JP2004258964A - Automatic operation method of computer system, and computer system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a computer system which reduces rerun overhead of job when the computer system stops by occurring trouble of the system. <P>SOLUTION: The computer system provided with a computing server 101, a control unit 103 and an environment monitoring unit 104 which monitors operation environment of the computing server 101 and the control unit 103 acquires information about a job executing on the computing server 101 when the control unit 103 receives abnormal information report, decides the priority of the job based on the acquired information, saves continuation executing information in relation to the job from the job of the decided higher priority in order to a storage unit 102, and begins stop processing of the computer system 11 when a grace time of beginning of system stop processing of the computer system 11 in relation to the abnormal condition of the operation environment set beforehand is expired. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an automatic operation method and a computer system of a computer system, and more particularly to a computer system that automatically stops a computer system due to a failure occurring in the computer system or an abnormality in an installation environment of the computer system. And a technique for improving the continuity of the processing.
[0002]
[Prior art]
In the operation of the computer system, if the installation environment does not satisfy the recommended conditions, there is a risk that the operation will continue and cause serious damage to the computer system.
[0003]
For example, if the room temperature of the room where the computer system is installed becomes abnormally high, the components that make up the computer system may fail due to heat. If the data cannot be supplied, a data storage device such as a disk device may malfunction and data may be damaged.
[0004]
In the event of a system failure such as an operating system panic in addition to the environmental abnormalities described above, it is necessary to stop the system immediately and remove the abnormal state in order to minimize damage to the system. is there.
[0005]
In order to cope with such various failures in the computer system, a method of automatically stopping the operation of safely stopping the computer system as shown in FIG. 11 has conventionally been adopted.
[0006]
This is because, in addition to the calculation server 101 and the control device 103 for managing the same, a system called an environment monitor device 104 that detects an abnormality in the environment where the computer system is installed is used to automatically stop the computer system. Processing is realized.
[0007]
For example, in the event of a power failure, a power failure is automatically detected by an uninterruptible power supply (UPS), this is notified to the control device 103 through the environment monitoring device 104, and the system is automatically activated using an auxiliary power supply such as a storage battery. Is being stopped.
[0008]
The same applies to abnormalities in the installation environment such as room temperature. A temperature sensor or the like is provided in the room to constantly monitor the room temperature, and when the environment monitoring device 104 detects that the room temperature has reached an abnormal room temperature, the system is automatically activated. Stopping, etc. are taken.
[0009]
Known examples of such a technique for automatically stopping a computer system due to a failure occurring in the computer system or an installation environment of the computer include a computer system described in Patent Document 1 and an electronic computer system described in Patent Document 2 And so on.
[0010]
On the other hand, when the computer system is stopped, the job that has been executed on the computer system is usually stopped, and the job needs to be re-executed.
[0011]
On the other hand, the status of the computer system, including the contents of the memory area used by the currently executed jobs and processes, is stored in a non-volatile storage device such as a disk device. The job is saved (e.g., job freeze) by saving the job, and even if the job is stopped due to a system stop, the status inside the computer system saved by the job freeze is saved. A computer system that has a function of reading a file from a storage device to continuously execute a job and reduce overhead due to a system stop, a so-called job freeze / restart function, has been proposed.
[0012]
Also, a computer system having a checkpoint recording function of periodically recording the internal state of the computer system as described above and performing continuous execution of a job based on the last recorded internal state when a failure occurs has been proposed. ing.
[0013]
Conventionally, with these functions, the computer system is stopped, and after the computer system is restarted, the internal state of the computer system is restored, and the job is continuously executed from the state before the computer system was stopped. As a result, the overhead caused by re-execution of a job with respect to a periodic shutdown of the computer system has been reduced, and the computer system has been operated efficiently.
[0014]
[Patent Document 1]
JP-A-10-240390
[0015]
[Patent Document 2]
JP-A-5-152942
[0016]
[Problems to be solved by the invention]
For a sudden system shutdown due to a failure occurring inside or outside the computer system, the occurrence of the failure cannot be foreseen, so it is important to quickly restore the system after the failure has occurred.
[0017]
However, conventionally, in the method using the job freeze function, when a failure occurs, all the running jobs are job-frozen, so that there is a problem that the job freeze takes time.
[0018]
For example, in order to perform a job freeze for all jobs at the time of a power failure, it is necessary to prepare an uninterruptible power supply having a large-capacity storage battery corresponding to the time. Therefore, it is difficult to freeze a job for all jobs in advance, and it is necessary to re-execute the job after the system is restored.
[0019]
Further, in the method using the checkpoint recording function, since the internal state is periodically recorded, a process of periodically recording the internal state is added at the time of executing the job, so that the execution time of the job is lengthened. There was a problem.
[0020]
SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and without causing an increase in hardware cost of a computer system or a prolonged job execution time, a sudden failure or alarm inside or outside the system can be achieved. It is therefore an object of the present invention to provide a computer system capable of reducing the overhead due to re-execution of a job at the time of system recovery as much as possible.
[0021]
The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.
[0022]
[Means for Solving the Problems]
The following is a brief description of an outline of typical inventions disclosed in the present application.
[0023]
An automatic operation method of a computer system according to the present invention is an automatic operation method for performing a stop process and a start process of a computer system, wherein the operation environment of the computer system is monitored, and when an abnormality occurs in the operation environment, the computer system is executed on the computer system. Information about the current job, determine the priority of the job based on the obtained information, save the continuous execution information for the job in the non-volatile storage device in order from the job having the highest determined priority, and set the job in advance. The computer system is stopped when the system stop processing start delay time of the computer system for the abnormality of the operating environment is passed.
[0024]
Further, the computer system according to the present invention monitors at least one calculation server that executes a job, a control device that manages the operation of the calculation server, and monitors the operation environment of the calculation server and the control device. Then, in the computer system including the environment monitoring device that reports the abnormality information to the control device, the control device receives information about the job being executed on the calculation server when the environment monitoring device reports the abnormality information. The priority of the job is determined based on the acquired information, the continuous execution information for the job is saved in the nonvolatile storage device in the order of the determined priority job, and the operation environment of the preset operating environment is saved. When the system stop processing start grace time for the abnormality has elapsed, the computer system starts the stop processing.
[0025]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In all the drawings for describing the embodiments, the same members are denoted by the same reference numerals, and a repeated description thereof will be omitted.
[0026]
(Embodiment 1)
FIG. 1 is a configuration diagram showing a configuration of a computer system according to Embodiment 1 of the present invention.
[0027]
In the figure, a computer system 11 includes one or more calculation servers 101, a storage device 102, a control device 103 for managing these calculation servers, and an air temperature for monitoring the room temperature and humidity of a room where the system is installed. A sensor device for the external environment, such as a power / humidity sensor 105 and an uninterruptible power supply (UPS) 106 for monitoring a power failure such as a power failure, and an external environment condition such as a room temperature and a power supply state received from these external environment sensor devices. And an environment monitoring device 104 for notifying the control device of the content of the abnormality as an alarm when an abnormality occurs.
[0028]
The control device 103 is connected to a storage device 114 that holds an operating system, various software, setting information, and the like necessary for the operation of the control device.
[0029]
In the control device 103, server management software 111, which is software for performing operation management of the entire computer system, is running, and a job management table for managing jobs currently being executed in the calculation server 101. In addition to 112, it has a failure / alarm operation setting table 113 which holds a computer system stop processing start grace time allowed until the system is stopped when a failure / alarm occurs.
[0030]
The job management table 112 manages information relating to the currently executing job, such as the user who executed the job and the execution start time. In addition, the failure / alarm operation setting table 113 includes a grace period until the start of the system stop processing with respect to the assumed failure / alarm, that is, the system stop time after the occurrence of the failure / alarm. The processing start grace time is indicated.
[0031]
Here, an example of the contents of the job management table 112 and the failure / alarm operation setting table 113 will be described.
[0032]
FIG. 2 is a diagram showing an example of the contents of the job management table 112, and FIG. 3 is a diagram showing an example of the contents of the failure / alarm operation setting table 113.
[0033]
The job management table 112 shown in FIG. 2 records information on a currently executing job. When the job starts to be executed, the job name 1121, the executed computer user name 1122, the priority 1123, the execution start time 1124 is recorded, and when the job is completed, the information of the job is deleted. The priority 1123 is added by, for example, being set by a computer user at the start of job execution, becomes priority information, and is used as a factor for determining the priority when a job freeze is performed.
[0034]
In addition, the failure / alarm operation setting table 113 shown in FIG. 3 includes a failure factor 1131 that is the content of the failure / alarm to be assumed, and a grace time allowed for the job freeze process for each failure / alarm. The system stop processing start grace time 1132 is defined. These pieces of information are determined from the specifications of the server, storage, UPS, and the like used in the system when the system is introduced.
[0035]
For example, as the grace period for starting the system stop processing at the time of power failure, the time during which the storage battery can be used as a backup power supply at the time of power failure may be set based on the capacity of the storage battery built into the UPS and the scale of the system.
[0036]
Next, the operation of this embodiment will be described.
[0037]
FIG. 4 is a processing flow at the time of occurrence of a failure / alarm in the computer system of this embodiment, and shows a processing flow of the computer system, particularly server management software, from the occurrence of the failure / alarm until the system stops. I have.
[0038]
FIG. 4 illustrates an example in which an indoor temperature abnormality occurs in a room where the computer system 11 is installed. However, even when another failure or alarm such as a power failure is issued, the system performs the same processing flow. Is performed.
[0039]
First, when the indoor temperature abnormality is detected by the environment monitor device 104 (S201), this information is notified as an alarm to the server management software 111 running on the control device 103.
[0040]
Next, when the server management software 111 receives the alarm of the room temperature abnormality from the environment monitor device 104, first, the time until the system stop processing is started for the alarm is started from the failure / alarm operation setting table 113 (FIG. The system stop processing start grace time is acquired (S202).
[0041]
Next, with respect to the job currently being executed by the calculation server, information necessary for determining the priority of the job saved by the job freeze is acquired (S203).
[0042]
Next, the priority of the job for which the job is to be frozen is determined based on the information obtained in S203, and the calculation server 101 is instructed to start the job freeze in accordance with the determined priority (S204).
[0043]
FIG. 4 shows a case where the priorities are determined in the order of job1, job2, and job3. In this case, the calculation server 101 stores the jobs in order from job1 in accordance with a job freeze start instruction from the server management software 111. The apparatus 102 or the like starts a job freeze in which information that can be re-executed after the computer system is restarted (continuous execution information) is saved.
[0044]
First, when the job freeze is completed for job1, the calculation server 101 notifies the server management software 111 that the freeze processing of job1 has been completed, and the server management software 111 notifies the server management software 111 of the completion of the freeze processing of job1. When received, the fact that job1 is frozen is recorded in the storage device 114 connected to the control device 103 as the frozen job table 115 (S205).
[0045]
Hereinafter, the job freeze processing proceeds on the calculation server 101 in the order of job2 and job3. When the system stop processing start grace time previously acquired in S202 elapses, an instruction to stop the currently executed freeze processing is issued to the calculation server 101. 101, and starts the stop processing of the computer system 11 (S206).
[0046]
In the example of FIG. 4, during the execution of the freeze processing of job3, the time for starting the system stop processing has been reached, so the freeze processing of job3 is stopped.
[0047]
As described above, in this embodiment, a job freeze is automatically performed for a job with a high priority when the computer system 11 is stopped, and when the computer system 11 is restored, these jobs are in a state when the computer system 11 is stopped. Can be executed continuously.
[0048]
When the computer system 11 is actually restored, the server management software 111 on the control device 103 refers to the list of frozen jobs recorded in the storage device 114, and On the other hand, by performing the restart process in the calculation server 101, the continuous execution may be started using the information saved in the storage device 102.
[0049]
Next, an example of a method for determining the priority of a job when executing the job freeze according to this embodiment will be described.
[0050]
FIG. 5 shows an example of a method for determining the priority of a job when executing a job freeze. FIG. 5 corresponds to the information acquisition processing of the running job in S203, the determination of the priority in S204, and the freeze processing start processing in the processing flow of FIG.
[0051]
First, in the job information acquisition process S203, the execution time and the used memory amount are inquired to the calculation server 101 for the currently working job (S2031), and the job, the execution time and the memory amount information are acquired from the calculation server 101. (S2032).
[0052]
Next, in the priority determination and freeze processing start processing S204, first, the processing time required to freeze the job is calculated from the memory capacity of each job acquired in S2032 (S2041).
[0053]
The job freeze is mainly a process of recording the information of the memory used by the job in the storage device 102 as it is, and the transfer speed from the calculation server 101 to the storage 102, the writing speed of the storage 102, and the like. The processing time can be calculated from the amount of memory.
[0054]
If this job freeze time is longer than the time up to the start of the system stop processing, there is no possibility that the job will be terminated even if the job freeze is performed. It is deleted from the freeze target (S2042).
[0055]
Next, for the remaining jobs, a job freeze start instruction is given to the calculation server 101 in order from the job with the longest execution time (S2043).
[0056]
According to this priority determination method, jobs that have already been executed for a long time and that can be job-frozen are preferentially job-frozen, so that the overhead of job execution time due to system stoppage can be reduced as a whole.
[0057]
Next, another example of a method for determining the priority of a job when executing the job freeze according to this embodiment will be described.
[0058]
FIG. 6 shows another example of a method for determining the priority of a job when executing a job freeze. FIG. 6 corresponds to the information acquisition processing of the running job in S203, the determination of the priority in S204, and the freeze processing start processing in the processing flow of FIG.
[0059]
In the example of FIG. 6, priority information is set for the job, and the priority information added to the job is used as information for determining the priority of the job.
[0060]
First, in the job information acquisition process S203, since the priority information is managed for each job in the job management table 112 as shown in FIG. 2, for example, the server management software 111 The priority is acquired together with the job (S2033).
[0061]
Next, in the priority determination and freeze processing start processing S204, a job freeze start instruction is issued to the calculation server 101 in order from the job with the highest priority acquired in S2033 (S2044).
[0062]
By this priority determination method, for example, for the execution of a core business processing program of a company, by appropriately setting the priority according to the importance of the processing, even if a failure occurs, the processing overhead due to this is reduced. be able to.
[0063]
The example of the method of determining the priority of the job freeze has been described with reference to FIGS. 5 and 6. However, the present invention is not limited to this method of determining the priority, and other methods of determining the priority of the job may be used. Alternatively, the priority may be determined using various information such as information of various tables in the control device 103 and information from the calculation server 101 depending on the operation mode of the computer.
[0064]
Further, in this embodiment, when a job being executed in a job freeze is saved, information that can be re-executed after the restart of the computer system is saved in the storage device 102 connected to the calculation server 101. Alternatively, the information related to the job may be saved in another nonvolatile storage device other than the storage device 102. In this case, when the computer system is restored, the restart process may be performed by the computer server 101 based on the information in the nonvolatile storage device from which the job has been saved, and the job may be continuously executed.
[0065]
(Embodiment 2)
In this embodiment, the operation of automatic operation when a failure or an alarm occurs in the computer system in the first embodiment is described. In particular, the execution of a job on a computer by using a company as a customer such as an application service provider (ASP) is executed. And provide it as a new service to businesses that act on behalf of.
[0066]
Hereinafter, a new business using the automatic driving method of the computer system of the present invention will be described.
[0067]
FIG. 7 is a diagram showing an outline of an application service provider (ASP) business according to the second embodiment of the present invention.
[0068]
In the figure, an ASP company 1 has a computer system 11 similar to that in the first embodiment, and provides a customer 2 with the use of a computer by acting for a job or by remote access via the Internet or the like. ing.
[0069]
The ASP company 1 gives the use customer 2 permission to use the priority that can be added to the job when the customer 2 executes the job in the computer system 11, and sets the priority usage fee set according to the priority level. Collected from customer 2 The priority is a parameter that determines the priority of a job to be subjected to a job freeze when a failure / alarm occurs in the computer system 11.
[0070]
Therefore, obtaining the right to add a high priority to the job by the user 2 leads to obtaining the right to reduce the overhead due to the occurrence of the failure when the failure occurs in the computer system 11 when the failure occurs in the computer system 11.
[0071]
Next, the relationship between the ASP 1 and the customer 2 in this embodiment will be described.
[0072]
FIG. 8 is a diagram illustrating a relationship between an ASP operator and a user according to the second embodiment.
[0073]
First, the phase of the contract phase 3 in which a contract regarding the use of a computer system is made between the ASP 1 and the customer 2 will be described.
[0074]
The user 2 applies to the ASP 1 for use of the computer system (S21). At this time, as a contract condition, in addition to the method of use, an application for use of an available priority that leads to the priority of job freeze is also made.
[0075]
When the ASP company 1 receives the application for use in S21, it notifies the user 2 of the use permission of the computer including the use permission of the priority, and charges the use fee and the contract fee (S22). On the other hand, the customer 2 makes a contract by paying the usage fee to the ASP company 1 (S23).
[0076]
Next, the stage of normal operation 4, which is the actual operation, will be described.
[0077]
First, the customer 2 requests the ASP 1 to execute a job, or directly executes remote access to the computer system 11 to execute a job (S24). At this time, it is possible to execute the program by adding the priority permitted at the time of contract.
[0078]
In response to the job execution request from the customer 2 in S24, the ASP company 1 executes the job, and returns the job execution result to the customer 2 (S25).
[0079]
As the contents of the execution result, the result is notified if the processing ends normally, and if the execution fails, the fact is notified. The case where the execution is failed includes the case where the customer 2 tries to use a priority for which use permission has not been obtained in advance.
[0080]
Next, a description will be given of the relationship between the ASP and the customer at the stage of the failure operation 5 in which a failure occurs in the computer system during operation.
[0081]
First, when the customer 2 requests the ASP company 1 to execute a job (S26), and after the ASP company 1 starts executing the job, a failure occurs in the computer system 11 (S27). Depending on the added priority, the usage status of the computer system 11, and the content of the failure, it is automatically determined in the computer system whether the job is to be continuously executed or the job cannot be continuously executed and the execution is stopped.
[0082]
If it is determined that continuous execution is to be performed, the job is evacuated to the storage device 102 or the like due to a job freeze, continued execution is performed by restart processing after the restoration of the computer system, and if it is determined that execution is stopped, The execution of the job is stopped, and the job is re-executed after the computer system is restored (S28). Thereafter, when the job is completed, the execution result is reported to the user (S29).
[0083]
Next, a system configuration for realizing the business mode of this embodiment will be described.
[0084]
FIG. 9 is a system configuration diagram for realizing the business mode of the second embodiment.
[0085]
In the figure, a customer terminal 14 used by a customer 2 is connected to a computer system 11 managed by an ASP operator 1 through an ASP gateway server 12 and the Internet 13.
[0086]
Further, the control device 103 includes a job management table 112, a failure / alarm operation setting table 113, and a user setting table 116 for managing the priority permitted to the customer 2 who uses the computer system. In addition, a management terminal 15 is connected for the ASP operator 1 to set the user setting table 116.
[0087]
In response to a priority use request from the customer 2, the ASP operator 1 sets the priority permission by editing the user setting table 116 through the management terminal 15.
[0088]
A job execution request is remotely sent from the customer terminal 14 to the computer system 11 through the Internet 13 and the gateway server 12. The execution result is also notified from the computer system 11 to the customer terminal 14 through the gateway server 12 and the Internet 13.
[0089]
Next, the processing operation of a job with a priority according to this embodiment will be described.
[0090]
FIG. 10 is an operation flow of the computer system at the time of a job execution request by a user according to the second embodiment. In the computer system when the user submits a job with priority to the computer system, especially the server management software. The operation is shown.
[0091]
First, the customer 2 issues a request to the computer system 11 to execute a job to which a priority has been added (S31).
[0092]
When the execution request in S31 is accepted by the server management software 111, first, the user setting table 116 is referred to (S32), and it is determined whether the customer 2 has permission to use the priority added to the job. Is performed (S33).
[0093]
When it is determined that the use is permitted in S33, an execution start instruction is issued to the calculation server 101 (S34). If it is determined in S33 that the use is not permitted, the client 2 is notified that the job is not executable (S35).
[0094]
As described above, in this embodiment, the priority of use permission is given to the customer 2 of the computer system 11, and the service of reducing the overhead of job execution when a failure occurs in response to the request of the customer 2 is provided. Can be provided.
[0095]
As described above, the invention made by the inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and can be variously modified without departing from the gist thereof. Needless to say.
[0096]
【The invention's effect】
The effects obtained by typical aspects of the invention disclosed in the present application will be briefly described as follows.
[0097]
(1) When there is a report of abnormal information from the environment monitoring device, the control device obtains information on the job being executed on the calculation server, and determines the priority of the job based on the obtained information. The job execution process is saved in the non-volatile storage device in the order from the job having the highest determined priority to the non-volatile storage device. Is started, when a failure occurs in the computer system and the system must be stopped, the job can be efficiently saved, and the overhead of re-executing the job can be reduced. This has the effect.
[0098]
(2) Since checkpoint recording is not performed periodically, and the overhead of a job for a sudden failure is reduced, there is an effect that the job execution speed during normal operation is not impaired.
[0099]
(3) By applying the present invention to the ASP business, it is possible to realize a new business form in which the right to raise the priority of job retraction when a failure occurs is compensated.
[Brief description of the drawings]
FIG. 1 is a configuration diagram showing a configuration of a computer system according to a first embodiment of the present invention.
FIG. 2 is a diagram showing an example of the contents of a job management table according to the first embodiment of the present invention.
FIG. 3 is a diagram showing an example of the contents of a failure / alarm operation definition table according to Embodiment 1 of the present invention;
FIG. 4 is a processing flow when a failure / alarm occurs in the computer system according to the first embodiment of the present invention.
FIG. 5 illustrates an example of a method of determining a priority for a job when executing a job freeze according to the first embodiment of the present invention.
FIG. 6 shows another example of a method for determining a priority for a job when executing a job freeze according to the first embodiment of the present invention.
FIG. 7 is a diagram showing an outline of an application service provider (ASP) business according to the second embodiment of the present invention.
FIG. 8 is a diagram showing a relationship between an ASP operator and a customer in Embodiment 2 of the present invention.
FIG. 9 is a system configuration diagram for realizing a business mode according to a second embodiment of the present invention.
FIG. 10 is an operation flow of the computer system when a user executes a job execution request according to the second embodiment of the present invention.
FIG. 11 is a processing flow when a failure / alarm occurs in a conventional computer system.
[Explanation of symbols]
1. Application service provider (ASP) operators
2 Customers
11 Computer system
12 Gateway server
13 Internet
14 Customer terminal
15 Management terminal
101 Calculation server
102 Storage device
103 control device
104 Environmental monitoring device
105 Temperature / humidity sensor
106 Uninterruptible Power Supply (UPS)
111 Server management software
112 Job management table
113 Fault / alarm operation setting table
114 Storage
115 Freeze processed job table
116 User setting table
1121 Job name
1122 User name
1123 Priority
1124 Job start time
1131 Failure factor
1132 System stop processing start grace time

Claims (10)

In an automatic operation method for performing a stop process and a start process of a computer system,
The operating environment of the computer system is monitored, and when an abnormality occurs in the operating environment, information on a job executed on the computer system is obtained, and the priority of the job is determined based on the obtained information. Then, the continuous execution information for the job is saved in the non-volatile storage device in the order of the determined jobs having the highest priority, and when the system stop processing start grace time for the computer system with respect to the abnormality of the operating environment that has been set in advance elapses, the computer An automatic operation method of a computer system, wherein the computer system is stopped. The method for automatically operating a computer system according to claim 1,
The priority of the job is determined based on the elapsed time of execution of the job and the amount of memory used in the computer system, and the elapsed time of execution of the job is long. An automatic operation method for a computer system, wherein a priority of a job to be completed before a processing start delay time elapses is determined to be high. The automatic operation method for a computer system according to claim 1,
The computer according to claim 1, wherein the priority of the job is determined based on priority information added in advance when the job is executed, and the priority of the job with the higher priority specified by the priority information is determined to be higher. Automatic operation of the system. At least one calculation server for executing the job,
A control device that manages the operation of the calculation server;
In a computer system comprising: an operation environment of the calculation server and the control device; and when an abnormality occurs in the operation environment, the environment monitor device reports the abnormality information to the control device.
The control device, when there is a report of abnormal information from the environmental monitoring device, acquires information on a job being executed on the calculation server, and determines the priority of the job based on the acquired information. In the order from the job having the highest priority determined, the continuous execution information for the job is saved in the non-volatile storage device, and when the system stop processing start grace time of the computer system for the abnormality of the operating environment which has been set in advance elapses, the computer system Computer system for starting a stop process of a computer. At least one calculation server for executing the job,
A control device that manages the operation of the calculation server;
In a computer system comprising: an operation environment of the calculation server and the control device; and when an abnormality occurs in the operation environment, the environment monitor device reports the abnormality information to the control device.
The control device is configured to start a system stop process of the computer system with respect to each of the first table for managing the job being executed in the calculation server and the abnormality information reported from the environment monitoring device. A second table storing an allowed system stop processing start delay time, and when there is a report of abnormal information from the environment monitoring device, information on a job being executed on the calculation server is stored in the first table. And the priority of the job is determined based on the acquired information, and the continuous execution information for the job is saved in the nonvolatile storage device in order from the job having the determined high priority. From the table, a system stop processing start grace time of the computer system with respect to the abnormality of the operation environment is obtained, and the obtained system stop Computer system, characterized in that to start with the expiration of physical start window time stopping process of the computer system. The computer system according to claim 4 or 5,
The control device determines the priority of the job based on the elapsed time of execution of the job and the amount of memory used in the computer system, the long elapsed time of execution of the job, and saves the information in the memory. A computer system characterized in that the priority of a job to be completed before a system suspension process start delay time of the computer system has elapsed is determined to be high. The computer system according to claim 4 or 5,
The control device may determine the priority of the job based on priority information added at the time of executing the job in advance, and determine a higher priority of the job with a higher priority specified in the priority information. A computer system characterized by the following. The computer system according to claim 7,
The control device determines the priority of the job based on the execution elapsed time of the job and the memory usage in the computer system for the job with the same priority specified in the priority information. A computer system, wherein a job execution elapsed time is long, and saving of information in the memory determines a high priority of a job to be completed before a system stop processing start delay time of the computer system elapses. 9. The computer system according to claim 7, wherein
A customer terminal connected to the computer system via a network,
The control device, the manager of the computer system, issued to the customer using the computer system via the customer terminal, priority information according to the priority use request of the user customer, To the customer, and when the customer uses the computer system via the network, accepts the execution of the job to which the priority information is added, and Is executed on the calculation server. The computer system according to claim 9,
The control device, the manager of the computer system, issued to the customer using the computer system via the customer terminal, priority information according to the priority use request of the user customer, A user setting table storing information of permission to execute the job with the priority information stored in the user setting table when the execution of the job added with the priority information from the customer is received. Comparing with the information, when the user is a job to which the priority information that is not permitted is added, stop the execution of the job and notify the customer that the job cannot be executed. Characteristic computer system.

Images