CN102056207A

CN102056207A - Method and system for realizing disaster recovery and switching

Info

Publication number: CN102056207A
Application number: CN2009102367387A
Authority: CN
Inventors: 章恩华
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2009-10-29
Filing date: 2009-10-29
Publication date: 2011-05-11
Anticipated expiration: 2029-10-29
Also published as: CN102056207B; WO2010145428A1

Abstract

The invention discloses a method and a system for realizing disaster recovery and switching, wherein the method comprises the following steps: judging whether synchronous switching conditions are met or not according to service quality test result of a data access link between the front end or the back end after the disaster recovery and the switching and the back end or the front end without the disaster recovery and the switching of a network device for the network device performing the disaster recovery and the switching at one of the front end and the back end; and performing synchronous switching on the back end or the front end without the disaster recovery and the switching of the network device according to synchronous switching rules when the judgment result is that the synchronous switching conditions are met. With the help of the invention, not only the unnecessary switching between the FE (front end) and the BE (back end) can be avoided, but also the service performance of the system can be ensured through automatic synchronous switching between the FE and the BE, and the service interruption time caused by manual switching can be further avoided.

Description

Implementation method that disaster tolerance is switched and system

Technical field

The present invention relates to the communications field, relate in particular to implementation method and system that a kind of disaster tolerance is switched.

Background technology

Attaching position register (home location register, abbreviate HLR as) be used to store mobile communication the whole network contracted user's data, along with the high speed development of mobile communication business and to the increase day by day of fail safe and integrated services demand, the active and standby and active and standby Disaster Tolerant Scheme of N+1 is evolved to distributed HLR from simple 1+1 to impel HLR.

At present, distributed HLR has following disaster tolerance: (1) is responsible for front end (the Front End of business logic processing, abbreviate FE as) and rear end (the Back End of responsible storage of subscriber data, abbreviate BE as) separated from one another, FE and BE both can be deployed in same geographic sites, also can be deployed in different geographic sites; (2) FE and BE can dispose Disaster Tolerant Scheme separately respectively, realize symmetry or asymmetric deployment, and under the situation of asymmetric deployment, the quantity of FE and BE are unequal, can realize the centralized stores of data.

Usually, in traditional active and standby HLR Disaster Tolerant Scheme, no matter be that the master uses website, or the disaster tolerance website, all need to dispose FE and BE, and FE can only visit the BE of this website.And in distributed HLR, FE can visit all BE, thereby can effectively avoid switching the restriction that back FE needs and the BE binding is brought.Will make distributed HLR and traditional HLR Disaster Tolerant Scheme switch difference below the existence in system like this: traditional HLR Disaster Tolerant Scheme provides a website switching to the another one website, particularly, leading with website in the process that the disaster tolerance website is switched, comprised that Business Processing is switched to disaster tolerance FE from main with FE, and data access is switched to disaster tolerance BE from main with BE; And for distributed HLR, FE and BE can switch respectively, that is, the master is broken down with FE, only Business Processing need be switched to standby FE, does not need BE to carry out synchronous switching.Same, when BE need switch, do not require that FE carries out synchronous switching.

Fig. 1 shows the deployment architecture of the active and standby disaster tolerance system of traditional HLR 1+1.As shown in Figure 1, main data with HLR 401 are synchronized to redundancy HLR 402 by the copy mode based on storage, database or application, simultaneously, at Signalling Transfer Point (signaling transfer point, abbreviate STP as) 101,102 main alternate routing 601 with HLR route 801 is set, and point to redundancy HLR 402.Before being switched to redundancy HLR 402, STP 101,102 sends to all service request main with HLR 401, and is with among the HLR 401, main with the main processing of finishing service request with BE 301 of FE 201 visits main.Main with FE 201 or main with after BE 301 faults, when being switched to redundancy HLR 402, stop the master and duplicate with the data of BE 301 and disaster tolerance BE302, and blocking-up STP 101,102 and main signaling link with FE 201, all service request send to standby HLR 402 automatically.

Fig. 2 shows the deployment architecture of distributed HLR.As shown in Figure 2, the master all can be deployed in different geographical position with FE 201, standby FE202, master with BE 301, standby BE 302.When the master used FE 201 faults, business will be switched to standby FE 202, and standby FE 202 can visit the main processing of finishing service request with BE 301.

Because FE allows to be deployed on the different geographic sites with BE, therefore, the service quality between FE and the BE just becomes the factor that restriction FE visits the BE data performance.Owing to exist geographical far and near difference etc. to influence the factor of data performance between different FE and the BE, can cause FE to visit some BE and can obtain optimal data performance.For example, in system deployment shown in Figure 2, main may be best with FE at the main data performance during with BE of visit, and the data performance when visiting standby BE is relatively poor.If only FE switches, BE does not carry out synchronous switching, can make that the service quality of switching between back BE and the FE is very poor, and data performance reduces, and when data performance can not satisfy business demand, can carry out by traffic affecting; In like manner, under the situation that BE switches separately, equally this problem can appear.

In like manner, except the HLR of distributed deployment, other can carry out distributed deployment, and the network equipment that business logic processing and storage are carried out independently of one another (for example, comprise authentication, authorize, (Authentication charges, Authorization, Accounting abbreviates AAA as) server, home subscriber server (Home Subscriber Server, abbreviate HSS as), equipment identity register (EquipmentIdentity Register, abbreviate EIR as)), all do not exist owing to FE and BE do not carry out synchronous switching and cause the affected problem of quality of service.

At not being guaranteed even can reducing data performance and then the professional problem of normally carrying out of influence owing to BE and FE carry out the asynchronous data performance of switching between BE after causing switching and the FE in the network equipment of distributed deployment in the correlation technique, effective solution is not proposed as yet at present.

Summary of the invention

At in the network equipment of distributed deployment in the correlation technique owing to FE and BE carry out asynchronously switching between FE after causing switching and the BE that data access performance can not be guaranteed even can be because the reduction of data access performance influence professional problem of normally carrying out, the present invention proposes the implementation that a kind of disaster tolerance is switched, and can guarantee to switch the data performance between back FE and the BE.

Technical scheme of the present invention is achieved in that

According to an aspect of the present invention, the implementation method that provides a kind of disaster tolerance to switch, the network equipment that is used for that distributed deployment and business logic processing are separated with storage is realized switching.

The method according to this invention comprises: for the side in the front-end and back-end network equipment that disaster tolerance is switched takes place, rear end that disaster tolerance switches does not take place in front end after disaster tolerance is switched or rear end and the network equipment or the service quality testing result of the data access link between the front end judges whether to satisfy the synchronous switching condition according to carrying out; In judged result is to satisfy under the situation of synchronous switching condition, according to the synchronous switching rule rear end or front end that disaster tolerance switches is not taken place the network equipment and carries out synchronous switching.

After the front end of the network equipment or rear end disaster tolerance takes place were switched, this method can further comprise: the service quality that takes place to detect the data access link in front end that disaster tolerance switches or rear end obtains that rear end that disaster tolerance switches does not take place for service quality testing result and/or the network equipment or the service quality of front-end detection data access link obtains the service quality testing result.

Preferably, the service quality testing result of data access link comprise following one of at least: the request response time in the request sum in the predetermined amount of time between the two ends of data access link, the predetermined amount of time between the two ends of data access link is greater than the average request response time between the two ends of data access link in the request quantity of response time threshold value, the predetermined amount of time.

And, one of at least under the situation, the judgment data access link satisfies the synchronous switching condition: the request response time in the predetermined amount of time between the two ends of data access link surpasses default amount threshold greater than the request quantity of response time threshold value below the service quality testing result satisfies, the average request response time in the predetermined amount of time between the two ends of data access link is greater than the average response time threshold value, request response time in the predetermined amount of time between the two ends of data access link surpasses default proportion threshold value greater than request quantity shared ratio in total request amount of the first default response time threshold value.

Preferably, the synchronous switching rule can comprise the rear end of the data access link that the foundation of front end after switching with disaster tolerance or rear end is new or the priority of front end.

Preferably, the network equipment can be for one of following: authentication, authorize accounting server; Attaching position register; Equipment identity register; Home subscriber server.

According to a further aspect in the invention, the realization system that provides a kind of disaster tolerance to switch, the network equipment that is used for that distributed deployment and business logic processing are separated with storage is realized switching, and this system can comprise the performance monitoring module, switches decision-making module and switch control module.

Particularly, the performance monitoring module is arranged at the front end of each network equipment and/or is arranged at the rear end of each network equipment, be used to detect the front end at this performance monitoring module place or rear end and, obtain the service quality testing result with front end or the rear end of back end communications or the service quality between the front end at this place; Switch decision-making module and be used to receive service quality testing result, and judge front end or the rear end after network equipment generation disaster tolerance is switched and disaster tolerance does not take place according to the service quality testing result and switch the service quality of the data access link between rear end or the front end and whether satisfy the synchronous switching condition from the performance monitoring module; Switching that control module is used in the judged result of switching decision-making module is to satisfy under the situation of synchronous switching condition, according to the synchronous switching rule rear end or front end that disaster tolerance switches is not taken place the network equipment and carries out synchronous switching.

Wherein, the service quality testing result of data access link can comprise following one of at least: the request response time in the request sum in the predetermined amount of time between the two ends of data access link, the predetermined amount of time between the two ends of data access link is greater than the average request response time between the two ends of data access link in the request quantity of response time threshold value, the predetermined amount of time.

Preferably, one of at least under the situation, switch decision-making module and can the judgment data access link satisfy the synchronous switching condition: the request response time in the predetermined amount of time between the two ends of data access link surpasses default amount threshold greater than the request quantity of response time threshold value below the service quality testing result satisfies, the average request response time in the predetermined amount of time between the two ends of data access link is greater than the average response time threshold value, request response time in the predetermined amount of time between the two ends of data access link surpasses default proportion threshold value greater than request quantity shared ratio in total request amount of the first default response time threshold value.

By means of technique scheme of the present invention, by the service quality between FE and the BE (data performance) is monitored, coordinate the synchronous switching of FE and BE, not only can avoid unnecessary FE and BE to switch, can also guarantee the system business performance by the automatic synchronous switching of FE and BE, and auto switching has also been avoided the manual service outage duration that causes of switching.

Description of drawings

Fig. 1 is the deployment architecture according to the active and standby disaster tolerance system of HLR 1+1 of correlation technique;

Fig. 2 is the deployment architecture according to the distributed HLR of correlation technique;

Fig. 3 is the flow chart of the implementation method of switching according to the disaster tolerance of the inventive method embodiment;

Fig. 4 is the flow chart of the disaster tolerance of the system according to the invention embodiment realization system of switching.

Embodiment

At distributed deployment in the correlation technique, and the network equipment that business logic processing is separated with storage (for example, HLR, aaa server, HSS, EIR etc.) go up that a side among FE and the BE switches that back the opposing party does not carry out synchronous switching and FE after causing switching and the data access performance between the BE can't guaranteed problems, though the present invention considers FE and the BE in the distributed HLR system and can independently switch, but in some cases, after a side among FE and the BE switches, if carrying out synchronous switching, the opposing party can make system obtain better data access performance, therefore, give chapter and verse service quality between each FE and each BE of the present invention determines whether the opposing party is carried out synchronous switching after the switching in FE or BE, if after having carried out switching, data access performance between FE and the BE is relatively poor, then a side who does not switch among FE and the BE is carried out synchronous switching.

Below in conjunction with accompanying drawing, describe embodiments of the invention in detail.

Method embodiment

In the present embodiment, the implementation method that provides a kind of disaster tolerance to switch, the network equipment (for example, can comprise equipment such as HLR, HSS, aaa server, EIR) that is used for that distributed deployment and business logic processing are separated with storage is realized switching.

In realizing process of the present invention, at first need according to the configuration of the communication conditions in the network by switching rule.Particularly, for FE, can determine this FE can be preferentially and which BE set up the data access link so that this FE is called mainly when use FE after disaster tolerance is switched, be the BE that this FE selects optimum; Similarly, for BE, can determine this BE can be preferentially and which FE set up the data access link so that this BE is called mainly when use BE after disaster tolerance is switched, be the FE that this BE selects optimum; Afterwards, alternatively, can will should rule preserve by form, for example, as shown in table 1.

Table 1

The list item numbering	The visit main body	Performance ordering when the visit main body conducts interviews
			1001	FE201	BE301 BE302
1002	FE202	BE302 BE301
			1003	BE301	FE201 FE202
1004	BE302	FE202 FE201

This is switched, and rule is pre-configured can the network planning time, and its configuration result is relevant with the factors such as geographical position relation of each FE, BE, and this paper is not described in detail.

In addition, also needing each front end and/or backend arrangement with all-network equipment is to monitor the service quality of the data access link relevant with this front end and/or rear end.

For example, in network design structure shown in Figure 2, main with FE 201 can detect its with main with the service quality between the BE301, can detect itself and master and use service quality between the BE 301; Same, mainly also can detect it and lead the service quality of using between the FE 201 with BE 301, standby BE 302 also can detect the service quality between itself and the standby FE 202, and, can carry out two-way detection between FE and the BE when being necessary.

In the actual service quality of carrying out when detecting, can adopt a variety of methods, fairly simple and effective and efficient manner is exactly the mode of request, promptly, send request and wait until the response that the other side returns to the other side among FE and/or the BE, according to the time detecting service quality of response.

At statistics network as a result the time, for each FE and BE, can write down request sum, average response time in the certain hour section therebetween, surpass the information such as number of responses of default response time threshold value.

Below in conjunction with accompanying drawing, describe processing procedure of the present invention in detail.

As shown in Figure 3, the implementation method of switching according to the disaster tolerance of present embodiment comprises: step S302, the network equipment that disaster tolerance is switched takes place for the side among FE and the BE, according to carrying out FE after disaster tolerance is switched or BE and the network equipment BE that disaster tolerance switches or the data access link between the FE do not take place (promptly, after disaster tolerance is switched, newly-established data access link) service quality testing result judges whether to satisfy the synchronous switching condition, the synchronous switching condition can have multiple situation, for example, can think and satisfy the synchronous switching condition when service quality on the data access link can not satisfy business demand; Step S304 is to satisfy under the situation of synchronous switching condition in judged result, according to synchronous switching rule (this rule can obtain by pre-configured) BE or FE that disaster tolerance switches is not taken place the network equipment and carries out synchronous switching.

In step S302, after the FE of the network equipment or BE disaster tolerance takes place switch, FE that disaster tolerance switches or BE take place need detect the service quality of data access link and obtain the service quality testing result and/or the network equipment service quality that BE that disaster tolerance switches or FE detect the data access link does not take place obtains the service quality testing result.

Preferably, the service quality testing result of data access link can comprise following one of at least: the request response time in the request sum in the predetermined amount of time between the two ends of data access link, the predetermined amount of time between the two ends of data access link is greater than the average request response time between the two ends of data access link in the request quantity of response time threshold value, the predetermined amount of time.

When satisfying the judgement of business demand whether, if at least one in meeting the following conditions then thought and can not be satisfied business demand: the quantity that surpasses the response of predetermined response to time threshold surpasses certain numerical value, surpass that the ratio that the quantity of the response of predetermined response to time threshold occupies in overall response quantity surpasses the certain proportion threshold value, average response time surpasses default average response time threshold value.

Because different business is to the requirement difference of service quality, so above-mentioned condition can be used singly or in combination.For example, for the undemanding business of quality of service requirement, can only judge with a condition in the above-mentioned condition.

For example, suppose in network design structure shown in Figure 2, if main switched to standby FE 202, just need the standby FE 202 after judgement is switched this moment and whether the master that switches can satisfy business demand with the service quality of the data access link between the BE 301 with FE 201; If judged result is for being (that is, not satisfying the synchronous switching condition), then do not carry out synchronous switching with BE 301, otherwise need switch to standby BE 302 with BE 301 main to main, cooperate with standby BE 302 by standby FE 202 and carry out the business of deal with data.

Switch at first, judge afterwards for BE whether FE carries out the situation of synchronous switching, and processing mode is similar, no longer describe in detail here.

When selecting the object of synchronous switching, as shown in table 1, each list item has been specified the auto switching strategy of FE and BE, for example, in list item 1001, the priority of FE 201 visit BE 301 is higher than BE302, therefore, (can not satisfy business demand if the performance of FE 201 visit BE 302 is lower than the setting threshold values, promptly, satisfy the synchronous switching condition) time, switching that control module can switch BE 301 automatically is main usefulness, does not then allow on the contrary; Configuration item 1003, the performance of FE 201 visit BE 301 are higher than FE 202 visit BE301, and therefore, if the performance of FE 202 visit BE 301 is lower than when setting threshold values, switching that control module can switch FE 201 automatically is main usefulness, does not then allow on the contrary.

By above-mentioned processing, can automatic decision switch with the synchronous disaster tolerance of control FE and BE, the system that makes runs under the optimum performance all the time, particularly for the asymmetric deployment of FE and BE, can avoid because the manual service outage duration that is produced of switching.

System embodiment

In the present embodiment, the realization system that provides a kind of disaster tolerance to switch, the network equipment (for example, comprising HLR, HSS, EIR, aaa server etc.) that is used for that distributed deployment and business logic processing are separated with storage is realized switching.

The realization system of switching according to the disaster tolerance of present embodiment comprises the performance monitoring module, switches decision-making module and switches control module, particularly:

The performance monitoring module is arranged at the FE of each network equipment and/or is arranged at the BE of each network equipment, be used to detect BE that the FE at this performance monitoring module place or BE communicate by letter with FE or BE with this place or the service quality between the FE, obtain the service quality testing result; Switch decision-making module and be used to receive service quality testing result, and judge FE or the BE after network equipment generation disaster tolerance is switched and disaster tolerance does not take place according to the service quality testing result and switch the service quality of the data access link between BE or the FE and whether satisfy pre-configured synchronous switching condition from the performance monitoring module; Switching that control module is used in the judged result of switching decision-making module is to satisfy under the situation of synchronous switching condition, according to pre-configured synchronous switching rule BE or FE that disaster tolerance switches is not taken place the network equipment and carries out synchronous switching.

To be deployed as example with distributed HLR below and describe the realization system of switching in detail according to the disaster tolerance of the embodiment of the invention.

Fig. 4 is the structured flowchart of the realization system of switching according to the disaster tolerance of present embodiment.

As shown in Figure 4, the service processing equipment of distributed HLR is made of with FE 201 and standby FE 202 main, is deployed in different geographical position; The data storage device of distributed HLR is made of with BE 301 and standby BE 302 main, is deployed in different geographical position, and the master can use FE 201, standby FE 202 not in same geographical position with the master with BE 301 and standby BE 302.

In the present embodiment, suppose in FE deploy performance monitoring module, promptly, as shown in Figure 4, performance monitoring module 101 is deployed in main with on the FE 201, performance monitoring module 102 is deployed on the standby FE 202, and performance monitoring module 101 is used to monitor the main response time with FE 201 visit BE 301, and performance monitoring module 102 is used to monitor the response time of standby FE 202 visit BE 302.

Switch decision-making module 401 and switch control module 501 and can be deployed on the identical node, also can be deployed in respectively on the different nodes, preferably, can be deployed in same geographical position switching decision-making module 401 and switching control module 501.

Below with main be example with FE 201 faults, specify disaster tolerance according to the present invention switch in the realization system FE break down switch after, according to the performance coordination FE of FE visit BE and the disaster tolerance reversed process of BE.

Main with after FE 201 faults, the signaling link of STP and the main FE 201 of using interrupts, and according to the route stand-by setting, STP sends to standby FE 202 with all service request.Because the master does not break down with BE 301, therefore, the main processing of finishing service request with the user data of BE 301 of FE 202 visits.

Performance monitoring module 102 can be monitored the main response time with BE 301 of FE 202 visit in real time, and sends in the mode that regularly reports and to switch decision-making module 401.Performance monitoring module 102 can pre-configured exception response time threshold values T1, makes that reporting the information of switching decision-making module 401 can comprise that total request quantity, average response time and response time in this time period is greater than the request quantity of exception response time threshold values T1.

Switch decision-making module 401 and collect the performance data that performance monitoring module 102 reports, and gather and analyze the performance data of all properties monitoring module of FE 202 in the pre-configured time period T2.Switching decision-making module 401 uses pre-configured performance index threshold values to determine whether to need to trigger the BE auto switching.The performance index threshold values can comprise average response time threshold values T3 and exception response request ratio threshold values T4, if in gathering time period T2, average response time surpasses threshold values T3, perhaps the response time accounts for the ratio of total request greater than exception response request ratio threshold values T4 greater than the request of exception response time threshold values T1, switch decision-making module 401 and determine to need to trigger the BE auto switching, and notice is switched control module 501 and begun to switch.

Switch after control module 501 receives the indication of switching decision-making module 401, need judge whether to allow the BE auto switching, and select a target BE to use BE as new master.If distributed HLR allows auto switching, and standby BE 302 running statuses are normal, then switching control module 501 at first controls and main switches to standby with FE 301, controlling then that standby FE 302 switches is main using, data copy to BE 301 from BE 302, and FE 202 visit BE 302 finish the processing of service request.

Below will with main be example with BE 301 faults, specify according in the disaster tolerance exchanging system of the present invention BE break down switch after, switch step according to the performance coordination FE of FE visit BE and the disaster tolerance of BE.

Main with after BE 301 faults, needing standby BE 302 to switch is main usefulness, and that STP still sends to all service request is main with FE 201, and the main user data that can only visit BE 302 with FE 201 is finished the processing of service request.

Main response time of performance monitoring module 101 monitoring,, send to and switch decision-making module 401 in the mode that regularly reports with FE 201 visit BE 302.Performance monitoring module 101 can pre-configured exception response time threshold values T1, makes that reporting the information of switching decision-making module 401 can comprise that total request quantity, average response time and response time in this time period is greater than the request quantity of exception response time threshold values T1.

Switch decision-making module 401 and collect the performance data that performance monitoring module 101 reports, and gather and analyze the performance data of all properties monitoring module of the main FE 201 of using in the pre-configured time period T2.Switching decision-making module 401 uses pre-configured performance index threshold values to determine whether to need to trigger the FE auto switching.The performance index threshold values can comprise average response time threshold values T3 and exception response request ratio threshold values T4, if in gathering time period T2, average response time surpasses threshold values T3, perhaps the response time accounts for the ratio of total request greater than exception response request ratio threshold values T4 greater than the request of exception response time threshold values T1, switch decision-making module 401 and determine to need to trigger the FE auto switching, and notice is switched control module 501 and begun to switch.

Switch after control module 501 receives the indication of switching decision-making module 401, need judge whether to be configured to allow the FE auto switching, and select a target FE to use FE as new master.Among Fig. 2, if distributed HLR is configured to allow auto switching, and standby FE 202 running statuses normal (and the service quality between FE 202 and the BE 302 can satisfy business demand), then switching control module 501 at first controls FE 202 and removes to block signaling link, control FE 201 blocking-up signaling links then, STP sends to FE 202 with all service request, and FE 202 visit BE 302 finish the processing of service request.

Preferably, when switching control module 501 selection auto switching target FE or target BE, can use the priority form shown in the table 1.

Each priority list item in the table 1 has been specified the auto switching strategy of FE and BE, for example, in list item 1001, the priority of FE 201 visit BE 301 is higher than BE 302, therefore, if the performance of FE 201 visit BE 302 is lower than when setting threshold values, switching that control module can switch BE 301 automatically is main usefulness, does not then allow on the contrary; Configuration item 1003, the performance of FE 201 visit BE 301 are higher than FE 202 visit BE 301, and therefore, if the performance of FE 202 visit BE 301 is lower than when setting threshold values, switching that control module can switch FE 201 automatically is main usefulness, does not then allow on the contrary.

For the situation that has plural FE and BE in the system, can switch according to the method described above equally, detailed process no longer repeats here with described process is similar before.

In addition, except shown in Fig. 4 the situation that the performance monitoring module is set on the FE, also can the performance monitoring module be set at BE, thereby the response of monitoring FE, and, in the case of necessary, the performance monitoring module can all be set on FE and BE, thereby reach the purpose of mutual detection.

In addition, in the present invention, switch control module and can block or go to block Signaling System Number 7 link, foundation or discharge switching of means control FE such as SCTP coupling and BE, concrete implementation is known in those skilled in the art, and this paper enumerates no longer one by one.

By said system, can automatic decision switch with the synchronous disaster tolerance of control FE and BE, the system that makes runs under the optimum performance all the time, particularly for the asymmetric deployment of FE and BE, can avoid because the manual service outage duration that is produced of switching.

In sum, by means of technique scheme of the present invention, on the fault auto switching basis of FE and BE, by the data performance between FE and the BE is monitored, coordinate the synchronous switching of FE and BE, can avoid unnecessary FE and BE to switch, can also guarantee the system business performance by the automatic synchronous switching of FE and BE, and can avoid the manual service outage duration that causes of switching.

The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims

1. implementation method that disaster tolerance is switched is used for the network equipment that distributed deployment and business logic processing are separated with storage is realized switching, and it is characterized in that described method comprises:

For the side in the front-end and back-end network equipment that disaster tolerance is switched takes place, rear end that disaster tolerance switches does not take place in front end after disaster tolerance is switched or rear end and the described network equipment or the service quality testing result of the data access link between the front end judges whether to satisfy the synchronous switching condition according to carrying out;

In judged result is to satisfy under the situation of described synchronous switching condition, according to the synchronous switching rule rear end or front end that disaster tolerance switches is not taken place the described network equipment and carries out synchronous switching.

2. method according to claim 1 is characterized in that, after the front end of the described network equipment or rear end disaster tolerance takes place were switched, described method further comprised:

The service quality that takes place to detect described data access link in described front end that disaster tolerance switches or rear end obtains that rear end that disaster tolerance switches does not take place for described service quality testing result and/or the described network equipment or the service quality of the described data access link of front-end detection obtains described service quality testing result.

3. method according to claim 1, it is characterized in that, the service quality testing result of described data access link comprise following one of at least: the request response time in the request sum in the predetermined amount of time between the two ends of described data access link, the predetermined amount of time between the two ends of described data access link is greater than the average request response time between the two ends of described data access link in the request quantity of response time threshold value, the predetermined amount of time.

4. method according to claim 3 is characterized in that, under one of at least the situation, judges that described data access link satisfies described synchronous switching condition below described service quality testing result satisfies:

Request response time in the predetermined amount of time between the two ends of described data access link surpasses default amount threshold greater than the request quantity of described response time threshold value, the average request response time in the predetermined amount of time between the two ends of described data access link is greater than the average response time threshold value, request response time in the predetermined amount of time between the two ends of described data access link surpasses default proportion threshold value greater than request quantity shared ratio in total request amount of the first default response time threshold value.

5. method according to claim 1 is characterized in that, described synchronous switching rule comprises the rear end of the data access link that the foundation of described front end after switching with disaster tolerance or rear end is new or the priority of front end.

6. according to each described method in the claim 1 to 5, it is characterized in that the described network equipment is one of following: authentication, authorize accounting server; Attaching position register; Equipment identity register; Home subscriber server.

7. realization system that disaster tolerance is switched, the network equipment that is used for that distributed deployment and business logic processing are separated with storage is realized switching, and it is characterized in that described system comprises the performance monitoring module, switches decision-making module and switches control module, wherein

Described performance monitoring module is arranged at the front end of each network equipment and/or is arranged at the rear end of described each network equipment, be used to detect the front end at this performance monitoring module place or rear end and, obtain the service quality testing result with front end or the rear end of back end communications or the service quality between the front end at this place;

The described decision-making module of switching is used to receive service quality testing result from described performance monitoring module, and judges front end or the rear end after network equipment generation disaster tolerance is switched and disaster tolerance does not take place according to described service quality testing result and switch the service quality of the data access link between rear end or the front end and whether satisfy the synchronous switching condition;

Describedly switch control module to be used in described judged result of switching decision-making module be to satisfy under the situation of described synchronous switching condition, according to the synchronous switching rule rear end or front end that disaster tolerance switches are not taken place the described network equipment and carry out synchronous switching.

8. system according to claim 7, it is characterized in that, the service quality testing result of described data access link comprise following one of at least: the request response time in the request sum in the predetermined amount of time between the two ends of described data access link, the predetermined amount of time between the two ends of described data access link is greater than the average request response time between the two ends of described data access link in the request quantity of response time threshold value, the predetermined amount of time.

9. system according to claim 8 is characterized in that, under one of at least the situation, the described decision-making module of switching judges that described data access link satisfies described synchronous switching condition below described service quality testing result satisfies:

10. according to each described system in the claim 7 to 9, it is characterized in that the described network equipment is one of following: authentication, authorize accounting server; Attaching position register; Equipment identity register; Home subscriber server.