JP3621634B2 - Redundant configuration switching system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is a system in which a failure is expected in a specific functional module, and the reliability is maintained by switching the functional module to maintain the function and the service is continued. The present invention relates to a redundant configuration switching technique for a high-reliability system that requires both immediacy and no immediacy. Specifically, the present invention relates to a redundant configuration switching technology for a system that realizes reliability and immediate switching of communication devices, switching devices, LANs, computer networks, various computers on a network, server machines, and the like.
[0002]
[Prior art]
Conventionally, in order to improve the reliability of the system, a redundant configuration or an N + 1 redundant configuration has been adopted, the system constituting the system or the functional module in the device is configured redundantly, and the active function module or the active functional module in the device has failed. Is configured to maintain the function by switching to the spare device or the spare function module in the device and to continue the service.
[0003]
FIG. 3 is a diagram showing a redundant configuration by duplication in a general highly reliable system as a conventional example. A description will be given based on a redundant redundant configuration of functional modules. In FIG. 3, reference numerals 1-0 and 1-1 are functional modules, and reference numerals 2-0 and 2-1 are operational data necessary for operating the functional modules in the current use.
[0004]
In a duplex configuration, two modules with the same functions, processing capabilities, and installed data are arranged, one being used as the active and the other as a backup. In the case of FIG. 3, the function modules 1-0 and 1-1 have the same function.
[0005]
The same operation data (simply referred to as data in the figure) is also stored in the data memories 2-0 and 2-1. In FIG. 3, the active module is set as the function module 1-0 and the spare module is set as the function module 1-1. If the current function module 1-0 fails, the current function module 1-0 and the spare function module 1-1 are replaced at the time of failure detection, and the operation of the apparatus is continued without interruption. , Continue the service.
[0006]
At that time, since the same operation data is loaded in the data memories 2-0 and 2-1, the processing can be continued without interruption of service. Thus, in the duplex configuration, a highly reliable system is constructed with the expectation that the probability that two devices having the same function will simultaneously fail is very low.
[0007]
FIG. 4 is a diagram showing an embodiment of an N + M redundant configuration as a conventional example. The N + 1 configuration is positioned as a configuration when M = 1. The operation of the conventional N + M redundant configuration will be described by taking M = 1 as an example. In the conventional N + 1 redundant configuration, it is assumed that N current functional modules and one spare functional module are arranged and processing is performed by load distribution.
[0008]
In FIG. 4, reference numerals 10-1 to 10-n + 1 are function modules that realize the same function, and reference numerals 20-1 to 20-n are processes in which each function module realizes a function by load distribution when all the function modules are normal. This is a data memory for storing operational data 1 to n required for each functional module 10-1 to 10-n + 1.
[0009]
If a failure occurs in any of the N current function modules 10-1 to 10-n, the function of the failure module is stopped and the spare function module 10-n + 1 is activated to replace the failure module. In this configuration, the system service is continued without interrupting the function of the apparatus or degrading the performance.
[0010]
[Problems to be solved by the invention]
The problem with the conventional duplex configuration shown in FIG. 3 is that in order to realize this, it is necessary to deploy two functional modules having complete functions in order to realize the service, and a system having a high processing capacity. In computers and servers that require a large amount of equipment and communication systems that require large-scale devices, two high-performance, large-scale modules are always required. As a result, the problem with the duplex configuration is that the scale of the apparatus is large and the economy is inferior.
[0011]
The problem of the conventional N + M redundant configuration shown in FIG. 4 is that the switching process is complicated due to the transfer of operation data necessary for realizing the function, and that the takeover time, that is, the service interruption time becomes long. .
[0012]
In FIG. 4, when the functional module 10-i out of the N current functional modules 10-1 to 10-n + 1 fails, the spare functional module 10-n + 1 performs the processing of the functional module 10-i. It is necessary to take over. At this time, it is necessary to take over operational data (simply referred to as data in the figure) i necessary for executing the processing.
[0013]
The operational data i of the functional module 10-i is loaded into the spare functional module 10-n + 1 from a higher-level management system or the like so that the function of the functional module 10-i can be taken over. At this time, it is not possible to predict which active functional module will cause the failure, so it is not possible to set data in the spare functional module in advance, and switching processing including transfer of these operational data is required when a failure occurs. Become.
[0014]
In this way, in the conventional N + M redundant configuration, when taking over the processing of a faulty module, it is necessary to load data, adjust the status, etc., and it is necessary to execute complicated procedures when recovering the system when a fault is discovered In order to continue the process and make the service the same as before the failure, a recovery time is required.
[0015]
Therefore, it is desirable to develop a method that combines the high-speed processing handover that is the advantage of the duplex configuration and the simplification of the device scale that is the advantage of the N + M redundant configuration, and solves both problems.
[0016]
The present invention is made in such a background, and provides a redundant configuration switching system capable of maintaining the reliability of the system with the minimum resources and avoiding the transient deterioration of the service accompanying the process takeover. For the purpose.
[0017]
[Means for Solving the Problems]
In the redundant configuration switching system of the present invention, N active devices or current functional modules that perform load distribution processing and one spare device or spare functional module are arranged as a device for ensuring reliability. Thus, there is no need to deploy two large-scale devices that can completely realize the service, and the device can be expected to be smaller and more economical.
[0018]
Further, as in the conventional N + M redundant configuration, when the active device or the current functional module fails, the data transfer of the operation data between the failed active device or the current functional module and the spare device or the spare functional module. And data matching processing are not required, the processing time for switching the device or functional module due to a failure is greatly shortened, and a transient service interruption or deterioration during switching can be avoided.
[0019]
Thus, the present invention can realize high-speed switching at the time of failure between small-sized and load-balanced devices or functional modules, and is inexpensive and highly reliable in communication devices, computers on large networks, and servers. A redundant configuration switching system can be provided.
[0020]
That is, the present invention includes N + 1 functional modules having the same function, and N of the N + 1 functional modules are arranged as active functional modules and one is arranged as a spare functional module. The system according to the present invention is characterized in that the N + 1 functional modules are assigned unique and consecutive module numbers, and the processing of the failed functional module is performed when a failure occurs in the functional module of module number #i. The function module having the module number # i + 1 or more is provided with means for taking over the processing of the function module having one module number.
[0021]
As a result, it is possible to perform the takeover from the working to the standby at a high speed while adopting an N + M redundant configuration with a small apparatus scale.
[0022]
That is, when a failure occurs, it is known in advance which functional module will take over the operation mode of which functional module, and an apparatus configuration for quickly taking over can be obtained.
[0023]
As an example, the function module of module number # i + 1 includes instructions for holding the operation data of the function module of module number #i together with the operation data of its own function module and the means for taking over the processing. Accordingly, it is possible to provide means for switching the own operation data to the operation data of the functional module having the module number #i.
[0024]
As a result, the takeover of the operation data can be quickly executed only by the processing inside the own functional module, and therefore the takeover time can be shortened.
[0025]
DETAILED DESCRIPTION OF THE INVENTION
The configuration of the redundant configuration switching system according to the embodiment of the present invention will be described with reference to FIG. 1 and FIG. FIG. 1 is an overall configuration diagram of a redundant configuration switching system according to a first embodiment of the present invention. FIG. 2 is an overall configuration diagram of the redundant configuration switching system according to the second embodiment of the present invention.
[0026]
As shown in FIG. 1, the redundant configuration switching system according to the first embodiment of the present invention includes N + 1 functional modules 10-1 to 10-n + 1 having the same function, and the N + 1 functional modules 10-1 to 10-10. In the redundant configuration switching system, N of −n + 1 are deployed as active functional modules 10-1 to 10-n, and 1 is deployed as a spare functional module 10-n + 1.
[0027]
Here, the feature of the present invention is that the N + 1 functional modules 10-1 to 10-n + 1 are assigned unique and consecutive module numbers # 1 to # n + 1, and the functional module 10 having the module number #i is assigned. -I When a failure occurs, the processing of the failed functional module 10-i is stopped, and each of the functional modules 10-i + 1 to 10-n + 1 whose module number is # i + 1 or more has one young module number It is in the place of taking over the processing of.
[0028]
In the redundant configuration switching system according to the second embodiment of the present invention, as shown in FIG. 2, the function module 10-i + 1 having the module number # i + 1 includes the operation data i + 1 of its own function module 10-i + 1 and the module number # i + 1. The operation data i of the function module 10-i is also held, and the own operation data i + 1 is switched to the operation data i of the function module 10-i with the module number #i according to the takeover instruction.
[0029]
In the following, embodiments of the present invention will be described in more detail.
[0030]
(First Example)
A first embodiment of the present invention will be described with reference to FIG. In FIG. 1, reference numerals 10-1 to 10-n + 1 denote functional modules that are arranged in the system and execute processing by load distribution. The N + 1 functional modules 10-1 to 10-n + 1 arranged in the system for providing the service are N modules of the functional modules 10-1 to 10-n in the operation in a state where all the modules are normal. Performs the process as active, and the functional module 10-n + 1 stands by as a spare.
[0031]
When a failure of a specific functional module 10-i is detected and it is determined that the processing cannot be continued in the functional module 10-i, the processing of the functional module 10-i is stopped and the functional module 10-i Each of the functional modules having a module number of # i + 1 or higher, such as the processing of the function module 10-i + 1 and the processing of the function module 10-n of the function module 10-n + 1. By taking over this process, it is possible to continue the service without impairing the system function and processing performance.
[0032]
(Second embodiment)
A second embodiment of the present invention will be described with reference to FIG. In FIG. 2, reference numerals 10-1 to 10-n + 1 are functional modules that are arranged in the system and execute processing by load distribution, and reference numerals 20-1 to 20-n are data for storing data groups necessary for the processing of the functional modules. In the operation with all the functional blocks in a normal state, the function module 10-1 executes processing based on the operation data 1, and the function module 10-i executes processing based on the operation data i.
[0033]
Each functional module 10-i is set with both operational data i and operational data i-1. Only the operation data 1 is set in the active function module 10-1, and the operation data n is set in the function module 10-n + 1 in standby as a backup. When a failure of a specific functional module 10-i is detected and it is determined that the processing cannot be continued in the functional module 10-i, the processing of the functional module 10-i is stopped and the functional module 10-i The function module 10-i + 1 takes over the processing of the function module 10-n, and the function module 10-n + 1 takes over the processing of the function module 10-n + 1. By taking over the processing of the functional modules it has, it is possible to continue the service without impairing the system functions and processing performance.
[0034]
Operation data required for processing takeover is switched between two previously set operation data, so that no new operation data needs to be transferred or set between any function modules when a failure occurs in the function module 10-i. Accordingly, the service data can be continued by autonomously switching the operation data held by each functional module.
[0035]
Thereby, it is possible to configure a redundant configuration switching system that realizes high reliability and immediate switching as well as economy. The redundant configuration switching system of the present invention can be applied to communication devices, switching devices, LANs, computer networks, various computers on a network, server machines, and the like.
[0036]
【The invention's effect】
As described above, according to the present invention, it is possible to realize high-speed switching at the time of failure with a small and economical device, and it is inexpensive in information systems such as communication systems, computers on large-scale networks, and servers. A highly reliable redundant configuration switching system can be provided.
[Brief description of the drawings]
FIG. 1 is an overall configuration diagram of a redundant configuration switching system according to a first embodiment of the present invention.
FIG. 2 is an overall configuration diagram of a redundant configuration switching system according to a second embodiment of the present invention.
FIG. 3 is a diagram for explaining a conventional duplex configuration switching system;
FIG. 4 is a diagram for explaining a conventional N + M redundant configuration switching system;
[Explanation of symbols]
1-n Operation data 10-1 to 10-n + 1 Functional modules 20-1 to 20-n Data memory

Claims (1)

N + 1 functional modules having the same function are provided,
In a redundant configuration switching system in which N of the N + 1 functional modules are deployed as active functional modules and one is deployed as a spare functional module.
The N + 1 functional modules are given unique and consecutive module numbers,
When a failure occurs in the functional module of module number #i, the processing of the functional module that has failed is stopped and each of the functional modules having a module number of # i + 1 or higher takes over the processing of the functional module having one module number. With means ,
In the function module of module number # i + 1, the operation data of the function module of module number #i is also stored together with the operation data of the function module of itself, and the operation data of itself is received according to the instruction of the means to take over the processing. A redundant configuration switching system, comprising: means for switching to operation data of the functional module of module number #i .

Images