable control pieces as backup node, according to the backup policy of setting, the forwarding element information and the network information on services of this node control periodically backuped to a plurality of backup nodes;
B, this node need be accepted the backup request of other control piece nodes, preserve the information and the network information on services of forwarding element that other nodes are controlled;
A plurality of backup node tabulations of C, this node are handed down to the forwarding element of being controlled, and behind this node failure, the forwarding element of being controlled selects suitable control piece node to connect once more according to the loading condition of backup node, accepts its control.
Understand and realization the present invention for the ease of persons skilled in the art, now further specify technical scheme of the present invention in conjunction with the accompanying drawings, provide a kind of embodiment of the method for the invention.Fig. 2 is the principle schematic of embodiments of the invention, having two labels is respectively 1 and 2 control piece and three labels and is respectively 10,11 and 12 forwarding element and forms a ForCES networkware, wherein the control piece administrative unit is present in each control piece by the form of configuration file, and the forwarding element administrative unit is by realizing by the configuration file in the control piece is copied on the forwarding element.Forwarding element has a plurality of network interfaces that are connected with external network, interconnects realization by Ethernet interface via an Ethernet switch between forwarding element and control piece and intercoms mutually.Embodiments of the invention comprise two control pieces, and they back up mutually, and common burden is to the control of forwarding element.High-availability mechanism of the present invention is mainly reflected in high availability (HA) software module, may operate on the different operating system and hardware platform through this module after the less modification.In an embodiment of the present invention, the IXP2400 network processing unit (containing necessary peripheral circuit) that forwarding element uses Intel Company to produce constitutes, and adopts Monta Vista Linux embedded OS; The control piece formation that uses a computer adopts the Redhat (SuSE) Linux OS.
Present embodiment at first uses the C language to write information interaction, data backup and other program related to the present invention according to the method described in the present invention.Compilation tool compiling separately obtains program command code separately through control piece and forwarding element, and packs in separately the memory by loading appliance separately.In order to realize the high availability of control piece, by on control piece and forwarding element with finger daemon mode independent operating HA software module, its major function is the information interaction of this node of management and peer node, make the work that a plurality of nodes can be harmonious, thereby realized the monitoring and the isolation of control piece redundancy backup, fault, and the equilibrium of control piece load.The HA software module can be monitored the running status of local task or thread, and the task or the thread of opposite end are monitored, and finds software fault and reports and handle, and a result who handles is reported to peer node.After the operation of HA software module, press the method for the invention exchange message continually between forwarding element and control piece,, guarantee that control piece has high availability in the networkware at the different strategy of all kinds of unusual selection of control piece and forwarding element.Carry out ad-hoc mode in an embodiment of the present invention, the state of two control pieces does not have active and standby branch, both possess identical software and hardware module, data back up mutually, share jointly the forwarding element control task, thereby be implemented in the performance that improves system under the prerequisite of guaranteeing the control piece high availability.
Mechanism and message interaction process to two mutual redundancy backups of control piece in the present embodiment is described in detail below:
Step a: before networkware starts, at first hold consultation, determine available control piece tabulation, the tabulation of acceptable forwarding element, load balancing strategy, data backup policy and some required parameters of other ForCES agreements operation by control piece administrative unit and forwarding element administrative unit.Control piece administrative unit and forwarding element administrative unit are modules in logic, existing definition in the ForCES agreement, can be used as a software module is present on control piece or the forwarding element, agreement by oneself is carried out mutual information interaction, also can only be a configuration file or is provided with by hand by the networkware keeper.Adopt the configuration file mode to realize its function in the present embodiment, by keeper's manual edit configuration file, on each control piece, preserve the configuration file of available control piece tabulation and the tabulation of acceptable forwarding element, and on forwarding element, only preserve the configuration file of available control piece tabulation.
Step b: control piece related software module mainly comprises ForCES agreement middleware, high availability module two large divisions.Wherein ForCES agreement middleware is responsible for and forwarding element communicates, and the management forwarding element on-the-fly modifies data handling procedure and and user mutual in the forwarding element.The high availability module comprises heartbeat module, load allocating module and data backup module, and heartbeat module is used for sending and receives heartbeat signal with monitoring control piece state variation; Load allocating function mould carries out load allocating according to control piece and forwarding element state variation; The data backup module according to backup policy with the information back-up of forwarding element that control piece is controlled to a plurality of control pieces.According to the manual load balancing strategy of setting, control piece 1 is responsible for control forwarding element 10 and route service two class loads in the present embodiment, and control piece 2 is responsible for control forwarding element 11 and 12, realizes load balancing thus.The high availability module need be responsible for the load data of oneself is backed up, the forwarding element of being responsible for as control piece 1 standby control element 2 11 with 12 relevant data, and the forwarding element of being responsible for as control piece 2 standby control element 1 10 and route service relevant data.The workflow of control piece as shown in Figure 3.
Step c: the related software module of forwarding element mainly comprises ForCES agreement middleware, high availability module two large divisions.Wherein ForCES agreement middleware is responsible for communicating with control piece, accepts its control, and gives control piece with the data forwarding that oneself can not handle, such as routing protocol data.The high availability module comprises heartbeat module, control piece selection module.Heartbeat module is mainly used in and receives the state variation of heartbeat signal with the monitoring control piece; Control piece selects module then according to the control piece list information, comprise the performance of control piece, being born of task and its load balancing strategy, select suitable control piece, and this control piece IP address passed to ForCES agreement middleware, connect by itself and control piece, report LFB example and ability, receive the configuration of control piece, realize user's data designated forwarding capability LFB.The workflow of forwarding element as shown in Figure 4.
Steps d: content and the opportunity of selecting data backup control piece and data backup mainly finished in data backup.Selection data backup control piece can be by manual and automatic dual mode, manual mode can be by the priority list of keeper according to an available data standby control element of control piece appointment, and automated manner then can be selected suitable standby control element tabulation according to the loading condition of all the other available control pieces.Secondly need to consider the number of backup, can preestablish by the keeper.In the data backup process, need to detect the state of standby control element, if unusual, then from the standby control element tabulation, select next control piece to back up.Consider on the content of data backup and opportunity: at first data content is divided into synchrodata and bursty data from following two aspects, treat with a certain discrimination the opportunity of different pieces of information backup then, back up after when synchrodata can select network traffics idle, bursty data must be backed up in realtime.
Step e: the load balancing strategy is mainly finished the quantification to all load treating capacities in the networkware, and its load comprises the support two big classes of the control and the network service of forwarding element, distributes according to the performance of each control piece afterwards.Its strategy is divided into static and dynamic dual mode, and static mode quantizes to be estimated in advance by the keeper to load, distributes according to the performance and the number of control piece then; Dynamical fashion is to be consulted mutually by the HA module of control piece in the networkware running, in real time load is quantized, in case thereby load imbalance just takes load to redistribute the realization load balancing.Adopt static mode to realize load balancing in the native system, control piece 1 is responsible for control forwarding element 10 and route service two class loads, and control piece 2 is responsible for control forwarding element 11 and 12.
By above several steps, whole networkware begins operate as normal, comes first step explanation the present invention in the mechanism that control piece is realized high availability at the handling process of four class abnormal conditions below:
1, certain forwarding element adds networkware:
When new forwarding element need add this networkware,, obtain available control piece tabulation at first by communicating by letter with the forwarding element administrative unit.By the HA module of forwarding element HA,, comprise the loading condition and the load balancing strategy of each control piece then from its details of obtaining current available control piece tabulation according to control piece tabulation access control part.If specified the load balancing strategy, forwarding element then and according to the control piece that strategy determined connects, otherwise the loading condition of each control piece is sorted, and selects the control piece of present load minimum to connect.
2, certain forwarding element part that exits network
When certain forwarding element withdraws from this networkware, may be for maintenance upgrade, also may be hardware and software failure.The control piece ForCES protocol module that the withdrawing from of forwarding element at first passes through to be controlled finds, and informs the HA module, by HA module broadcasting load variations, and removes data on the relevant standby control element.Consult to carry out load balancing afterwards between control piece mutually.
3, certain control piece adds networkware
When new control piece adds this networkware, have two big classes may: a kind of is for control piece is carried out adding behind the upgrade maintenance, also can be after the control piece operation makes mistakes, and misarrangement recovers the back and adds.Another kind of is to add new control piece to realize redundancy backup, and in order to obtain better system performance.In order to allow control piece add this networkware, at first need to dispose the control piece administrative unit, revise the tabulation of available control piece, send to each control piece, control piece HA module is received and is changed refresh control part list information after the message, and sends to forwarding element HA module.Each control piece HA module is according to the strategy of load balancing plan road decision carrying out load transfer, and adjustment data backup node.
4, certain control piece part that exits network
Control piece withdraws from this networkware, have two kinds may: a kind of is for control piece is carried out upgrade maintenance, and another kind is to make mistakes in the control piece operation.For first kind, the keeper can revise available control piece tabulation by the control piece administrative unit, with this control piece deletion, gives all control piece and forwarding elements with this information broadcast.The forwarding element of being responsible for by this control piece will be transferred to relevant standby control element with related service, for the transfer of forwarding element and related service, by the HA module of this forwarding element of control piece deletion message trigger press fixed backup recovery policy carry out.For second kind of situation, makeing mistakes of control piece monitors discovery by control piece administrative unit, other control pieces and forwarding element by heartbeat, loses efficacy in case find certain control piece, then triggers corresponding backup recovery policy.
Though described the present invention by embodiment, those of ordinary skills know, the present invention has many distortion and variation and do not break away from spirit of the present invention, wish that appended claim comprises these distortion and variation and do not break away from spirit of the present invention.
Claims (6)
1, a kind of method that realizes the control piece high availability under forwarding and control separated network spare framework is characterized in that, comprises the steps:
Determine available control piece tabulation by the control piece administrative unit when 1) initial, and inform forwarding element by the forwarding element administrative unit;
2) each control piece is monitored the variation of other control piece states, all tabulation of available control piece and its loading conditions in the maintaining network spare;
3) the available control piece tabulation of forwarding element by obtaining from the forwarding element administrative unit, all available control piece tabulation, loading condition and load-balancing methods in the networkware that control piece request control piece is safeguarded are one by one initiatively selected control piece then;
4) two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, and many control pieces are not distinguished main or backup, and equity fully each other triggers forwarding element switching controls part according to the loading condition of control piece, realizes load balancing;
5) between many control pieces, carry out data backup, realize the consistency of related data.
2, according to claim 1 a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework, it is characterized in that describedly determining available control piece tabulation by the control piece administrative unit when initial, and inform forwarding element: determine available control piece tabulation by the control piece administrative unit when initial by the forwarding element administrative unit, be to use configuration file or order to inform the IP address of available control piece in the control piece administrative unit system in advance by administrative staff, the forwarding element administrative unit is obtained its available control piece tabulation by carrying out exchanges data with the control piece administrative unit afterwards,, when the forwarding element administrative unit is inquired, available control piece tabulation is informed at forwarding element.
3, according to claim 1 a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework, it is characterized in that described each control piece monitors the variation of other control piece states, all tabulation of available control piece and its loading conditions in the maintaining network spare: each control piece is by the monitoring of regular packet exchange realization to other control piece state variation, and wherein the control piece state comprises whether cpu busy percentage, internal memory operating position and the control piece of control piece can normally send heartbeat signal; The controlled product that can normally send heartbeat signal is defined as available control piece, and the relevant information by all available control pieces in the form record networkware, comprises IP address and its state information.
4, according to claim 1 a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework, it is characterized in that, the available control piece tabulation of described forwarding element by obtaining from the forwarding element administrative unit, all available control piece tabulations in the networkware that control piece request control piece is safeguarded one by one then, loading condition and load-balancing method, initiatively select control piece: forwarding element is by inquiring about to the forwarding element administrative unit, obtain available control piece tabulation, the available control piece of safeguarding to its control piece of available control piece request is one by one tabulated and state information then, by relatively the cpu busy percentage of all control pieces and the operating position of internal memory select the control piece of load minimum as the own control piece that will be connected.
5, according to claim 1 a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework, it is characterized in that described two above control pieces bear the network service function towards a plurality of forwarding elements simultaneously, many control pieces are not distinguished main or backup, reciprocity fully each other, loading condition according to control piece triggers forwarding element switching controls part, realize load balancing: two above control pieces are born the network service function towards a plurality of forwarding elements simultaneously, many control pieces are not distinguished main or backup, and are reciprocity fully each other; The threshold value that gap between the control piece of the control piece of load maximum and load minimum sets greater than the keeper, to trigger forwarding element switching controls part, first forwarding element that the control piece of load maximum is controlled is controlled by the control piece of load minimum, realizes load balancing.
6, according to claim 1 a kind of in the method for transmitting and control realization control piece high availability under the separated network spare framework, it is characterized in that the described data backup of between many control pieces, carrying out, realize the consistency of related data: between many control pieces, carry out data backup, each control piece will backup to other one or more control pieces with forwarding element of being controlled and the required data of operational network service of oneself, by timing data synchronously and bursty data upgrade the consistency of dual mode realization Backup Data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100597962A CN101237413B (en) | 2008-02-29 | 2008-02-29 | Method for realizing high-availability of control part under forward and control separated network architecture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100597962A CN101237413B (en) | 2008-02-29 | 2008-02-29 | Method for realizing high-availability of control part under forward and control separated network architecture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101237413A true CN101237413A (en) | 2008-08-06 |
| CN101237413B CN101237413B (en) | 2010-06-09 |
Family
ID=39920778
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100597962A Expired - Fee Related CN101237413B (en) | 2008-02-29 | 2008-02-29 | Method for realizing high-availability of control part under forward and control separated network architecture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101237413B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101834871A (en) * | 2010-05-14 | 2010-09-15 | 浙江工商大学 | Open type router device and method thereof |
| CN101883049A (en) * | 2010-06-29 | 2010-11-10 | 浙江工商大学 | Method for realizing business configuration router based on architecture of transmitting and controlling separate networkware |
| CN102546823A (en) * | 2012-02-18 | 2012-07-04 | 南京云创存储科技有限公司 | File storage management system of cloud storage system |
| CN103401695A (en) * | 2013-06-21 | 2013-11-20 | 浙江工商大学 | Failure detection and recovery method for control elements based on forwarding and control network element framework |
| CN104331353A (en) * | 2014-10-17 | 2015-02-04 | 云南远信数通科技有限公司 | Method for guaranteeing software high availability |
| CN104951552A (en) * | 2015-06-29 | 2015-09-30 | 努比亚技术有限公司 | Big data statistical method and system used for big data statistics |
| CN106789246A (en) * | 2016-12-22 | 2017-05-31 | 广西防城港核电有限公司 | The changing method and device of a kind of active/standby server |
| CN110493328A (en) * | 2019-08-07 | 2019-11-22 | 苏州浪潮智能科技有限公司 | A kind of state information processing method, device, server and readable storage medium storing program for executing |
| CN114546714A (en) * | 2020-11-27 | 2022-05-27 | 中国移动通信集团四川有限公司 | Data backup method, system, device and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1943184B (en) * | 2004-02-02 | 2011-08-03 | 艾利森电话股份有限公司 | Adaptive router architecture enabling efficient internal communication |
| CN100553257C (en) * | 2006-02-24 | 2009-10-21 | 浙江工商大学 | Transmit method with the exchange of control separated network spare internal information |
-
2008
- 2008-02-29 CN CN2008100597962A patent/CN101237413B/en not_active Expired - Fee Related
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101834871B (en) * | 2010-05-14 | 2013-02-27 | 浙江工商大学 | Open type router device and method thereof |
| CN101834871A (en) * | 2010-05-14 | 2010-09-15 | 浙江工商大学 | Open type router device and method thereof |
| CN101883049A (en) * | 2010-06-29 | 2010-11-10 | 浙江工商大学 | Method for realizing business configuration router based on architecture of transmitting and controlling separate networkware |
| CN101883049B (en) * | 2010-06-29 | 2011-11-30 | 浙江工商大学 | Method for realizing business configuration router based on architecture of transmitting and controlling separate networkware |
| CN102546823A (en) * | 2012-02-18 | 2012-07-04 | 南京云创存储科技有限公司 | File storage management system of cloud storage system |
| CN103401695B (en) * | 2013-06-21 | 2017-02-22 | 浙江工商大学 | Failure detection and recovery method for control elements based on forwarding and control network element framework |
| CN103401695A (en) * | 2013-06-21 | 2013-11-20 | 浙江工商大学 | Failure detection and recovery method for control elements based on forwarding and control network element framework |
| CN104331353A (en) * | 2014-10-17 | 2015-02-04 | 云南远信数通科技有限公司 | Method for guaranteeing software high availability |
| CN104951552A (en) * | 2015-06-29 | 2015-09-30 | 努比亚技术有限公司 | Big data statistical method and system used for big data statistics |
| CN106789246A (en) * | 2016-12-22 | 2017-05-31 | 广西防城港核电有限公司 | The changing method and device of a kind of active/standby server |
| CN110493328A (en) * | 2019-08-07 | 2019-11-22 | 苏州浪潮智能科技有限公司 | A kind of state information processing method, device, server and readable storage medium storing program for executing |
| CN114546714A (en) * | 2020-11-27 | 2022-05-27 | 中国移动通信集团四川有限公司 | Data backup method, system, device and storage medium |
| CN114546714B (en) * | 2020-11-27 | 2023-07-21 | 中国移动通信集团四川有限公司 | Data backup method, system, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101237413B (en) | 2010-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101237413B (en) | Method for realizing high-availability of control part under forward and control separated network architecture | |
| KR101099822B1 (en) | Redundant routing capabilities for a network node cluster | |
| CN107209710B (en) | Node system, server device, scaling control method, and program | |
| US7225356B2 (en) | System for managing operational failure occurrences in processing devices | |
| CN107465721B (en) | Global load balancing method and system based on double-active architecture and scheduling server | |
| US7577720B2 (en) | Migration method for software application in a multi-computing architecture, method for carrying out functional continuity implementing said migration method and multi-computing system provided therewith | |
| CN100470494C (en) | Cluster availability management method and system | |
| US6332198B1 (en) | Network device for supporting multiple redundancy schemes | |
| US7389367B2 (en) | Method of managing I/O interface modules in a computer system | |
| CN101483555A (en) | Method for redundant backup implemented by dual ForCES control piece | |
| US20030037150A1 (en) | System and method for quality of service based server cluster power management | |
| CN101706781B (en) | Method and system for centralized management of database caches | |
| WO2006065661A2 (en) | Systems and methods providing high availability for distributed systems | |
| CN110224871A (en) | A kind of high availability method and device of Redis cluster | |
| US20080301489A1 (en) | Multi-agent hot-standby system and failover method for the same | |
| US20080307254A1 (en) | Information-processing equipment and system therefor | |
| US7319664B2 (en) | Redundant link management switch for use in a stack of switches and method thereof | |
| CN102916825A (en) | Management equipment of dual-computer hot standby system, management method and dual-computer hot standby system | |
| CN112948063B (en) | Cloud platform creation method and device, cloud platform and cloud platform implementation system | |
| CN103532753A (en) | Double-computer hot standby method based on memory page replacement synchronization | |
| CN111371625A (en) | Method for realizing dual-computer hot standby | |
| CN111800484B (en) | Service anti-destruction replacing method for mobile edge information service system | |
| WO2006005251A1 (en) | Method and system for carrying out the function of switching in a communication system | |
| CN100496030C (en) | Integrated access media gateway device based on the high-available system | |
| WO1997049034A1 (en) | Job taking-over 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100609 Termination date: 20210228 |