JPH09297692A - Multiplexing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To carry on the work of a faulty computer via another normal computer by registering a standby processor and the shared resources against a working processor and loading a work program into an independent space via the standby computer when the working computer has a fault. SOLUTION: A processor 1 of a standby system 5 is previously set on a constitution information table 6 against a processor 1 of a working system 3. If the fault of the processor 1 of the system 3 is detected when the corresponding processor 1 of the system 5 is decided based on the table 6, a work program 4 of the processor 1 of the system 3 is loaded into an independent space 2 and the work of the faulty processor 1 is carried on. In such cases, another processor 1 of the system 3 is set as a processor 1 of the system 5 contained in the table 6. When a fault occurs, the processor 1 of the system 5 loads the program 4 of the faulty processor 1 into another independent space 2 different from the work under operation. Thus, plural tasks are carried out in parallel to each other.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multiplexing system for multiplexing processing devices.

[0002]

2. Description of the Related Art Conventionally, there is a system in which a computer in operation and a computer in standby are provided, and when the computer in operation stops for some reason, the computer in standby continues to operate in place of the down computer. . In this system, one standby computer is provided in association with one operating computer.

[0003]

Therefore, it is necessary to prepare a standby computer in a one-to-one relationship in association with a computer in operation, and one computer is in operation for N computers in operation. If a standby computer is associated and one of the computers fails, the standby computer can continue its work, or
When one of the operating computers fails, another computer cannot flexibly set the configuration such that other computers can continue the work of the failed computer together with their own work. was there.

[0004] The present invention solves these problems,
A configuration information table that registers the standby system processing devices and shared resources in association with the active system processing devices is provided, and according to the information set in this table, the standby system processing devices can be The purpose is to load a business program into an independent space, continue the business, and flexibly associate the operating system and the standby system of the processing device.

[0005]

Means for solving the problem will be described with reference to FIG. In FIG. 1, a processing device 1 loads a business program 4 to perform processing, and here, is composed of a space 2 and a configuration information table 6 and the like.

[0006] The space 2 is for loading business programs 4 and performing business, and is independent of each other. Here, the business program 4 is loaded into the space 2 of the operation system 3 to perform business.

The configuration information table 6 pre-sets the processing device 1 and the shared resource 7 of the standby system 5 in association with the processing device 1 of the active system 3. The shared resource 7 is various resources shared in the space where the business program 4 is loaded, and is, for example, a disk device or a printer device.

Next, the operation will be described. The processing device 1 of the standby system 5 is set in advance in the configuration information table 6 in association with the processing device 1 of the active system 3, and when the processing device 1 is found to be the standby system 5 by referring to this configuration information table 6 When a failure of the processing device 1 of the active system 3 is detected, the business program 4 of the processing device 1 is loaded into the independent space 2 to continue the business.

At this time, the standby system 5 in the configuration information table 6
As the processing device 1 of the other active system 3, the processing device 1 of the other active system 3 is set, and the business of the processing device 1 in which the failure occurs in the independent space 2 different from the business in which the processing device 1 of the standby system 5 is operating. The program 4 is loaded to execute a plurality of jobs in parallel.

Therefore, the configuration information table 6 for registering the processing device 1 of the standby system 5 and the shared resource 7 in association with the processing device 1 of the active system 3 is provided, and the processing of the active system is performed according to the information set in this table. Standby system 5 when a failure occurs in device 1
When the processing device 1 of FIG. 1 loads the business program 4 into the independent space 2 and continues the business, the processing device 1
It has become possible to flexibly associate the active system 3 and the standby system 5 of.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, an embodiment and operation of the present invention will be described in detail with reference to FIGS.

FIG. 1 is a block diagram showing one embodiment of the present invention.
FIG. 1A shows a configuration diagram. In FIG. 1A, the processing device 1 loads the business program 4 into the space 2 independently to perform various processes, and loads the business program into the space 2 according to a preset configuration information table 6. This is done (described later with reference to FIG. 2). The processing device 1 is composed of a space 2 for loading a business program 4 and a configuration information table 6 here.

The space 2 is for loading and activating a business program of a class according to the configuration information table 6 and performing business processing using the shared resource 7. When the space 2 is different, the business processing is independently performed without being affected, and when the business programs are loaded into a plurality of spaces in the one processing device 1 and activated, each is affected. Work independently. Here, the active system 3 loads the business program 4 into the space of the processing device 1 set as the active processing device in the configuration information table 6 to perform the business process. On the other hand, when the standby system 5 detects a failure in the active system processing device 1 corresponding to the processing system 1 set as the standby system processing device in the configuration information table 6,
The business program 4 is automatically loaded into the space 2 to continue business processing. The business programs 4 loaded into the space 2 are programs that are different for each business, such as A business programs, B business programs, and C business programs, and are programs set in the configuration information table 6.

As shown in FIG. 1B, the configuration information table 6 is used to set the classes, business programs, operating system / standby system distinction, shared resources, etc. for the processing apparatus 1.

The shared resource 7 is a resource used for each space 2 and is a shared resource such as a printer device and a tape device in addition to the illustrated Disk (hard disk device). This shared resource 7 is a business program 4 in the space 2 of the active system 3.
The connection is changed so that it is used by the business process loaded and started, and when a failure occurs, the business process 4 that loads and starts the business program 4 in the standby space 2 is automatically used (see S7 in FIG. 2). ).

FIG. 1B shows an example of the configuration information table. As shown in the figure, the configuration information table 6 is a table in which items corresponding to the class, the business program, the active processing device, the standby processing device, and the shared resource record are set in advance. Here, the class is an attribute given to each record in association with the operating system 1 of the operating system, and in the case of the operating system of class 0, the A business program loaded in the independent space of the processing system # 1 processes. Do class 0
In the case of the standby system, the A business program loaded in the independent space of the processing device # 3 performs the processing.

The business (business program) is a program loaded by the active / standby processor 1 into the independent space. The operational processing device is an identifier (address) of a processing device that loads the business program 4 into the independent space and performs business processing in a normal state.

The standby system processing device is the active system processing device 1.
Identifier (address) of the processing device that loads the business program 4 into the independent space and performs business processing when a failure is detected
It is.

The shared resource 6 is a shared resource 6 used when the active processing unit 1 or the standby processing unit 1 performs processing. Next, the operation of the configuration of FIG. 1 will be described in detail according to the order shown in the flowchart of FIG.

In FIG. 2, S1 sets configuration information. For this, the administrator sets the configuration information table 6 shown in FIG. 1B, for example. In S2, the business program is loaded based on the configuration information. For example, the processor 0 of the operating system of class 0 in the configuration information table 6 of FIG.
The following information classes shown in the figure are: Business Operation System Processing Device Standby System Processing Device Shared Resource 0 A Business # 0 # 3 Disk0 is taken out, and as shown in FIG. 1 (a), the space 2 created by the processing device # 0. The A resource program is loaded and activated, and the shared resource Disk0 is connected, and initialization processing is performed so that the processing of the A resource can be executed. Similarly, each of the active processing devices 1 set in the configuration information table 6 of FIG. 1B performs initialization processing.

At S3, the operation is started. In this case, the processing devices of each operation system initialized in S2, for example, the processing devices # 0, # 1, and # 2 of FIG. In S4, the arrival of the alive message from another node is confirmed in a fixed time period. This means that an alive message from another node (another processing device of FIG. 1A) (the processing device of the active system transmits to all the other processing devices at regular intervals and is in normal operation). Message to inform)
Confirm that the arrival of is received in a certain period of time.

In step S5, it is determined whether an alive message from a specific processing device has arrived. In the case of YES, it is determined that the alive message from the specific processing device 1 does not arrive (receive) even after a certain time has passed.
Proceed to 6. On the other hand, if NO, S4 is repeated.

In S6, if YES in S5, the specific processing device 1
Since it was found that the alive message from the server did not arrive even after a certain period of time, the configuration information table 6 in FIG.
Refer to to determine whether the failed processing device is the active system.
In the case of YES, it progresses to S7. On the other hand, if NO, the process returns to S4 and is repeated.

In S7, since the processing unit in which the failure has occurred is found to be the active system in YES in S6, the processing unit which backs up the failure processing apparatus has the space required for the operation of the class operating in the failure processing apparatus. A processor that creates an operating system of the class ・ Connects the shared resources belonging to the class to its own node and associates it with the created space ・ Restores the shared resources ・ Space where the business program belonging to the class is created To load. That is, if a failure occurs in the active processing device,
All the processing devices having the same configuration information table 6 in FIG. 1B have detected that a failure has occurred in the active processing device (detected by not receiving an alive message even after a certain period of time). In response to this, in order for the standby processing device 1 to back up the failed operating system processing device, the shared resources belonging to the class of the failed active processing device must Processing unit) and associate it with the newly created independent space.

A shared resource connected to the self-processing device in
For example, there is a high possibility that the data on the magnetic disk drive is partially damaged, so repair the damage when it is found (for example, when backup data is used to repair it, or when data reading and writing is completed). The process is returned to, the incomplete data is erased, and a repair process is performed such as a process for the erased data).

The business program of the operating system in which the failure has occurred is loaded into the independent space created in step 3 to bring it into a workable state. Perform

In step S8, the operation is restarted. By the above,
When a failure occurs in the active processing device 1 by referring to the configuration information table 6 in FIG. 1B, the standby processing device 1 detects the failure and detects the failure of the active processing device 1. After switching the connection of the shared device to the processing device of the standby system and repairing it, it is possible to create an independent space, load the business program of the operation system into the space, restart operation, and continue business processing. Become.

By doing so, (b) of FIG.
As shown in the configuration information table 6 of FIG. 1, only a simple setting operation of setting a business program, a standby system processing device, and a shared resource in association with an active system processing device is automatically performed. When a processor loads a business program into an independent space and starts normal operation, and when a failure occurs in the active processor, the standby processor detects the failure and switches the shared device, and the shared device It was possible to build a system for repairing the above, creating an independent space, loading a business program into the independent space, and automatically restarting the operation.

Next, with reference to FIGS. 3 to 6, the state of backup when the configuration information table 6 in FIG. 1B is specifically described. FIG. 3 shows a specific example (1) of the present invention. According to the configuration information table 6 of FIG. 1B, this is the processing device # that started normal operation when the power is turned on.
The states of 0, # 1, and # 2 are shown. In this state, the state is as follows.

The active processor # 0 loads the A business program into the space and connects the shared resource Disk00 to perform business processing. The active processing device # 1 loads the B business program into the space, connects the shared resource Disk01, and performs the B business processing.

The active processor # 2 loads the C business program into the space, connects the shared resource Disk02, and performs the C business processing. The standby-system processing device # 3 is in a standby state without generating a space during normal operation (a dotted line rectangle represents an ungenerated space in a standby state). FIG. 2) is shown. This shows a state when a failure occurs in the active processing device # 0 in the configuration information table 6 of FIG. 1B. In this state,
It becomes the following state.

The processing device # 0 of the active system is stopped due to a failure. The active processor # 1 loads the B business program into the space, connects the shared resource Disk01, and continues the B business processing.

The active processor # 2 loads the C business program into the space, connects the shared resource Disk02, and continues the C business processing. The processing device # 3 of the standby system starts space A, loads the A business program, switches the shared resource Disk00 to connect and repair, and then starts the A business processing.

FIG. 5 shows a specific example (3) of the present invention. This shows a state in the case where a failure has occurred in the active processing device # 1 in the configuration information table 6 of FIG.
In this state, the state is as follows.

The processing device # 0 of the active system is brought into a stopped state due to a failure. 'The operating system processor # 1 is in a stopped state due to a failure. The active processor # 2 loads the C business program into the space, connects the shared resource Disk02, and continues the C business processing.

The standby processing device # 3 further creates a new space, loads the B job program, switches the shared resource Disk01, connects and restores, and then starts the B job process. As a result, the standby processing device # 3
Is in a state where the A business program is loaded into one space and the shared resource Disk00 is connected to perform business processing, and the B business program is loaded into another space in parallel to connect the shared resource Disk01. Then, the B business process is started.

FIG. 6 shows a specific example (4) of the present invention. This shows a state in which a failure has occurred in the active processing device # 2 in the configuration information table 6 of FIG. 1B.
In this state, the state is as follows.

The processing device # 0 of the active system is stopped due to a failure. 'The operating system processor # 1 is in a stopped state due to a failure. 'The processing device # 2 of the active system is stopped due to a failure.

'''The standby processing device # 3 further creates a new space, loads the C job program, switches the shared resource Disk02, connects and restores it, and then starts the C job process. As a result, the standby processing device # 3
Loads the A business program and the B business program into the two spaces independently and shares the shared resources Disk00
And the shared resource Disk01 are connected to each other to perform the business processing independently, and the C business program is loaded in parallel to another space to share the shared resource Disk0.
2 will be connected and C business processing will be started.

As described above, the configuration information table 6 shown in FIG.
(B) of FIG.
As shown in Fig. 2, when the processing devices # 0, # 1, and # 2 of the operating system operate, respectively, and when a failure occurs in one of them, the standby processing device (here, the processing device # 3) operates in an independent space. Load programs, switch shared resources
It is possible to repair and continue the business processing in which each failure has occurred.

FIG. 7 shows an example of the shared resource table (processing device # 3) of the present invention. This is an example of the shared resource table of the processing device # 3 of FIG. In the state of FIG. 6, the standby processing device # 3 has failed
Shared resources of 0, # 1, and # 2 (see the configuration information table 6 in FIG. 1B) Disk00, Disk01, and Disk
02 are exclusively used, so this relationship is set and stored.

Next, with reference to FIGS. 8 and 9, a system configuration will be described in which only the processing devices 1 in the active system back up each other and the processing devices are effectively used. FIG. 8 shows another embodiment (FIG. 9) of the present invention. This shows the state at the start of operation according to the configuration information table 6 of FIG.

The processing device # 0 is an active processing device of class 0 in the configuration information table of FIG. 9, and loads the A business program into the space and connects the shared resource Disk00 to perform the A business processing. .

The processing device # 1 is an operating system processing device of class 1 in the configuration information table of FIG. 9, and loads the B business program into the space and connects the shared resource Disk01 to perform the B business processing. .

The processor # 2 is an operating processor of class 2 in the configuration information table of FIG. 9, and loads the C business program into the space and connects the shared resource Disk02 to perform the C business processing. .

In this state, for example, when a failure occurs in the processing device # 0, the processing device # 0 is stopped, and the processing device # 1 of the standby system detects the failure occurrence and the standby system of the class 0 indicated by the dotted line. Space, load A business program, switch shared resource Disk00, and restore A
Continue work. Similarly, when a failure occurs in a processing device of another active system, the processing device of another active system set as a standby system continues to work.

Only by making the settings as shown in the configuration information table of FIG. 9 above, under normal conditions, the processing devices # 0, # 1, and # 2 of the operating system respectively operate as shown in FIG. When a failure occurs, one of the standby processing devices and the set operating processing devices # 0, # 1, and # 2 loads a business program in an independent space, switches and repairs shared resources, and a failure occurs. It becomes possible to continue the business processing. In this case, during normal operation, there is no processing unit to wait and it can be used efficiently, and when a failure occurs, another processing unit backs up and processing speed is slightly reduced, but reliability can be improved. Therefore, even if the configuration of the operating system / standby system of such a processing apparatus is as shown in FIG.
As shown in the figure, the configuration information table 6 can be easily set by the administrator without making a distinction from the case where the standby processing device of FIGS. 3 to 6 described above is exclusively provided. Becomes

[0048]

As described above, according to the present invention,
A configuration information table 6 for registering the processing device 1 of the standby system 5 and the shared resource 7 is provided in association with the processing device 1 of the active system 3, and a failure occurs in the processing device 1 of the active system according to the information set in this table. Since the processing system 1 of the standby system 5 sometimes loads the business program 4 into the independent space 2 to continue the business, the active system 3 of the processing system 1 and the standby system 5 are flexibly associated with each other. , It becomes possible to easily set up and realize backup, and improve the reliability of the system.

[Brief description of drawings]

FIG. 1 is a configuration diagram of one embodiment of the present invention.

FIG. 2 is a flowchart illustrating the operation of the present invention.

FIG. 3 is a specific example (1) of the present invention.

FIG. 4 is a specific example (2) of the present invention.

FIG. 5 is a specific example (3) of the present invention.

FIG. 6 is a specific example (4) of the present invention.

FIG. 7 is an example of a shared resource table of the present invention (processing device # 3)
It is.

FIG. 8 is another specific example (FIG. 9) of the present invention.

FIG. 9 is an example of a configuration information table (FIG. 8) of the present invention.

[Explanation of symbols]

 1: Processor 2: Space 3: Operating system 4: Business program 5: Standby system 6: Configuration information table 7: Shared resource

Claims (2)

[Claims] 1. A multiplexing system for multiplexing processing devices, wherein a configuration information table for setting a standby processing device in association with an active processing device is provided, and this configuration information table is set as the active system. When all the processing devices load the business program into the space and execute the processing respectively, and when the processing device set as the standby system detects the failure occurrence of the corresponding active processing device, the business program of the processing device is executed. A multiplexing system characterized by loading in an independent space and continuing work. 2. A processing device of another active system is set as a processing device of the standby system in the configuration information table, and when a failure occurs, a failure occurs in an independent space different from the work in which the standby processing device is operating. 2. The multiplexing system according to claim 1, wherein a plurality of business tasks are executed in parallel by loading the business program of the processing device.