JP2000057116A - Computer complex system operation control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an operation control method which prevents operation control discrepancy after operation change occurrence, is safe and also has high operability. SOLUTION: The configuration of an operation controller 400 is provided with a system state managing part 407, a system sate management table 408 and a system state storage file 409 which stores the content of the table 408. When the part 407 receives a processing request from an operation controlling part 401, it reports the content of a configuration definition data file 406 and the content of the table 408 or the requested content of the table 408 to the part 401 in accordance with the request content and issues a control instruction. When the part 407 receives control completion response from the part 401, it changes the content of the table 408 and updates it according to the response. After that, when it receives a processing request from the part 401, the part 407 makes a report based on the updated content.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multi-computer system operation control system, and more particularly to a system operation control system suitable for grouping and operating and monitoring devices to be operated.

[0002]

2. Description of the Related Art Conventionally, in a complex computer system, an operation control device has been used to efficiently operate a plurality of computer systems. FIG. 1 shows a general configuration example of a compound computer system to which an operation control device is applied. In FIG. 1, a plurality of processing apparatuses 101, 102, 103
Is connected to the route switching device 301 on the one hand, and to the operation control device 400 via the network 501 on the other hand. I / O devices 201, 202, 203, 20
4 is also connected to the route switching device 301 on the one hand, and to the operation control device 400 via the network 502 on the other hand, similarly to the processing device. Route switching device 301
Are the processing devices 101, 102, 103 and the input / output device 20
Connection or disconnection between connection points 1, 202, 203 and 204 (hereinafter referred to as ports) (hereinafter referred to as path switching)
And data can be transferred between the connected ports. A route switching instruction to the route switching device 301 is executed from the operation control device 400 via the network 503.

The operation control device 400 includes a network 50
1 to start, stop, and monitor the operation state of the processing devices 101, 102, and 103; monitor the start and stop of the input / output devices 201, 202, 203, and 204 via the network 502; monitor the operation state; An operation function such as path switching control of the switching device is executed. The console device 410 is connected to the operation control device 400, and allows the operator to give an operation instruction to the operation control device 400, and the processing devices 101, 102, 103, and the input / output device 20
1, 202, 203, and 204 are used to display the connection relationship between the processing device and the input / output device based on the connection status of the path switching device 301.

FIG. 2 is a block diagram showing a configuration of an operation control device 400 according to a conventional operation control method. The operation control device 400 includes an operation control unit 401, a processing device control unit 40
2. I / O device control unit 403, path switching device control unit 40
4, a console device control unit 405, and a configuration data file 405. Processing device control unit 402
Are connected to the processing apparatuses 101 and 1 via the network 501.
02 and 103, determine the content of the data received according to the prescribed protocol, report to the operation control unit 401, and receive the control instruction from the operation control unit 401, and send the control instruction according to the prescribed protocol. To the processing devices 101, 102 and 103. The input / output device control unit 403 transmits, via the network 502, the input / output devices 201, 202, 203, and 204
The content of the data received according to the determined protocol is determined and reported to the operation control unit 401, and a control instruction is received from the operation control unit 401, and the input / output device 201, 20
2, 203, and 204. The route switching device control unit 404 determines the content of the data received from the route switching device 301 via the network 503 according to the determined protocol, and reports the content to the operation control unit 401. It receives a control instruction and sends the control instruction to the path switching device 301 according to a predetermined protocol.

[0005] The console device control unit 405 determines the contents of the data input to the console device 410 and reports the data to the operation control unit 401, receives display data from the operation control unit 401, and transmits the data to the console device 4.
The control for displaying on 10 is performed. The configuration data file 406 stores information necessary for controlling the processing devices 101, 102, and 103, the input / output devices 201, 202, 203, and 204, and the route switching device 301, and information on a typical operation method. , Is referred to by the operation control unit 401 when executing the operation control. As described above, the operation control unit 401 refers to the configuration data file 406 in response to the control instruction from the console device control unit 405, and refers to the processing device control unit 402 and the input / output device control unit 40.
3. While giving a control instruction to the path switching device control unit 404, the processing device control unit 402 and the input / output device control unit 40
3. An operation control device 4 that issues a display instruction to the console device 410 to the console device control portion 405 based on a status report from the route switching device control portion 404.
00 controls the overall processing.

FIG. 3 is a diagram showing an example of the configuration and contents of data stored in the configuration definition data file 406. The configuration definition data includes pattern definition data 406A,
It comprises processing device (CPU) group definition data 406B, input / output device (IO) group definition data 406C, CPU definition data 406D, and IO definition data 406E. The pattern definition data 406A is data defining a predetermined operation method (hereinafter referred to as an operation pattern), and includes a pattern name 406A1, a CPU group name 406A2 which is a target of the operation pattern,
A command 406A3 indicating control content for the group,
IO group name 406A4 that is the target of the operation pattern,
Command 406A indicating control contents for IO group
5, and IO group connection destination CPU group name 406
A. The commands 406A3 and 406A5 are typically START, STOP, and NOP. START means starting the device, STOP means stopping the device, and NOP means not operating.

The CPU group definition data 406B is the CPU group name 406A of the pattern definition data 406A.
2 is data defining the names of the CPUs that make up the CPU group shown in FIG.
1 and a group 406B2. Similarly,
The IO group definition data 406C is an IO group name 406A4 of the pattern definition data 406A.
This is data defining the names of the IOs that make up the group, including the IO group name 406C1 and the group configuration 4
06C2. The CPU definition data 406D is
This is data defining control information of the CPU name indicated by the configuration 406B2 of the U group definition data 406B,
PU name 406D1, CPU control network 501
It is composed of the above address 406D2 and the port number 406D3 of the path control device to which the CPU is connected. Similarly, the IO definition data 406E is data defining control information of the IO name indicated by the configuration 406C2 of the IO group definition data 406C, and includes IO names 406E1, I
Address 406E2 on the O control network 502,
And port number 406 of the path control device to which the IO is connected
E3. The operation control unit 401 performs operation control with reference to the configuration definition data.

FIG. 4 shows an operation control unit 401 based on the contents of the pattern definition (PTN1) of the configuration definition data shown in FIG.
FIG. 4 is a diagram showing an outline of the processing of FIG. When receiving the PTN1 execution instruction from the console device control unit 405, the operation control unit 401 reads the configuration data from the configuration data file 406 in the following procedure to determine the control target and the content.

First, in step S101, the CPU group name 406A2 of the record whose name 406A1 of the pattern definition data 406A is PTN1, and the command 406
A3, IO group name 406A4, and command 406
A, the connection destination 406A6 is read, and control contents (commands) for the CPU group and the IO group are set in the internal area. In the example of PTN1, CPUG1 is STAR
T, CPUG2 is STOP, IOG1 is START and C
Connected to PUG1, IOG2 is STOP and no CPUG connection processing (NOP).

Next (S102), the name 406B of the CPU group definition data 406B is obtained in order to determine the configuration of the CPU group and the IO group designated by PTN1.
1 is the configuration 406B2 of CPUG1, CPUG2, and I
The name 406C1 of the O group definition data 406C is IO
The configuration 406C2 of G1 and IOG2 is read, and the CPU name and IO name to be controlled are set in the internal area. In this example, CPUG1 is CPU1, CPU2, and CPUG1.
2 is CPU2, CPU3, IOG1 is IO1, IO
2. In IO3, IOG2 is IO2, IO3, IO4.

Next (S103), in order to obtain the control address of the target device and the port number of the path control device,
The name 406D1 of the CPU definition data 406D is the CPU
1. The control address 406D2 and port number 406D3 of the CPU2 and CPU3 are set in the internal area with the name 406E1 of the IO definition data 406E as the control address 406E2 and the port number 406E3 of IO1, IO2, IO3 and IO4. In this example, CPU 1 has control address 0
1, port number 1 (expressed as 01: 1), CPU2
Is 02: 2, CPU3 is 03: 3, IO1 is 11:
4, IO2 is 12: 5, IO3 is 13: 6, IO4 is 1
4: 7.

Then, based on the contents obtained from S101, S102 and S103, finally in S104, the CPU,
The command corresponding to the IO control address and the contents of the port connection instruction to the path switching device are obtained. In this example of PTN1, CPU control addresses 01 and 02 are START,
03 is a STOP, IO control addresses 11, 12, and 13 are START, 14 is a STOP, and an instruction to the path switching device connects port numbers 4 and 1 (expressed as 4: 1). 1, 5: 2, 6: 1 and 6: 2 are connected.

Finally (S105), the control command obtained in S104 is transmitted to the processing device control unit 402, the input / output device control unit 403, and the path switching device control unit 404 to instruct control execution of each device. The completion of the command is confirmed from the control unit 402, the input / output device control unit 403, and the path switching device control unit 404, the completion report is sent to the console device control unit 405, and the processing of the operation control unit 401 is completed. I do.

As described above, in the operation of the multifunction computer system according to the conventional operation control method, an instruction is received from the console device, and the multifunction computer system is controlled based on configuration definition data defined in advance. Needless to say, from the console device to the processing device, the input / output device,
Individual control instructions for the path switching device are also possible.

[0015]

The system operation using the operation control device according to the conventional operation control method has the following problems. During system operation, unexpected system configuration changes may be required. For example, when a disk device fails, the disk device connected to another CPU may be temporarily changed and used until the repair is completed, or the temporary print volume may be changed by a printing subsystem or the like. In order to cope with the increase, a printer device of another subsystem may be connected. In such a case,
Generally, the system operation manager responds to a situation by controlling the path switching device with the console device of the operation control device to change the connection between the CPU and the IO. in this case,
The change in the IO connection changes the configuration of the IO group. However, since the configuration definition data in the operation control device is fixed, the next time the operation pattern control is executed, control is performed using the fixed definition data without considering the IO whose connection has been changed. However, there is a problem that incorrect control is performed, which does not match the current operation state.

This problem will be described with reference to the configuration definition data shown in FIG. First, it is assumed that the system is started using the operation pattern PTN1. In this state, the IO group IOG1, ie, IO1, IO2, IO3,
PU group CPUG1, that is, CPU1, CPU2
Connected to Next, IO4 is required, and IO4 is connected to CPU1 and CPU2. Thereafter, it is assumed that the operation pattern PTN2 is executed to change the operation mode.
PTN2 sets IO group IOG2 to CPU group C
Because it is defined to connect to PUG2, IO2, I2
O3 and IO4 are connected to CPU2 and CPU3. As a result, IO2, IO3, and IO4 are connected to all CPUs. Although IO2 and IO3 are as originally planned and have no problem, since IO4 is not assumed to be connected to all CPUs, there is a possibility of incorrect operation.

An object of the present invention is to solve such a conventional problem, to prevent inconsistency in operation control after an operation change occurs, and to provide an operation control method for a multifunction computer system that is safe and operable. .

[0018]

In order to achieve the above object, the present invention relates to a multi-computer system operation control system for grouping and operating and monitoring hardware of a multi-computer system comprising a plurality of computer systems. System state storage means for holding operating state information of each hardware group of a plurality of computer systems, information indicating an interconnecting relationship between the groups, and identification information of hardware constituting each group; Reference to the storage contents of the means, system state management means for managing updates, and operation control means for controlling the operation contents of the plurality of computer systems, the system state management means,
By referring to the contents of the system state storage means according to the instruction from the operation control means, reporting the control target hardware and control contents corresponding to the instruction to the operation control means, and executing the operation control from the operation control means. Is received, and the contents of the system state storage means are updated based on the response information.

The system state storage means may be a fixed control state, which is a state of system operation based on a predetermined fixed control, or a change control state, in which the contents of the predetermined fixed control are changed. Is stored.

[0020] Further, there is provided storage means connected to the system state management means for permanently storing the contents of the system state storage means, and the apparatus provided with the system state storage means is continuously operated even after stopping and restarting. In this way, the operation control of the complex computer system is enabled.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS. FIG. 5 is a block diagram showing a configuration of an embodiment of an operation control device 400 to which the operation control method of the present invention is applied. Operation control device 400
Is, like the conventional operation control device, an operation control unit 401,
A processing device control unit 402, an input / output device control unit 403, a path switching device control unit 404, a console device control unit 405,
And a configuration definition data file 406. In the operation control device to which the operation control method of the present invention is applied, in addition to the above, the system state management unit 407, the system state management table 408, and the system state storage file 409
It is equipped with.

The operation control unit 401 requests the system state management unit 407 to perform a process of determining the control target and the contents with reference to the configuration data file 406 in response to a control instruction from the console device control unit 405. I do. Subsequent operations for the processing device control unit 402, the input / output device control unit 403, and the path switching device control unit 404 are the same as in the related art. The system state management unit 407 includes an operation control unit 401
In response to the request from the server, the processing for determining the control target and the contents with reference to the configuration file 406 and the system state management table 408 is performed. The system state storage file 409 holds the same contents as the system state management table 408, and stores the operation state of the computer system at that time when the operation control device 400 stops alone. When the operation control device 400 is restarted, it is used to restore the previous operation state to the system state management table 408.

FIG. 6 shows the system status management table 408.
FIG. 4 is a diagram showing an example of the structure and contents of data managed by (the same applies to a system state storage file 409). The system state management data includes three elements: a pattern state 408A, a CPU group state 408B, and an IO group state 408C. The pattern state 408A includes an operation pattern name 408A1 that has been most recently executed, and whether or not the contents of the operation pattern are contents defined by the configuration definition data 406, and whether or not the contents have been changed. . If the change presence / absence state data 408A2 is the same as the configuration data, “OR
"IGINAL", otherwise "MODIFIED"
Is stored. The CPU group status 408B indicates that the CPU
The data includes a group name 408B1, data 408B2 indicating the current operation state of the CPU group, and data 408B3 indicating the names of the CPUs forming the current group.
The status data 408B2 includes the command contents “START”, “STOP”, and “NO” specified by the operation pattern.
P "is stored. The IO group status 408C is IO
The data includes a group name 408C1, data 408C2 indicating the current operation state of the IO group, and data 408C3 indicating the name of the IO constituting the current group. The state data 408C2 includes command contents “START”, “STOP”, “NOP”,
And the name of the connection destination CPU group.

Next, with reference to FIG. 7 to FIG. 10, the respective processes of pattern control execution, IO group control execution, and IO group configuration change when the operation control device is started by the system state management unit 407 will be described. FIG. 7 is a diagram illustrating a process of the system state management unit 407 when the operation control device is activated. First, it is checked whether the pattern name 408A1 of the system state storage file 409 is blank (S2).
01). If it is blank, it means that there is no previously saved system state, so the system state management unit 407 performs initialization processing of the system state management table 408 shown in S202 to S205.

The initialization process is performed in the configuration data file 4
From 06, predetermined data is copied to the system status management table 408 by the following procedure. (1) The name 406B1 of the CPU group definition data 406B of the configuration data file 406 is copied to the group name 408B1 of the CPU group status 408B of the system status management table 408 (S202). (2) The configuration 406B2 of the CPU group definition data 406B of the configuration data file 406 is copied to the configuration 408B3 of the CPU group status 408B of the system status management table 408 (S203). (3) The name 406C1 of the IO group definition data 406C of the configuration data file 406 is copied to the group name 408C1 of the IO group status 408C of the system status management table 408 (S204). (4) The configuration 406C2 of the IO group definition data 406C of the configuration data file 406 is copied to the configuration 408C3 of the CPU group status 408C of the system status management table 408 (S205).

If the pattern name 408A1 is not blank, all the data of the system state storage file 409 is copied to the corresponding area of the system state management table 408. As described above, the preparation of the request for the control target and content determination processing from the operation control unit 401 thereafter is completed.

FIG. 8 is a diagram showing the processing of the system state management unit 407 accompanying the execution of the operation pattern control. First,
When a processing request is received from the operation control unit 401, the control command 406A3 of each CPU group name 406A2 of the corresponding pattern name 406A1 (PTN1 in this example) of the pattern definition data 406A of the configuration data file 406.
(Referred to as a CPU group command) and the CPU configuration 408B3 of each CPU group name 408B1 in the CPU group status 408B of the system status management table 408, and report them to the operation control unit 401 (S301).
Next, similarly for the IO group, the control command 406A5 of each IO group name 406A4 of the corresponding pattern name 406A1 (PTN1 in this example) of the pattern definition data 406A of the configuration data file 406, the connection destination CPU group name 406A6 (IO group Command), and I of each IO group name 408C1 in the IO group status 408C of the system status management table 408.
The O configuration 408C3 is read and reported to the operation control unit 401 (S302). Next, for the operation control unit 401,
An operation pattern control is instructed based on the data reported previously, and a completion response from the operation control unit 401 is waited (S30).
3). When a completion response is received from the operation control unit 401 (S
304), the CPU group command reported to the operation control unit 401 is transmitted to the CPU in the system state management table 408.
The state 408B2 of the group state 408B is set in the area 408B2, and at the same time, is written in the same area of the system state storage file 409 (S305). Similarly, the IOG command is set in the state 408C2 area of the IO group state 408C of the system state management table 408, and is also written in the same area of the system state save file 409 (S306). If an incomplete response is received from the operation control unit 401 or the response itself is not received within a predetermined time, the system management information table 4
08, the process is terminated without updating the system state storage file.

FIG. 9 is a diagram showing the processing of the system state management unit 407 accompanying the IO group control.
FIG. 9 is a diagram illustrating a case where an IOG2 is connected to a CPUG1 as an IO group control. First, when a process request is received from the operation control unit 401, the IO configuration 40 of the IO group name 408C1 (IOG2 in this example) of the IO group status 408C of the system status management table 408 is received.
8C3 is read out and reported to the operation control unit 401 (S4
01). Next, an instruction for IO group control is issued to the operation control unit 401, and a completion response from the operation control unit 401 is waited (S402). When a completion response is received from the operation control unit 401 (S403), the IO group control command and the connection destination CPU group name (START, CPUG1 in this example) reported from the operation control unit 401 are transmitted to the IO group in the system state management table 408. State 408 of IOG2 for group name 408C1 in state 408C
In addition to the setting in the C2 area, writing is also performed in the same area of the system state saving file 409 (S404). If an incomplete response has not been received from the operation control unit 401 or the response itself has not been received within a certain period of time, the processing ends without updating the system management information table 408 and the system state storage file.

FIG. 10 is a diagram showing a change in IO group configuration.
FIG. 10 is a diagram illustrating processing of the system state management unit 407, which illustrates IO1 of IOG1 as an IO group configuration change.
FIG. 7 is a diagram showing an example in which is removed from the configuration. First, when a processing request is received from the operation control unit 401, the IO group status 408C of the system status management table 408 is received.
The I / O configuration 408C3 of the I / O group name 408C1 (IOG1 in this example) is read and reported to the operation control unit 401 (S501). Next, an IO group configuration change execution instruction is issued to the operation control unit 401, and a completion response from the operation control unit 401 is waited (S502). Upon receiving a completion response from the operation control unit 401 (S503), the changed IO group configuration (IO2, IO3 in this example) reported from the operation control unit 401 is changed to the configuration of the IO group status 408C in the system status management table 408. 408C
3 areas and a system state save file 4
09 is also written in the same area (S504). If an incomplete response has not been received from the operation control unit 401 or the response itself has not been received within a certain period of time, the processing ends without updating the system management information table 408 and the system state storage file.

FIGS. 11, 12, and 13 show the contents of the system state management table 408 (similarly, the system state storage file 409) after the above-described operation example.

FIG. 11 shows an operation pattern control (PTN1).
This is the state after execution, and the pattern name 408A1 is "PT
N1 "is the change presence / absence state 4 immediately after the execution of the pattern.
08A2 is set to “ORIGINAL”, and the CPU group status 408B2 and the IO group status 408C2 are set to the contents defined by the pattern definition data 406A of the configuration file 406.

FIG. 12 shows the state of the system state management table 408 after the execution of the IO group control (connecting the IOG 2 to the CPU G 1), which is set to the contents changed by the change of the IO group control. That is, the pattern name 40
8A1 does not change in “PTN1” and the change presence / absence state 408
Since A2 has been changed from the pattern contents, "MODI
FIED "and the CPU group status 408B2
Is changed from STOP to START, and the state of CPUG2 in the IO group state 408C2 is changed from STOP, NOP to START, CPUG1.

FIG. 13 shows the state of the system state management table 408 shown in FIG.
Is removed from the configuration), and the status of the system status management table 408 after the execution is set to the content changed by the IO group configuration change. That is, the pattern name 408A1 does not change with "PTN1", and the change presence / absence state 408A2 does not change with "MODIFIED" because the pattern content has already been changed.
The IO group configuration of the IOG1 of the O group configuration 408C3 is (IO1, IO2, IO3) to (IO2, IO2).
Changed to 3). As described above, one embodiment of the present invention has been described, but the present invention is not limited to this, and it goes without saying that various modifications can be made without departing from the scope of the claims.

[0034]

As is apparent from the above description, according to the present invention, even if the IO group control or the IO group configuration change is performed after the execution of the operation pattern control, all the control contents and states are managed. Then, in any operation including the operation pattern operation, the control reflecting the current system state can be executed, and the state does not become invalid. In addition, by providing a system state storage file, the operation control device turns off the power by itself,
When the power is turned on again, control can be restarted from the previous state.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a general configuration example in which an operation control device is applied to a complex computer system.

FIG. 2 is a block diagram illustrating a configuration of an operation control device according to a conventional operation control method.

FIG. 3 is a diagram showing an example of the configuration and contents of configuration definition data.

FIG. 4 is a diagram showing a flow of processing of an operation control unit of a conventional operation control device.

FIG. 5 is a block diagram showing an embodiment of a configuration of an operation control device to which the operation control method of the present invention is applied.

FIG. 6 is a diagram showing an example of the configuration and contents of a system state management table and a system state storage file.

FIG. 7 is a diagram showing a flow of a process at the time of starting the operation control device of the system state management unit of FIG. 5;

FIG. 8 is a diagram showing a flow of a pattern control process of a system state management unit in FIG. 5;

FIG. 9 is a diagram showing a flow of an IO group control process of the system state management unit of FIG. 5;

FIG. 10 is a diagram showing a flow of an IO group configuration change process of the system state management unit of FIG. 5;

11 is a diagram illustrating an example of the contents of a system state management table and a system state storage file after executing the pattern control process of FIG. 8;

12 is a diagram illustrating an example of the contents of a system state management table and a system state storage file after the execution of the IO group control processing of FIG. 9;

13 is a diagram illustrating an example of contents of a system state management table and a system state storage file after execution of the IO group configuration change processing of FIG. 10;

[Explanation of symbols]

 101, 102, 103 Processing device 201, 202, 203, 204 Input / output device 301 Route switching device 400 Operation control device 401 Operation control unit 402 Processing device control unit 403 Input / output device control unit 404 Route switching device control unit 405 Console control unit 406 Configuration data file 407 System status management unit 408 System status management table 409 System status storage file 410 Console device 501, 502, 503 Network

Claims (3)

[Claims] 1. A multi-computer system operation control method for grouping and operating and monitoring hardware of a multi-computer system comprising a plurality of computer systems, comprising: operating state information of each hardware group of the plurality of computer systems; System status storage means for holding information indicating an interconnection relationship between the groups, and identification information of hardware constituting each group; and system status management means for managing reference and updating of storage contents of the system status storage means. And operation control means for controlling operation contents of the plurality of computer systems, wherein the system state management means refers to the contents of the system state storage means in response to an instruction from the operation control means and responds to the instruction. The control target hardware to be controlled and the contents of the control are reported to the operation control means. Receiving response information after completion of the execution of the control, the composite computer system operation control method and updates the contents of the system state storage means based on the response information. 2. The multi-computer system operation control method according to claim 1, wherein said system state storage means is in a fixed control state which is a state of system operation by a predetermined fixed control, or is a predetermined fixed form. A multi-computer system operation control method characterized by holding change status information indicating whether or not the content of the control is a change control state in which the control has been changed. 3. The multi-computer system operation control system according to claim 1, further comprising: a storage unit connected to the system state management unit, for permanently storing contents of the system state storage unit.
A multi-computer system operation control method characterized in that the multi-computer system operation control can be continuously performed even after the device provided with the system state storage means is stopped and restarted.