WO2010145428A1

WO2010145428A1 - Method and system for implementing disaster recovery switch

Info

Publication number: WO2010145428A1
Application number: PCT/CN2010/073261
Authority: WO
Inventors: 章恩华
Original assignee: 中兴通讯股份有限公司
Priority date: 2009-10-29
Filing date: 2010-05-26
Publication date: 2010-12-23
Also published as: CN102056207A; CN102056207B

Abstract

A method and system for implementing disaster recovery switch are disclosed. The method includes: for a network equipment whose disaster recovery switch occurs in one party of a front end (FE) and a back end (BE), judging whether the condition of the synchronization switch is satisfied, according to the detection result of quality of service (QoS) of the data access link between FE or BE after the disaster recovery switch and BE or FE of the network equipment in which the disaster recovery switch does not occur; if satisfying the condition of the synchronization switch, synchronously switching BE or FE of the network equipment in which the disaster recovery switch does not occur, according to the rule of the synchronization switch. It is possible to harmonize the synchronization switch of FE and BE and ensure the capability of the system service by monitoring QoS between FE and BE.

Description

Method and system

This relates to the field of communications, and in particular to methods and systems. Background

5 location register (ho e oca o Tegse ) for storing all the users, see the high speed, and the increase of security and integration requirements, so that R 1 1 + and + each program force distribution R

At present, the distribution R has the following characteristics: (1) the front end of the responsibility (Fo E d, FE) and the user-storage (Bac d, BE). FE and BE can be deployed in the same geographical station, or can be deployed in Different geographic stations (2) FE and BE can be configured separately, you or non-you deploy, and, in the case of non-deployment, FE and BE are not equal, and can be stored centrally.

5 Usually, in the various R schemes, yes, near

Both need to deploy FE and BE, and FE can only have this BE. In the distribution R, FE can be used for all BEs, but it can be effective for rabbit FE needs and BE.

Unrestricted. So that the R scheme of the distribution R and the system have the following R schemes provided by the following R schemes, which are provided to the other 0 stations, Hugh, in the main direction, including each FE to FE, and teach BE to BE and the distribution R, FE and BE can be respectively, that is, FE has a barrier, and only need to be connected to each FE, and no BE synchronization is required. Same, BE Need, does not require FE synchronization.

1 +1 deployment of each system. As shown in Fig. 1, the R 401 is synchronized to R 402 in the way of storage, or use, and is connected to (sg a g a s e po , ST ) 01 102

R is from 601 to 801, pointing to R402. Before reaching R402, STP 01 102 sends all of them to R401. In the main R401, F 01 BE301 completes each.

F 01 or BE301 barrier, to R402, stop BE301 and BE302, and STP 01 102 and F 01 signaling, all sent to each R402

2 Distribution R deployment. As shown in Figure 2, F 01, each F 02, BE301, and each BE302 can be deployed in different geographical locations.

FE 201 barriers, each to each FE 202, each FE 202 can

BE301 completes each.

Since FE and BE are allowed to be deployed on different geographical stations, FE and BE are the same, which limits the performance of FE BE. Because of the different performance characteristics of FE and BE, such as FE, the BE can be optimally ideal. For example, in the system shown in Figure 2, FE may have the best performance in BE, and each BE The performance is poor. If the FE is out, the BEs are not synchronized, so that the BE FEs are very different, the performance is degraded, and the performance cannot meet the requirements. The same reason, in the case of BE sheep, the same occurs. Similarly, except distributed Outside the deployment of R, other "distributed deployments, and each storage of each other (for example, including, , ( he ca o A hoTza o Acco, AAA, user Each (o eS bsc be Se Ve , SS, each register (qpe de y Regse ER)) exists because each of the FE and BE are not synchronized.

of .

In the distributed deployment of the phase, the BE FE with the non-synchronization of the BE and the FE cannot be obtained or even reduced in performance, and no effective solution has been proposed yet.

Content

In the distributed deployment of the phase, the performance of the FE BE due to the unsynchronized FE and BE cannot be obtained, or even

Performance is reduced and often

, FE BE with the same number of performance.

The wood of this is

In this aspect, the methods provided are distributed, distributed, and stored separately.

The method includes the front end of each of the front and the side, or each of the unfamiliar customers, in the case that the result satisfies the synchronization, the synchronization rules are not synchronized with the front end or the front end.

At each front end or a customer, the method can include the front end of each of the friends or each

The results, and/or each of the unoccupied or front ends each yields a result.

, each of the results includes the following at least the same within the same predetermined The same is the same as the same number of the same value, the same average within the same.

And, if each result satisfies at least the following,

The synchronization synchronization rule may include the front end or establish a new one, if the same number of the same number in the same period as the synchronization is greater than the same value, and the same average within the same time is greater than the average, and the same within the predetermined same size. Or front end first.

, each can force the following, each, each

Location register Each register user.

In another aspect of the present invention, the distributed, distributed deployment, and each storage system may include performance monitoring, decision making, and control.

Hugh, performance monitoring at each front end, and/or each of the front end of the performance monitoring or the number of the communication or front end, each result is determined to receive the results of the performance monitoring, each or the front end Whether the same number satisfies the synchronization

It is controlled that the synchronization rules are not synchronized with each other or the front end is synchronized when the result of the decision meets the synchronization.

Wherein, each result may include at least the following Same as the same inside

The same as the same value of the same value, the same as the same within the same.

In the case where the results satisfy at least the following, the decision can satisfy the synchronization

The same average of the same and greater than the same value, which is greater than the average, and the same within the same

Proportion of the same value of greater than the same

proportion .

Each of the following registers can be used for each of the register and user registers.

With the help of this board, the FE BE is controlled by the same (performance) line, the synchronization of FE BE, the unnecessary FE and BE of the rabbit, the self-synchronization performance of FE and BE, and the manual Each of the same. Description

1 is the deployment of +1 each system

2 is the distribution of the wood R deployment

3 is the flow of the method of the method

4 is the unified process of this system. Hugh way

The distributed deployment in the phase, and the parties in the FE and BE on each storage (for example, R, each, SS ER, etc.) The same performance of the FE BE that is not synchronized, this considers that FE and BE in the distribution system can be, but in some cases, the parties in FE and BE can be made more synchronized. Good performance, therefore, it is proposed that each FE of each FE is synchronized with each other in FE or BE, and if so, FE BE is the same

Poor performance, then FE and BE are not synchronized.

The following will be combined with this one.

Method

In this paper, methods are provided for distributed deployment, and each storage (for example, may include R SS AA, ER, etc.).

In this section, you first need the communication status configuration rule. In the case of FE, the FE can be established by BE first, so that the FE of the FE main FE FE is similar to that of the FE. For the BE, the BE can be established by the FE first, so that the BE is the BE of the BE BE. You can put the rules table, as shown in Table 1.

1

Performance sequencing between lounges

1001 FE201 BE301 BE302

1002 FE202 BE302 BE301

1003 BE301 FE201 FE202

1004 BE302 FE202 FE201 rules can be pre-configured, and the configuration results are FE BE The geographical location and other factors, this article is no longer.

This requires all of the front ends and/or the ability to configure the number of monitor front ends and/or phases.

For example, in the deployment shown in 2, F 01 can

BE301 is the same as the same, BE301 can also be the same as 01, each BE 302 can also be the same as each FE 202, and the necessary FE BE can be the same.

In the service, there are many ways to compare sheep and the effective way is /. The way to the right, that is, FE and / or BE in the same to return to the same.

In the result, in FE BE, information such as the same amount, the same amount, and the same value of mercury can be determined.

The following will be combined with this one.

As shown in Figure 3, the method of this includes

Step S302, in each of the FE and the BE,

Whether the results of the same number of BE or FE of the FE or BE (ie, the newly established number) satisfy the synchronization, the synchronization may have a multi-god situation, for example, each of the roads may not satisfy the requirements. Synchronization step 5304, in the case that the result satisfies the synchronization, the synchronization rule (the rule can be pre-configured) synchronizes the BE or FE of each non-guest.

In step S302, each FE or BE customer, each of the customer's FE or BE needs to obtain each And each of the unexposed BE or FE each obtained each result.

The results of , may include at least the same within the same predetermined

If the demand is met, if at least the following is satisfied, the force cannot satisfy the ratio of the predetermined value of each demand, the proportion of the predetermined value, and the average of the same average.

Due to the different requirements of different quantities, the sheep can be used or combined. For example, in each of the requirements, you can refer to the above.

For example, in the deployment result 01 to the FE 202 shown in FIG. 2, if each of the FEs 202 that are required by each FE 202 does not satisfy the respective requirements, that is, the result is (ie, the synchronization is not satisfied), then the BE301 is not synchronized. Otherwise, it is necessary to combine BE301 to each BE302, and each of the F 02 and each BE302.

In the case where the BE first and FE are synchronized, the method is similar, and it is no longer.

In the synchronization, as shown in Table 1, each of the self-strategies that specify FE and BE, for example, is higher than BE302 at 1001 F 01 BE301, so the performance of F 01 BE302 is lower than (the requirements cannot be met, ie , to meet the synchronization), control from the BE301 Force master, on the contrary, is not allowed to configure 1003, FE 201 BE 301 performance is higher than 02 BE301, therefore, the performance of F 02 BE301 is lower than

, control will be F 01 force master, but not allowed.

On, self-decision and control of the synchronization of FE and BE, so that the system is always in the best performance, especially for the non-deployment of FE and BE, rabbits are the same because of the manual exchange of households.

System

In this paper, the system is provided in a distributed deployment, and each storage (for example, including SS ER, etc.).

The system includes performance monitoring, decision making, and control.

The performance is monitored in each FE, and/or in each BE, and the FE or BE where the performance monitoring is located is the same as the BE or FE of the FE or BE where the performance is monitored, and each result is determined to receive the results of the performance monitoring. , each result of each customer's FE or BE does not have a customer BE or FE, whether each meets the pre-configured synchronization control, in the case that the result of the decision meets the synchronization, the pre-configured synchronization rules are not the raw BE or FE synchronization.

Wherein, each of the results may include the following at least the same

The same is the same as the same average of the same value. In the case that each result satisfies at least the following, the decision can satisfy the synchronization

The same number of the same number greater than the same value, the same as the same

The proportion of the same number of values greater than the sum of the quantities

proportion .

Each of the following registers can be used for each of the register registers. System.

4 is the unit of this.

4, the distribution of each of the main F 01 and each F 02, the number of storage in the different geographical distribution R is stored by the main BE301 and each BE302, deployed in different geographical locations, and BE301 and each BE 302 can 01. Each F 02 is not in the same geographical location.

In this example, performance monitoring is deployed on the FE, that is, as shown in FIG. 4, the performance monitoring 101 is deployed on the main F 01, the performance monitoring 102 is deployed on each F 02, and the performance monitoring 101 is monitored on the FE 201 BE 301. 102 monitors each F 02 and BE302.

The decision 401 and the control 501 can be deployed on the same or separately, and the decision 401 and the control 501 can be deployed in the same geographical location.

Below is the F 01 barrier.

In the FE barrier, FE BE performance FE and BE .

In the signaling path of F 01 , STP and F 01 , each and every STP will be sent to each F 02 . Since the BE301 is unobstructed, the users of the F 02 BE301 complete each. Performance monitoring 102 can monitor the same as F02 BE301 and make decisions 401 on a regular basis. The performance monitoring 102 can be pre-configured with the same T, so that the information of the upper decision 401 can include the same, the same and the same as the same T.

Decision 401 collects performance on performance monitoring 102 and analyzes the performance of all performance monitoring of pre-configured FE 202 within T2.

Decision 401 uses pre-configured performance to determine if BE is required. The performance may include an average of the same T3 and the ratio T4. If the ratio is the same as T3 in the same T2, or the ratio of the same T is greater than the ratio T4, the decision 401 requires BE self, and the control system 501.

Control 501 receives an indication of decision 401, whether or not to allow BE to be self-contained, and to introduce a new BE. If the fruit distribution is allowed, and each BE302 is in a row, the control 501 first controls each of the FE 301, and then controls each FE 302.

BE302 to BE301, F 02 BE302 complete each.

The following will be the BE301 barrier, the rest of the system in the BE barrier, FE BE performance FE and BE

step. The BE301 barrier requires each BE302 to be the master, while the STP still sends all of them to 01, 01 and only the BE302 users can complete each.

Performance Monitoring 101 monitors FE 201 BE 302 in the same way as decision-making 401. The performance monitoring 101 can be pre-configured with the same T, so that the information of the upper decision 401 can include the same, the same and the same as the same T.

Decision 401 collects performance on performance monitoring 101 and analyzes the performance of all performance monitoring of pre-configured FE 201 within T2. Decision 401 Use pre-configured performance Undetermined if FE is required. Performance can include averages with T3 and

Ratio T4, if it is within the same T2, the average is the same as T3, or the ratio of the same is greater than the same T is greater than

For the ratio T4, the decision 401 requires FE self-control and 501 control.

Control 501 receives an indication of decision 401, whether FE is required to be configured, and FE is the new FE. 2, the fruit distribution configuration is allowed, and each F 02 line is always (and the same amount of FE202 BE302 can meet the requirements), then control 501 first FE 202 command, then control FE 201 command, STP will send all to F 02, F 02 BE302 complete each.

, Control 501 from FE or BE can use the first table shown in 1.

Each of the 1s specifies the self-policy of FE and BE, for example, In 1001, F 01 BE301 is higher than BE302 first. Therefore, if the performance of FE 201 BE 302 is lower than that of BE BE 301, the opposite is not allowed to configure 1003. The performance of FE 201 BE 301 is higher than F 02 BE301. Therefore, the performance of F 02 BE301 is lower than

Control self-pressing F 01 force master, but not allowed.

In the case of the above FE and BE in the system, the same can be done according to the above method, before the break, no longer. Therefore, in addition to the performance monitoring on the FE shown in 4, it is also possible to monitor the FE performance, and monitor the FE, and, if necessary, the performance monitoring on the FE and BE, and to the mutual measurement. the goal of.

In addition, in the present invention, the control can be either stipulated, established or SCTP, and BE, and the way of Hugh is in this field, and this article is no longer.

Up-to-date, self-determination and control of the synchronization of FE and BE, so that the system always performs at the best performance, especially for the non-deployment of FE and BE, the rabbits are replaced by the manual exchange of households.

With the help of the above-mentioned wood, in the FE and BE barriers, the FE BE has the same performance line monitoring, FE BE synchronization, rabbit unnecessary FE and BE, near FE and BE self-conformity performance And the rabbits are handmade. The above is better than the limit, and any modifications, equivalents, etc. made within this spirit and principle are included in this document.

Claims

Claim

1. Method for distributed deployment, and each

Each of the stored items, the method includes

Everyone in the front and the middle of the square,

In the case where the result satisfies the synchronization, the synchronization rules are not synchronized with each other or the front end.

2. The method of claim 1 is characterized in that, in each of the front ends or the customer, the method step comprises

The front end of the friend

To each result, and/or to each

3. The method of claim 1 is characterized in that each of the results includes at least the same, the same within the same predetermined, the same number greater than the same value, and the same average within the predetermined same.

4. The method of claim 3, characterized in that

If at least the following conditions are met, the same and greater than the same within the synchronization schedule are satisfied.

The number of the same value, the same as

The same average is greater than the average, the same as the same within the same

5. The method of claim 1 is characterized in that the synchronization rule comprises first.

6. The method of any one of the items 1 to 5 is characterized in that each of the following registers and each register of each position register is used.

7. System, distributed deployment, and

Each of the storage includes a performance monitoring, a decision, and a control, wherein

Performance is monitored at each front end, and/or each of the front ends of the performance monitoring or the same number of communications or front ends.

Satisfy synchronization

It is controlled that the synchronization rules are not synchronized with the front end or the front end when the result of the decision is satisfied.

8. The system of claim 7 is characterized in that each of the following results includes at least the same, the same and the same within the predetermined same size, the same number of the same value, and the same average within the same time.

9. The requirement of 8 is that the decision is made when the results satisfy at least the following

Satisfy the synchronization

The same and the same within

The number of the same value, the same as

The same average is greater than the average and the same as the same as stated in the same

The ratio of the proportion of the same value to the proportion of the quantity.

10. The system of any one of the items 7 to 9 is required, and each of the registers of the respective registers is different from each other.