JPH0689197A - Hot standby system - Google Patents

Abstract

PURPOSE:To improve the use efficiency of a computer system by eliminating the standby system which is not operated in the normalcy in the hot standby system where plural computer systems are so operated that they back up one another. CONSTITUTION:Each of computer systems 5A to 5C is provided with a current application 11RE, a standby application 11ST which can be substituted with the current application of another computer system, and subsystems which correspond to these applications respectively and can be substituted with each other. At least one of subsystems corresponding to standby applications 11ST in computer systems 5A to 5C is started; and if any computer system is down, this computer system is immediately backed up by the standby application 11ST of another computer system corresponding to the current application system 11RE of this down computer system and the sub-system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a hot standby system in which a plurality of computer systems are operated so as to back up each other.

Computer systems are used in many fields, including financial institutions such as banks and securities companies.
The need for continuous operation for 365 days is increasing. Therefore, in case of a failure of the business software, it is usual to use a hot standby system that takes into consideration the fact that the backup can be performed immediately.

In such a case, there is a demand for an efficient hot standby system that effectively uses resources, especially CPU without making them idle.

[0004]

2. Description of the Related Art In a conventional hot standby system, there is a one-to-one correspondence between an active system, which is a computer system that normally performs business, and a standby system (or a standby system), which is a computer system for backing up the system. is doing. In addition, a standby system is always required even in a configuration using a plurality of active systems.

For example, in order to improve reliability, financial institutions are operated by two independent systems, an accounting system involving financial transactions and an information system other than that, and a standby system for them. Is provided.

FIG. 4 shows a conventional hot standby system 8
It is a block diagram showing 0. The hot standby system 80 is an active system 81 that performs business in normal times.
82 and standby system 8 for their backup
3 and 3.

One active system 81 is, for example, an accounting system, and the other active system 82 is an information system. The standby system 83 is used for both the active systems 8
The status of 1,82 is constantly monitored, and the current system 8
Be prepared for the occurrence of 1,82 failures.

In the standby system 83, an accounting system that requires more urgency when a failure occurs is set up,
When a failure occurs in the active system 81, the business of the accounting system is immediately taken over.

[0009]

However, the conventional hot standby system 80 described above has the following problems.

A standby system 83, which is separate from the active systems 81 and 82, is required. That is, the standby system 8
No. 3 is hardly operated when the current systems 81 and 82 are operating normally, and the utilization efficiency of the resources (especially the CPU) of the computer system is poor.

The standby of the active systems 81 and 82 is only the standby system 83. The active systems 81 and 82 do not serve as the standby of other systems, and the standby system is fixed. Therefore, it is difficult for one standby system 83 to sufficiently support a plurality of active systems 81 and 82.

When a plurality of active systems are supported by one standby system, when all the active systems have a failure and are operating in the standby system, even if one active system is not restored, the standby system If a failure occurs, the backup cannot be done and the whole system goes down.

In view of the above problems, it is an object of the present invention to eliminate a standby system that is not normally operated and improve the utilization efficiency of a computer system.

[0014]

In order to solve the above-mentioned problems, a system according to the invention of claim 1 has a plurality of computer systems 5A, 5B, 5 as shown in FIGS.
In the hot standby system 1 in which Cs are operated so as to back up each other, the active application 11 is provided in each of the computer systems 5A, 5B and 5C.
RE and standby application 11S that can replace the active application 11RE of another computer system
T and a sub-system 12 corresponding to each of the active application 11RE and the standby application 11ST, which are mutually substitutable, are provided, and in each of the computer systems 5A, 5B, and 5C, a sub-system corresponding to the standby application 11ST If at least one system 12 is started up and one of the computer systems 5A, 5B, and 5C goes down, the standby application 11ST of any other computer system corresponding to the active application 11RE of the down computer system, The corresponding subsystem 12 immediately backs up.

A system 5 according to a second aspect of the present invention includes a plurality of different applications 11, a plurality of sub-systems 12 that can be replaced with each other, and the sub-systems 1.
An identifier DS for identifying each of the two
When there is a service request from the application 14 and the means 14 for giving C, one of the subsystems 12 is selected by selecting one of the subsystems 12 based on the identifier DSC included as a parameter in the service request. And a means 13 for associating with each other.

A system 5 according to a third aspect of the present invention manages a plurality of different applications 11, a plurality of subsystems 12 having mutually replaceable and different identifiers DSC, and each of the subsystems 12. Management control table 42 for controlling the application 1
When there is a service request from No. 1, any one of the subsystems 12 is selected based on the identifier DSC included as a parameter in the service request and the management control table 42, and is associated with the application 11. And a means 13 for performing.

In the system 5 according to the fourth aspect of the present invention, each subsystem 12 has a control table 32 storing the address position of each subsystem, and the management control table 42 includes each of the subsystems. The control table 32 has a pointer PT2 that points to an address position.

In the method according to the fifth aspect of the present invention, the subsystems 12 are activated by providing the identifiers DSC for identifying each other, and the application 1
When 1 makes a service request for use of the subsystem 12, the monitoring program 43 is called with the identifier DSC as its parameter, and the monitoring program 43 selects the subsystem 12 designated by the application 11 and Corresponds to the application 11.

[0019]

Each computer system 5A, 5B, 5C is provided with an active application 11RE, a standby application 11ST, and a substitutable subsystem 12.

Each computer system 5A, 5B, 5C
In, at least one subsystem 12 corresponding to the standby application 11ST has been started up,
Any other computer system 5A, 5B, 5C
When is down, the standby application 11ST and the subsystem 12 corresponding to the down working application 11RE immediately back up.

That is, in each computer system 5A, 5B, 5C, a plurality of subsystems 12 of the same type can be activated simultaneously. The identifier DSC is used to identify the plurality of subsystems 12, and the corresponding subsystem 12 is associated with the pointer PT2 of the management control table 42 in response to the service request from the application 11.

[0022]

1 is a block diagram of a computer system 5 used in a hot standby system 1 according to the present invention.

The computer system 5 includes a plurality of applications (or subsystem users) 1 different from each other.
1 (11a to n), a plurality of sub-systems 12 (12a to n) that can be replaced with each other, a subsystem selecting mechanism 13, an identifier assigning mechanism 14, and the same subsystem starting mechanism 15.
have.

The system administrator determines the number of subsystems 12 and their identifiers DSC in consideration of the operation of the computer system 5. The determined identifier DSC is given to the identifier assigning mechanism 14. The identifier DSC is, for example, “0
For example, a numerical code in the range of 0 to 99.

The identifier assigning mechanism 14 manages the given identifier DSC and informs the same subsystem starting mechanism 15 of the required identifier DSC. The subsystem selection mechanism 13 is also notified of the identifier DSC.

The same subsystem activation mechanism 15 operates the required subsystems 12 according to the number of notified identifiers DSC, and the subsystems 12 are activated at each activation.
To each identifier DSC.

The subsystem selection mechanism 13 receives a service request from the application 11 and determines one of the subsystems 12 according to the identifier DSC attached as a parameter of the service request. At the same time, the application 11 and the subsystem 12 are associated with each other, and the request is transmitted to the subsystem 12.

That is, the subsystem 12 has the identifier DSC.
Are identified by each other and are associated with the corresponding subsystem 12 by the identifier DSC designated by the application 11. Therefore, the identifier D
By changing the value of SC, application 11
And the subsystem 12 can be freely combined.

FIG. 1 is a block diagram showing details of essential parts of a computer system 5 according to the present invention. The application 11 has application programs 21 (21a to 21n) different from each other and a control table 22 used by each application program 21. When issuing a service request to the subsystem 12, each application 11 writes the identifier DSC (DSCa-n) in the control table 22 as its parameter.

The subsystem 12 includes subsystem programs 31 (31a to 31n) for virtual communication access method (VTAM), and each subsystem program 3 for example.
Control table 32 (32a to n) used for No. 1 of FIG. Therefore, each subsystem 12 operates according to the respective control table 32.

The respective subsystem programs 31a to 31n are the same as each other and can therefore be replaced. Each subsystem program 31a-n is provided with an identifier DSCa-n at the time of its startup.

The subsystem selection mechanism 13 has a hardware fixed core (FLC: for storing fixed addresses of the control table).
A fixed low core (hereinafter referred to as “FLC”) 41, a main control table management control table (hereinafter referred to as “MCMC”) 42, a supervisor 43, and the like.

One fixed address is assigned to one type of subsystem in the FLC 41, and a pointer PT1 for pointing the MCMC 42 to the specific fixed address.
Is stored.

The MCMC 42 has, for each identifier DSC, a pointer PT2 pointing to each control table 32 of each subsystem 12 and an ASCB (Addre) indicating the space.
sSpace Controlled Block).

Therefore, when any of the applications 11 (11a to 11n) issues a service request, the control is passed to the supervisor 43 by the SVC instruction, and the pointer PT1 of the FLC 41 is read by the control of the supervisor 43. The MCMC 42 is pointed to by the pointer PT1, and the corresponding control table 32a-n is found by referring to the identifier DSC attached to the service request. A service request is issued to the relevant subsystem 12 based on the result.

In this way, a plurality of subsystems 12 are activated and simultaneously operate in response to service requests from a plurality of applications 11, and they are associated with each other.

FIG. 3 is a diagram for explaining an operating state of the hot standby system 1 using the computer system 5 configured as described above. The hot standby system 1 of FIG. 3 has three computer systems 5A, 5 having three applications 11 and three subsystems 12 (denoted by these) that are equivalent to each other.
It is a leash coupling system composed of B and 5C.

In FIG. 3A, in each of the computer systems 5A, 5B, and 5C, the applications (businesses) of ... Operate under the control of their respective subsystems as active applications. Further, another application that is not the active application, for example, the application in the computer system 5A, is a standby application that can replace the active application of the other computer system 5.

Each of these computer systems 5A, 5
B and 5C perform hot standby of other computer systems with each other. In FIG. 3, the active application is a double circle and the standby application is 1
Shigemaru, respectively. Also, the current application is code 11RE and the standby application is code 11ST.
, Respectively.

Here, if the computer system 5A or the subsystem for processing the business is stopped due to a failure, the processing of the business is executed by the computer system 5B as shown in FIG. 3 (b). In other words, in the computer system 5B, the subsystem of business operates.

In this state, if the computer system 5B now fails, as shown in FIG. 3C, the remaining two subsystems in the hot standby mode in the computer system 5C take over the processing of the task. To do.

Then, after a while, the computer system 5
The fault of A is recovered. Then, as shown in FIG. 3D, the computer system 5A starts hot standby of the jobs processed by the computer system 5C. After that, if a failure occurs in the computer system 5C that performs three tasks,
As shown in (e), the restored computer system 5
A will take over those duties.

As described above, the three computer systems 5A, 5B, and 5C perform hot standby for the other computer systems 5 as long as no fault occurs, and a fault occurs in a plurality of computer systems 5. At least one computer system 5
Can operate, it is possible to avoid the suspension of the entire work of the hot standby system 1.

Moreover, in a normal state, all computer systems 5 are executing the active application, each job is smoothly processed, and each resource, particularly the expensive CPU, is effectively used.

In the above-described embodiment, the unit or the number of applications or jobs can be various. The configuration of each part of the computer system 5 or the hot standby system 1, the processing contents, the processing order, etc.
Various changes can be made along the gist of the present invention.

[0046]

According to the present invention, it is possible to improve the utilization efficiency of a computer system by eliminating a standby system that is not normally operated.

According to the second, third and fifth aspects of the present invention, it is possible to simultaneously operate a plurality of sub-systems of the same type which are mutually replaceable. According to the invention of claim 4, subsystems of the same type can be easily identified from each other and a plurality of subsystems can be operated at the same time.

[Brief description of drawings]

FIG. 1 is a block diagram showing details of essential parts of a computer system according to the present invention.

FIG. 2 is a block diagram of a computer system used in the hot standby system according to the present invention.

FIG. 3 is a diagram for explaining an operating state of the hot standby system according to the present invention.

FIG. 4 is a block diagram showing a conventional hot standby system.

[Explanation of symbols]

 1 Hot Standby System 5, 5A, 5B, 5C, Computer System 11 Application 11RE Working Application 11ST Standby Application 12 Subsystem 13 Subsystem Selection Mechanism (Means for Correlating) 14 Identifier Granting Mechanism (Means for Granting) 32 Control Table 42 MCMC (Management control table) 43 Supervisor (monitoring program) DSC identifier PT2 pointer

Claims (5)

[Claims] 1. A plurality of computer systems (5A, 5A)
B, 5C) is a hot standby system (1) operated so as to back up each other, wherein each computer system (5A, 5B, 5C) has an active application (11RE) and an active computer of another computer system. Application (1
1RE) standby application (11S)
T) and sub-systems (12) that can be replaced with each other corresponding to the active application (11RE) and the standby application (11ST), respectively, and the respective computer systems (5A, 5B, 5)
In C), the standby application (11ST)
At least one subsystem (12) corresponding to is started up, and any one of the computer systems (5A, 5B, 5C)
When a down occurs, the standby application (11ST) of any other computer system corresponding to the active application (11RE) of the down computer system and the subsystem (12) corresponding to it
A hot standby system characterized by immediate backup by. 2. A plurality of computer systems (5A, 5A)
A computer system (5) used in a hot standby system (1) in which B, 5C) are operated so as to immediately back up each other, and a plurality of mutually different applications (11) and a plurality of sub-substitutes that can be mutually replaced. A system (12) and a means (1) for giving an identifier (DSC) for identifying each of the subsystems (12) to each other.
4) and when there is a service request from the application (11), based on the identifier (DSC) included as a parameter in the service request, the subsystem (1)
Select one of 2) and select the application (1
A computer system used for a hot standby system, comprising: a means (13) for associating with 1). 3. A plurality of computer systems (5A, 5A)
A computer system (5) used in a hot standby system (1) in which B, 5C) are operated so as to immediately back up each other, and which are mutually replaceable with a plurality of different applications (11). Different identifier (DS
C), a plurality of subsystems (12), a management control table (42) for managing each subsystem (12), and a service request from the application (11). Means (13) for selecting one of the subsystems (12) based on the identifier (DSC) included as a parameter in the service request and the management control table (42) and associating it with the application (11). ), And a computer system used in a hot standby system. 4. The computer system (5) according to claim 3, wherein each subsystem (12) has a control table (32) storing an address position of each subsystem, and the management control The table (42) is a pointer (PT) indicating the address position of the control table (32) of each subsystem.
A computer system used in a hot standby system, characterized in that the computer system comprises 2). 5. A method of operating a plurality of substitutable sub-systems (12) in a computer system used in a hot standby system (1) for identifying each sub-system (12) from each other. When the application (11) issues a service request for use of the subsystem (12) with the identifier (DSC) added and started, the identifier (DSC) is added as a parameter to the monitoring program (4).
3) and calls the monitoring program (43) to select the subsystem (12) specified by the application (11) and associate it with the application (11). How the subsystem works.