JPH06139087A - Check point restart system - Google Patents

Abstract

PURPOSE:To improve the job execution performance by asynchronously executing the job execution and the writing of the execution environment in a secondary storage device. CONSTITUTION:Check point information to be acquired is written in a magnetic disk 14 through an expansion memory device 12. In this case, the expansion memory device 12 is a nonvolatile memory backed up by the power supply and even when a fault occurs in writing in the magnetic disk device 14, the check point information is held in the memory 12 as it is. At the fault generation, the presence or absence of the effectivity of the check point information of the expansion memory device 12 is judged by referring to the flag of a management information area 3A. The recovery processing using the check point information of the expansion memory 12 is performed if it is effective and the recovery process using the check point information of the magnetic disk 14 if it is invalid.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a checkpoint restart method for restoring the execution environment of a computer system based on the contents of a checkpoint file.

[0002]

2. Description of the Related Art Generally, a computer system is provided with various fault countermeasure functions. Restart processing is often used as one of the failure countermeasure functions. This restart processing is executed when the job being executed or the entire system is stopped due to a failure.

In the restart processing, only a specific job is restarted (job restart) and the entire system is restarted (system restart).
Generally, job restart is often used for the purpose of speeding up the process.

As a job restart, a checkpoint restart processing method is well known. In this checkpoint restart, checkpoints are set in advance at key points of the job, and each time the job processing reaches a checkpoint, its execution environment is recorded as status information in the checkpoint file of the secondary storage device. If the job execution is interrupted due to a failure, the job is restarted from the latest checkpoint.

As described above, the conventional checkpoint restart saves the execution environment of the program being executed as a checkpoint file in the external secondary storage device, and saves it when the execution of the program is interrupted due to a failure. The old environment is restored on the basis of the information of the checkpoint file that has been stored, and re-execution is possible. By using this restart method, high-speed failure recovery can be performed.

However, in such a conventional checkpoint restart method, since the execution of the job and the writing of the execution environment into the secondary storage device are performed in synchronization, the CPU is shown in FIG. As described above, the execution of the job is delayed until the write request for instructing the write to the secondary storage device is issued and the response indicating the write completion is returned.

This is because if the next job is executed without waiting for the completion of writing to the secondary storage device, the execution environment on the main storage will change, so if a failure occurs during writing to the secondary storage device. If you do, the original execution environment will be lost.

Therefore, in the conventional checkpoint restart method, a waiting time for writing the execution environment into the secondary storage device is inserted at each checkpoint, which deteriorates the job execution performance. There was a flaw.

[0009]

Conventionally, since the execution of a job and the writing of the execution environment to the secondary storage device are performed in synchronization with each other, the waiting for writing the execution environment to the secondary storage device is performed at each checkpoint. Since time is inserted, there is a drawback that the job execution performance is degraded.

The present invention has been made in view of the above circumstances, and enables execution of a job and writing of an execution environment in a secondary storage device to be executed asynchronously, thereby improving job execution performance. The purpose is to provide a possible checkpoint restart method.

[0011]

Means and Actions for Solving the Problems
In the checkpoint restart method that restores the execution environment of the computer system based on the contents of the checkpoint file, a secondary storage device in which the checkpoint file is stored and checkpoint information to be written in the checkpoint file are stored. A nonvolatile memory having a buffer area and a management information area in which a flag indicating the validity of the checkpoint information of the buffer area is set;
Input / output means for executing data transfer to the next storage device, and an execution environment of a job being executed on the computer system is sampled for each checkpoint, and the collected information is stored in the buffer area, and the management information is also stored. Means for setting the flag in the area; means for issuing to the input / output means a request for writing the contents of the buffer area of the non-volatile memory in the checkpoint file of the secondary storage; In response to the write completion notification, referring to the means of resetting the flag of the management information area to release the corresponding buffer area of the non-volatile memory, and the flag of the management information area of the non-volatile memory when a failure occurs, Depending on the set / reset state of the flag, the contents of the buffer area of the nonvolatile memory or the secondary memory Characterized by comprising a means for restoring the execution environment of the device checkpoint file computer system using the.

In the checkpoint restart method, the checkpoint information to be collected is written in the secondary storage device via the non-volatile memory. in this case,
Since the memory is non-volatile, even if a failure occurs during writing to the secondary storage device, the checkpoint information is retained in the non-volatile memory as it is. For this reason, when a failure occurs, the flag is used to determine whether the checkpoint information in the non-volatile memory is valid. If it is valid, the restoration process using the checkpoint information in the non-volatile memory is performed. By performing the restoration process using the checkpoint information of the secondary storage device, the checkpoint restart using the non-volatile memory becomes possible. Therefore, it is not necessary to wait for the execution of the job until the writing to the secondary storage device is completed, and the execution of the job and the writing of the execution environment to the secondary storage device can be executed asynchronously. Can be improved.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows the configuration of a computer system according to an embodiment of the present invention. This computer system comprises a computer main body 11, an extended memory device 12, a power supply backup device 13, and a magnetic disk device 14. The computer main body 11 has the same configuration as a normal computer system, and includes a CPU 111 and a main memory 11 which are interconnected via a system bus.
2, I / O channel 113 and the like.

The CPU 111 controls the entire computer system and executes various jobs. Also, C
The PU 111 has a function for realizing checkpoint restart. In the checkpoint restart, checkpoints are set in advance at key points of the job, and the execution environment is recorded as checkpoint information in the checkpoint file of the magnetic disk device 114 each time the job processing reaches the checkpoint. If the job execution is interrupted due to a failure, the job is restarted from the latest checkpoint.

In the process of collecting check points,
The CPU 111 transfers the checkpoint information from the main memory 112 to the extended memory device 12, and then issues a request for writing the checkpoint information from the extended memory device 12 to the magnetic disk device 114. C like this
The checkpoint collection processing by the PU 111 is executed according to the checkpoint management program in the main memory 112.

The main memory 112 is a dynamic RAM
And a checkpoint management program stored therein. Further, the main memory 112 is used as a work area for job execution, and information indicating a job execution environment is set on the main memory 112. In the figure, 2A, 2B, 2C
Indicates a part of the execution environment of the program, and these are collected as checkpoint information.

The I / O channel 113 is an input / output device for executing read / write with respect to the magnetic disk device 14 in response to a request from the CPU 111, and is expanded according to a write request from the CPU 111 in the checkpoint sampling process. The checkpoint information of the memory device 12 is written in the checkpoint file 141 of the magnetic disk device 14. Also, the I / O channel 113
When the writing is completed, the CPU sends a writing completion notification to the CPU.
Hand it over to 111.

The expansion memory device 12 is a volatile memory such as a dynamic RAM which is connected to the computer main body 11 as needed, but here, the power supply backup device 1 is used.
It is configured to be used as a non-volatile memory by a backup power source from 3. The extended memory device 12 is allocated with buffer areas 3B1 and 3B2 in which the collected checkpoint information is temporarily stored, and a management information area 3A in which management information for generation management of checkpoint information is set. .

The magnetic disk device 14 is used as a secondary storage device of this computer system, and a checkpoint file 141 required for a checkpoint restart is accumulated and held therein. The checkpoint file 141 includes a management information area 6A and checkpoint information areas 6B1 and 6B2. In the management information area 6A, management information for performing generation management of checkpoint information is set. Here, the contents of the management information area 3A of the extended memory device 12 are reflected in the management information area 6A at any time. The contents of the buffer areas 3B1 and 3B2 of the extended memory device 12 are written in the checkpoint information areas 6B1 and 6B2. next,
Checkpoint information collection processing will be described with reference to the flowchart in FIG.

Here, a case will be described in which two buffers are used to support rollback processing using the extended memory device 12 and two-generation checkpoint information is secured in the extended memory device 12.

It is assumed that information (generation 1) to be stored as checkpoint information at the checkpoint (program quiescent point 1) of the user program is 2A, 2B, 2C. In this case, the CPU 111 causes the information 2A, 2
A buffer 3B1 for storing B and 2C is secured in the extended memory device 12 (step S11).

Next, the CPU 111 transfers the information 2A, 2B, 2C to be collected to the buffer 3B1 (step S12), and at this point, the information is stored in the management information area 3A in order to establish this checkpoint information. Two
The IDs (identifiers) of A, 2B, and 2C and the extended memory device 1
2 and a valid flag indicating that valid information is set (step S13).

Next, the CPU 111 loads the contents of the buffer 3B1 on the extended memory device 12 into the magnetic disk device 14.
Issue a write request (W1) for writing to the checkpoint file 141 of
The process returns to the execution of the user program (step S14).

When job execution is continued and the next checkpoint arrives, the CPU 111 causes the buffer 3B2 to store the information (generation 2) to be collected at that time.
Is reserved on the extended memory device 12 (step S1).
5). Next, the CPU 111 transfers the information to be collected to the buffer 3B2 (step S16), and at this point, in order to establish this checkpoint information,
In the management information area 3A, an ID (identifier) of the information and a valid flag indicating that it is set on the extended memory device 12 are set (step S17).

Next, the CPU 111 loads the contents of the buffer 3B2 on the expansion memory device 12 into the magnetic disk device 14.
Issue a write request (W2) for writing to the checkpoint file 141 of
The process returns to the execution of the user program (step S18).

After that, the buffer 3B1 becomes the target for use when the next checkpoint arrives, so that a final I / O completion wait is performed before the next checkpoint arrives (step S19). Then, in response to the write completion notification for the write request W1, the buffer 3B1 is released, and the buffer 3B1 is newly secured for the next new checkpoint information (step S2).
0). FIG. 3 shows the processing when the writing is completed. When the write completion notification is issued from the I / O channel 113 in step S19 of FIG. 2, the CPU 111 executes the process of FIG.

For example, in the case of the completion notification for the request W1 for writing the contents of the buffer 3B1 to the disk 14, the CP
U111 is the buffer 3B of the management information area 3A.
The flag corresponding to 1 is reset (step S2
1). Next, the CPU 111 releases the buffer 3B1 for the next checkpoint information (step S2).
2). Next, the restoration process when a failure occurs will be described with reference to the flowchart in FIG. When the computer system is down due to some failure, the CPU 111 executes the next restart process after the computer system is restarted.

That is, the CPU 111 first refers to the management information area 3A of the expansion memory device 12 to find the latest I
The state of the flag corresponding to the checkpoint information having D is checked (step S31). The CPU 111 determines whether the flag is set or reset (step S32), and selects the information used for the restoration process according to the set / reset of the flag.

If the flag is set, the checkpoint information in the extended memory 12 is valid and the checkpoint information has not yet been written to the disk 14. Therefore, when the flag is set, the CPU 111 executes the restoration process using the checkpoint information (buffer 3B1 or 3B2) of the extension memory 12 (step S33).

If the flag is reset, the checkpoint information in the extension memory 12 is invalid and the checkpoint information has already been written in the disk 14. Therefore, if the flag is reset, the CPU 111 checks the checkpoint information (6B1) of the checkpoint file 141 of the magnetic disk device 14.
Alternatively, the restoration process using 6B2) is executed (step S34).

As described above, in the checkpoint restart method of this embodiment, the checkpoint information to be collected is written in the magnetic disk device 14 via the extended memory device 12. In this case, the extended memory device 1
Reference numeral 2 is a non-volatile memory whose power is backed up. Therefore, even if a failure occurs during writing to the magnetic disk device 14, the checkpoint information is retained in the extended memory device 12 as it is. Therefore, when a failure occurs, the flag of the management information area 3A is referred to determine whether the checkpoint information of the extended memory device 12 is valid, and if it is valid, the extended memory device 12 is checked.
By performing the restoration process using the checkpoint information of No. 1 and the restoration process using the checkpoint information of the magnetic disk device 14 when the checkpoint information is invalid, the checkpoint restart using the extended memory device 12 becomes possible.

Therefore, as shown in FIG. 5, it is not necessary to wait for the execution of the job until the writing to the magnetic disk device 14 is completed, so the job execution is asynchronous with the writing of the execution environment to the magnetic disk device 14. The job can be executed, and the job execution performance can be improved.

In this embodiment, the extended memory device 12 is used.
Although the power supply backup device 13 is used to use the above as a non-volatile memory, it is also possible to use a non-volatile memory element such as an EEPROM for the extended memory device 12.

[0035]

As described above in detail, according to the present invention,
The execution of the job and the writing of the execution environment to the secondary storage device can be executed asynchronously, and the job execution performance can be improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a computer system according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating a checkpoint information collection processing operation according to the embodiment.

FIG. 3 is a flowchart illustrating an operation when writing of checkpoint information is completed in the embodiment.

FIG. 4 is a flowchart illustrating a restart processing operation according to the embodiment.

FIG. 5 is a diagram showing how execution of a job in the embodiment is executed asynchronously with a checkpoint information writing operation.

FIG. 6 is a diagram showing how a job is executed in synchronization with a checkpoint information writing operation in a conventional checkpoint processing method.

[Explanation of symbols]

11 ... Computer main body, 12 ... Extended memory device, 13 ... Power supply backup device, 14 ... Magnetic disk device, 111 ...
CPU, 112 ... Main memory, 113 ... I / O channel, 3A ... Management information area, 3B1, 3B2 ... Buffer area, 141 ... Checkpoint file.

Claims (1)

[Claims] 1. A checkpoint restart method for restoring an execution environment of a computer system based on the contents of a checkpoint file, a secondary storage device in which the checkpoint file is stored, and a checkpoint to be written in the checkpoint file. A nonvolatile memory having a buffer area for storing information and a management information area in which a flag indicating the validity of checkpoint information in the buffer area is set, and data transfer from the nonvolatile memory to the secondary storage device. And an input / output means for executing the above, and the execution environment of the job being executed on the computer system is collected for each checkpoint, the collected information is stored in the buffer area, and the flag is set in the management information area. Means and a buffer of the non-volatile memory A means for issuing a request to write the contents of the area to the checkpoint file of the secondary storage device to the input / output means, and resetting the flag of the management information area in response to the write completion notification from the input / output means. Means for releasing the corresponding buffer area of the non-volatile memory, and referring to the flag of the management information area of the non-volatile memory when a failure occurs, the buffer of the non-volatile memory according to the set / reset state of the flag. A checkpoint restart method comprising means for restoring the execution environment of the computer system by utilizing the contents of the area or the checkpoint file of the secondary storage device.