JP2005031892A - Job execution system and execution control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To avoid a service stop when a resource error occurs in a resource manager host which executes a job during execution of the job. <P>SOLUTION: Each of resource manager hosts A-X which execute jobs executes a job while managing resources such as a CPU, a memory and a disk, and reports a resource error to an integrated manager host 10 when the CPU efficiency is deteriorated, or the memory or disk is insufficient. The integrated manager host 10 searches another resource manager host which can execute the job performed in the resource manager host of the reporting source of the error or changes the job schedule to make another resource manager host execute the job performed in the resource manager host of the reporting source of the error. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a job execution system and an execution control method, and more particularly to an execution system and an execution control method for a scheduled job in an information processing system.
[0002]
[Prior art]
In recent information processing systems, in order to process business jobs unattended, job scheduling such as automatic start of a job by time reservation and automatic start of a connection job reserved next by the end of a preceding job is performed.
[0003]
In addition, one manager host is prepared as an information processing system so that jobs defined on the manager host can be executed on multiple other execution hosts, and execution results are centrally managed on the manager host. Systems that can do this are known.
[0004]
The job execution system as described above is re-executed on another host when a job that has been executed on a certain host cannot be executed on that host, as described in Patent Document 1 and the like. Can do.
[0005]
[Patent Document 1]
JP-A-11-353284
[0006]
[Problems to be solved by the invention]
The above-mentioned conventional technology that controls job execution by performing job scheduling determines the scheduling in consideration of machine information, execution job information, job execution conditions, etc. when building the job schedule. If the business data capacity is changed and the execution time of each job is significantly different from the forecast at the time of schedule construction, the execution of jobs assigned to the CPU of each host machine becomes inefficient. This causes a problem that the processing time of jobs scheduled in the system increases.
[0007]
In addition, the above-described conventional technology defines a job to be performed when an error occurs when a machine trouble occurs, such as a resource error such as insufficient memory or disk capacity, in the running host during system operation. Although troubles can be avoided, there is a problem that job execution in a normal case is stopped.
[0008]
An object of the present invention is to provide a job execution system and an execution control method that solve the above-described problems of the prior art and that can prevent a business operation from being stopped due to a resource error during job execution. is there.
[0009]
[Means for Solving the Problems]
According to the present invention, the object is to provide, in a job execution system for executing a scheduled job group, a plurality of resource manager hosts that execute jobs with a plurality of types of resources, and an integrated manager that manages job execution. A resource error detection means for detecting a resource error by monitoring the status of a resource during job execution, and a resource error notification means for reporting a resource error to the integrated manager host. In addition, the integrated manager host can execute a job performed on the resource manager host that has reported the resource error based on the resource status of the resource manager host other than the resource manager host that has reported the resource error. Relocation destination litho It is accomplished by having a means for determining a scan manager host.
[0010]
Further, the object is to provide a resource error report in a job execution control method for controlling execution of a scheduled job group by receiving a resource error report from a resource manager host that executes a job with a plurality of types of resources. Determine the migration destination resource manager host that can execute the job performed on the resource manager host that reported the resource error from the resource status of the resource manager host other than the resource manager host that performed the job. Is achieved.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of a job execution system and an execution control method according to the present invention will be described below in detail with reference to the drawings.
[0012]
FIG. 1 is a block diagram showing a configuration of a job execution system according to an embodiment of the present invention, FIG. 2 is a block diagram showing an internal configuration of an integrated manager host, and FIG. 3 is a block diagram showing an internal configuration of a resource manager host. 1 to 3, 10 is an integrated manager host, 20 to 40 are resource manager hosts, 50 to 90 are resources, 100 is a PP information database, 110 is a job schedule execution unit, 120 is a job execution unit, and 130 is a resource error Information receiving unit, 140 is a resource error information decoding unit, 150 is each resource usage status determining unit, 160 is a resource investigating unit, 170 is a resource deletion request unit, 180 is an installation instruction unit, 190 is an installation information transmitting unit, and 200 is a job Schedule definition change unit, 210 is a resource error detection unit, 220 is a resource error notification unit, 230 is a resource information notification unit, 240 is a command reception unit, 250 is a resource deletion unit, 260 is an installation information reception unit, and 270 is an installation execution unit It is.
[0013]
As shown in FIG. 1, the job execution system according to the embodiment of the present invention is configured by connecting an integrated manager host 10 and a plurality of resource manager hosts 20 to 40 via a network. The resource manager hosts 20 to 40 include various resources 50 to 90 such as CPUs, memories, and disks to be managed, or are connected to each other, and execute jobs. The integrated manager host 10 is connected to a PP information database 100 that stores and manages application (PP) information used by the resource manager host, and manages the schedule of jobs to be executed and the resource manager host that executes jobs. Manage.
[0014]
The integrated manager host 10 includes various functional units as shown in FIG. 2, and each of these functional units will be described next.
[0015]
The job schedule execution unit 110 controls the job schedule based on the already defined job schedule.
[0016]
The job execution unit 120 executes each job defined by the job schedule.
[0017]
The resource error information receiving unit 130 receives an error notification transmitted from the resource manager host.
[0018]
The resource error information decoding unit 140 decodes the error notification received by the resource error information receiving unit 130.
[0019]
The resource usage status determination unit 150 investigates how much the resource group of all resource manager hosts connected via the network is used, and a destination that can execute a job that has stopped due to a resource error Determine the resource manager host.
[0020]
When the resource manager host of the migration destination that can execute the job is determined, the resource investigating unit 160 determines information such as PP that must be installed on the resource manager host, and unnecessary information that exists on the resource manager host. Wash out.
[0021]
The resource deletion request unit 170 uses the resources of each resource manager host when none of the resource manager hosts has enough room to execute a job as a transfer destination as determined by the resource usage status determination unit 150. See the rate and request removal of unnecessary information to the resource manager who has resources that can be released.
[0022]
The installation instruction unit 180 requests the migration destination resource manager host to install information such as PP information necessary for execution of a job.
[0023]
The installation information transmission unit 190 transmits information to be actually installed on the machine to the resource manager host.
[0024]
The job schedule definition unit 200 determines in what order the job is used, which resource manager host is used, and in which time zone, and rewrites the job schedule table.
[0025]
Each resource manager host is configured to include various functional units as shown in FIG. 2, and each of these functional units will be described next.
[0026]
The resource error detection unit 210 detects a resource error in its own resource manager host, and the resource error notification unit 220 notifies the integrated manager host 10 of the acquired resource error information.
[0027]
The resource information notification unit 230 notifies the integrated manager host 10 of information on what resources are insufficient in the own resource manager host, and information such as PP necessary for job execution.
[0028]
The instruction receiving unit 240 receives and decodes the instruction transmitted from the integrated manager host 10.
[0029]
The resource deletion unit 250 releases resources based on information determined to be unnecessary notified from the resource deletion request unit 170.
[0030]
The installation information receiving unit 260 receives installation information necessary for job execution from the integrated manager host 10, and the installation execution unit 270 installs information necessary for job execution.
[0031]
Next, a processing operation for avoiding the suspension of a business due to a resource error during job execution in the job execution system configured as described above will be described. Before describing the processing operation, it is necessary in the processing. The configuration of the resource manager host resource monitoring screen, various data communication protocols, and various data tables will be described with reference to the drawings.
[0032]
FIG. 4 is a diagram showing an example of a resource monitoring screen used for monitoring resources of the own host by the resource manager host. This display screen displays the usage status of resources of the host registered in the resource manager host. In the illustrated example, the resource manager host displaying this screen displays the CPU as resources A to D. , Memory, disk, etc., resources A to C are used by job (1), resource A (CPU) usage rate is 30%, resources B (memory) and C (disk) are each 20 MB , 80 MB is being used.
[0033]
FIG. 5 is a diagram illustrating an example of a communication protocol for error information data. This protocol is used for notifying error information from the resource error notifying unit 220 of the resource manager host to the integrated manager host 10, and the start of error data indicating that this data is error information data at the head of the data. Followed by the resource manager host name where the error occurred, the resource type such as insufficient capacity that caused the error, the error type indicating what kind of error occurred, and the date and time when the error occurred Is set and configured.
[0034]
FIG. 6 is a diagram showing a configuration example of the error information data table. The error information data table is for the integrated manager host to store and manage the error information shown in FIG. 5 transmitted from the resource manager host. Indicate which error occurred in which resource on which resource manager host, the name of the resource manager host where the error occurred, the type of resource such as insufficient capacity that caused the error, and what kind of error occurred The error type indicating whether the error occurred and the date and time when the error occurred are managed. In the example of the record in the first row shown in the figure, in the resource manager host A, it is indicated that an error indicating “Empty” occurs in the date and time YYY / MM / DDhh / mm / dd in Memory1.
[0035]
FIG. 7 is a diagram illustrating an example of a communication protocol for resource information data. The resource information 1000 is resource information such as information on a job that the resource manager host having generated a resource error wants to move (being executed) transmitted to the integrated manager host, and PP information necessary for job execution.
[0036]
In the resource information 1000, a resource data start flag indicating that this data is resource information data is set at the head of the data. Subsequently, the transmission source resource manager host name, the ID of the job being executed, A transmission date and time is set, followed by the memory capacity necessary for job execution, disk capacity, recommended CPU performance, number of PPs, PP name and version, and finally installation of the PP to be transmitted after this resource information data The number of registry files necessary for the configuration and the number of information files necessary for the PP setting are set. In the illustrated example, there are PP names 1 and 2 of two PPs, and the version, the number of registry files, and the number of information files are set in each.
[0037]
FIG. 8 is a diagram illustrating a configuration example of the resource information data table. The resource information data table 1010 is obtained by dividing the resource information shown in FIG. 7 transmitted from the resource manager host on the integrated manager host side and storing it in the resource information data table.
[0038]
FIG. 9 is a diagram showing a configuration example of the performance information data table. This performance information data table 1011 is for the integrated manager host to check the resource usage status of other available resource manager hosts and to store the results. The managed resource manager host name, the number of resources, the date and time , Resource name, usage rate is stored. In the illustrated example, the managed resource manager host name HostA has three resources, and has the CPU, MEMORY, and DISK as resources in the date and time 2003/02/21 10:30, respectively, 50%, 30 % And 60%. In addition, the method of investigating a use condition can use a well-known method.
[0039]
FIG. 10 is a diagram for explaining the configuration of the PP information database 100 connected to the integrated manager host 10. In the PP information database 100, the PP name and version installed on the resource manager host managed by the integrated manager host are registered separately for each resource manager host. When a new PP is installed on the resource manager host to be managed, the PP name and version are sequentially registered in this database.
[0040]
FIG. 11 is a diagram illustrating an example of a communication protocol for an installation instruction. The installation instruction is an instruction that the installation instruction unit 180 transmits to the resource manager host when the information necessary for job execution is insufficient and the resource manager host is not in a job executable state.
[0041]
The installation instruction communication protocol 1100 sets an installation instruction data start flag indicating that this data is the data of the installation instruction at the head of the data, followed by the PP name to be installed, the version number, and after this data. The number of installation setting files necessary for installation to be transmitted, the number of registry information files necessary for setting the registry, and the number of information files necessary for environment setting of the PP to be installed are added.
[0042]
FIG. 12 is a diagram for explaining the installation information. The installation information 1110 is an information file group actually transmitted for installation that is transmitted after an installation instruction using the protocol described with reference to FIG. These file groups depend on the installer file to be installed, the installer setting file group in which information necessary for the installer is described, the registry file group necessary for setting the registry, and the PP environment setting file group necessary for setting the environment of the installed PP. Composed.
[0043]
FIG. 13 is a diagram illustrating an example of a communication protocol for a resource deletion request. This resource deletion request is executed when the resource manager host is requested to delete unnecessary information of resources when the resource manager host that can be executed is determined by redefining job scheduling. It is sent to the manager host. In this protocol, a deletion request start flag indicating that this data is the data of the deletion request is set at the head, and subsequently, the PP name to be deleted and the version number are assigned and transmitted.
[0044]
FIG. 14 is a flowchart for explaining the processing operation for avoiding resource errors in the integrated manager host when a resource error occurs in the resource manager host. This will be described next.
[0045]
(1) When all the jobs are executed normally, the integrated manager host is in a standby state waiting for resource error information to be transmitted from the resource manager host. When the resource error information is transmitted from the resource manager host in the form of the communication data protocol 900 described with reference to FIG. 5, this error information is received by the resource error information receiving unit 130 of the integrated manager host (step 300).
[0046]
(2) The resource error information receiving unit 130 passes the received resource error information to the resource error information decoding unit 140. The resource error information decoding unit 140 determines which resource of which resource manager host has an error and stores it in the error information data table 910 described with reference to FIG. 6 (step 310).
[0047]
(3) After decoding the information in step 310, the integrated manager host waits for resource information such as the information of the job that the resource manager host wants to move (that was being executed) and the PP information necessary for job execution, which is transmitted again from the resource manager host. It becomes a state. When the integrated manager host receives the resource information 1000 as described with reference to FIG. 7 from the resource manager host, the integrated manager host separates the resource information and stores it in the resource information table 1010 described with reference to FIG. The integrated manager host further receives the registry information file and the setting information file transmitted from the resource manager host, and temporarily stores these files in a user-specified directory (step 320).
[0048]
(4) Next, the integrated manager host acquires the resource performance information of each resource manager host (using a known function) by the resource usage status determination unit 150, and uses the performance table as described with reference to FIG. Create (step 330).
[0049]
(5) Next, the integrated manager host searches for resource manager hosts that can be used elsewhere. In this search, the resource in the performance information data table shown in FIG. 9 created in the process of step 330 and the job to be moved stored in the resource information table shown in FIG. This is done by comparing the required resources.
[0050]
For example, suppose that it is checked whether or not the job (1) executed on the resource manager host A can be transferred to the resource manager host B. As shown in FIG. 8, the maximum memory capacity required for job (1) is 71 MB, and the memory usage rate in the resource usage rate of the resource manager host B is shown in FIG. So that it is 50%. Therefore, it is checked whether the remaining capacity of the memory provided in the resource manager host B, in this case, the remaining capacity of 50% is 71 MB or more. Similarly, the disk capacity as a resource required by job (1) is 1 GB as shown in FIG. 8, and the disk usage rate in the resource usage rate of the resource manager host B having two disks is As shown in FIG. 9, they are 40% and 70%, respectively. Therefore, it is checked whether the remaining capacity of the capacity of the two disks provided in the resource manager host B, in this case, the remaining capacity of either 60% or 30% is 1 GB or more.
[0051]
If all resources are compared as described above, and the result of this comparison is that another resource manager host having a larger usable capacity than the resource required by the job to be moved is found, the available resource manager host is found. If it cannot be found, it is determined that there is no usable resource manager host (step 340).
[0052]
(6) If no other resource manager host that can be used is found in the determination in step 340, the integrated manager host reconfigures the job schedule, and as a result of the reassembly, the resource manager host that executes the job whose execution order is slow Is determined as an executable resource manager host. The job schedule can be reconfigured using a known means (steps 350 and 360).
[0053]
(7) After the process of step 360 or when the resource manager host that can be used is found in the determination of step 340, the resource investigating unit 160 needs to execute the job to the target resource manager host that executes the job. It is checked whether or not the resource manager host is in a job executable state, such as whether or not there is PP information or the like, or whether there is unnecessary information in the target resource manager host. The necessary PP is examined based on the PP information 101 of the management target resource manager host registered in the PP information database 100 connected to the integrated manager host described with reference to FIG. This is done by comparing the necessary PP names of information (step 370).
[0054]
(8) It is determined whether or not the resource manager host is in a job executable state based on the result of whether the necessary PP is registered in the PP information of the management target resource manager host in the investigation in step 370. That is, if the necessary PP is registered in the PP information of the managed resource manager host, it is determined that the resource manager host is in a job executable state, and if it is not registered, it is determined that the job is not in a job executable state. To do. If the resource manager host is not in an executable state, there is insufficient information necessary for job execution, or there are resources used by the postponed job, and as a result, there are insufficient resources Is determined (step 380).
[0055]
(9) If it is determined in step 380 that the information necessary for job execution is not sufficient and the resource manager host is not in a job executable state, the installation instruction unit 180 performs the installation instruction communication as described with reference to FIG. A protocol 1100 is created and sent to the resource manager host (step 390).
[0056]
(10) Subsequently, the installation information transmission unit 190 transmits the installation information 1110 which is a necessary PP installer and setting file group as described with reference to FIG. 12 stored in the user-specified directory to the resource manager. Thereafter, the processing returns to step 370 and the processing from the resource investigation is continued again (step 400).
[0057]
(11) If it is determined in step 380 that there is a resource used by the postponed job and, as a result, the necessary resource is insufficient and the resource manager host is not in a job executable state, that is, When it is determined in step 340 that an executable resource manager host cannot be found and the executable resource manager host is determined by redefining job scheduling, the resource deletion requesting unit 170 determines that resource. Request the manager host to delete the resource. Here, a deletion request protocol 1200 for unnecessary information on resources as shown in FIG. 13 is created and transmitted to the resource manager host. In determining the PP information to be deleted, the PP information of the job that has been delayed in the processing of Step 350 and Step 360 is set as the deletion target PP. Thereafter, the processing returns to the processing from step 370 and the processing from the resource investigation is continued again (step 410).
[0058]
(12) If it is determined in step 380 that the resource manager host is in a job executable state, the job schedule definition changing unit 200 reconfigures the job schedule, and the job execution unit 120 resumes job execution. To do. The reorganization of the job schedule and the job execution can be performed by known means (steps 420 and 430).
[0059]
FIG. 15 is a diagram for explaining an example of job schedule recombination performed in the process of step 350 described above, which will be described next.
[0060]
As indicated by 600 in FIG. 15A, jobs (1) and (2) are executed on the same or different resource manager hosts, and now jobs (3), (4), and (5) are different resource managers. Assume that a resource error occurs in the resource manager host that is executing job (3) in this state. At this time, when the process is shifted to the job schedule reassembling process in step 350 and the job schedule is reconfigured, the resource manager host executing the job (5) whose execution order may be delayed. Move job (3) to execute. That is, the resource manager host executing the job (5) does not need the resources necessary for executing the job (5), and uses the resources necessary for the execution of the job (5). Thus, it is possible to execute the process of job (3). Job (5) is scheduled to be executed after execution of job (3) or job (4), but since job (4) is already being executed, job (4) is job (3). The process ends before ▼. Therefore, the job (5) moves to the resource manager host that is executing the job (4). As a result of the reorganization of the job schedule, the schedule after the job schedule is changed becomes as shown by 601 in FIG. Note that the job {circle around (6)} that should have been executed last in FIG. 15A is executed last even in the case of the reassembly shown in FIG.
[0061]
FIG. 16 is a flowchart for explaining a processing operation for avoiding a resource error in the resource manager host when a resource error occurs in the resource manager host, which will be described next. The processing here includes both processing when a resource error occurs in its own resource manager host and when a resource error occurs in another resource manager host.
[0062]
(1) Each resource manager host monitors resources in its own host. Now, it is assumed that a resource shortage has occurred in the resource A-1 in the resource manager host A20 shown in FIG. In this case, the resource error detection unit 210 (resource monitoring screen described with reference to FIG. 4) of the resource manager host detects the resource shortage error. This error detection can be performed by known means, and a threshold value (user setting) of the usage rate is provided for each resource, and if it exceeds that, it is regarded as an error (step 500).
[0063]
(2) When the resource error detection unit 210 detects a resource error, the resource error information notification unit 220 of the resource manager host notifies the integrated manager host that a resource error has occurred. This notification is performed by creating the error information data communication protocol 900 described with reference to FIG. 5 and transmitting it to the integrated manager host (step 510).
[0064]
(3) Next, the resource manager host notifies the integrated manager host of resource information such as job information that cannot be executed and PP information necessary for job execution from the resource information notification unit 230. As described with reference to FIG. 7, this notification is generated by creating a resource information communication protocol, and then creating a registry information file from the registry information of the necessary PP. These files are copied and subsequently transferred to the integrated manager host after the resource information notification (step 520).
[0065]
After the processing of step 520, the resource manager host that has generated the resource error enters a state of waiting for an instruction from the integrated manager host side. In addition, the resource manager host in which no resource error has occurred is in a state of waiting for an instruction from the integrated manager host side while monitoring resources.
[0066]
(4) When receiving the command transmitted from the integrated manager host, the command receiving unit 240 of the resource manager host determines whether the command is a “resource deletion request”, “installation instruction”, or “job execution”. It is divided into some (step 530).
[0067]
(5) If the instruction received in step 530 is an installation instruction, the installation information receiving unit 260 receives the installation information sent next, and the installation executing unit 270 installs the installation information in the machine. To do. After that, it again enters a command waiting state (steps 550 and 560).
[0068]
(6) If the command received in step 530 is a resource deletion request, the resource deletion unit 250 deletes unnecessary information specified in the received deletion request data, and then again waits for a command. (Step 540).
[0069]
(7) If the command received in step 530 is job execution, the resource manager host starts job execution and starts monitoring each resource used during execution (step 570).
[0070]
Each processing in the above-described embodiment of the present invention can be configured as a processing program, and this processing program is stored in a recording medium such as HD, DAT, FD, MO, DVD-ROM, and CD-ROM and provided. can do.
[0071]
According to the above-described embodiment of the present invention, the resource manager host that has generated the resource error during job execution reports the occurrence of the resource error to the integrated manager host, and the job being executed by another resource manager host. Therefore, it is possible to avoid a business stop due to a resource error.
[0072]
In the above-described embodiment of the present invention, the case where a resource error has occurred in the resource manager host has been described as an example. However, in the present invention, a failure occurs in a function other than the resource of the resource manager host, and the resource manager This can also be applied when the job cannot be executed on the host. In this case, the integrated manager host may detect and respond to the fact that communication with the resource manager host is no longer possible.
[0073]
【The invention's effect】
As described above, according to the present invention, even when a resource error occurs in a resource manager host that is executing a job during job execution, it is possible to avoid a business stop.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a job execution system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing an internal configuration of an integrated manager host.
FIG. 3 is a block diagram showing an internal configuration of a resource manager host.
FIG. 4 is a diagram showing an example of a resource monitoring screen used for monitoring resources of the host by the resource manager host.
FIG. 5 is a diagram illustrating an example of a communication protocol for error information data.
FIG. 6 is a diagram illustrating a configuration example of an error information data table.
FIG. 7 is a diagram illustrating an example of a communication protocol for resource information data.
FIG. 8 is a diagram illustrating a configuration example of a resource information data table.
FIG. 9 is a diagram illustrating a configuration example of a performance information data table.
FIG. 10 is a diagram illustrating a configuration of a PP information database connected to an integrated manager host.
FIG. 11 is a diagram illustrating an example of a communication protocol for an installation instruction.
FIG. 12 is a diagram illustrating installation information.
FIG. 13 is a diagram illustrating an example of a communication protocol for a resource deletion request.
FIG. 14 is a flowchart for explaining a resource error avoidance processing operation in the integrated manager host when a resource error occurs in the resource manager host.
FIG. 15 is a diagram illustrating an example of job schedule reorganization.
FIG. 16 is a flowchart for explaining a resource error avoidance processing operation in the resource manager host when a resource error occurs in the resource manager host;
[Explanation of symbols]
10 Integrated manager host
20-40 Resource manager host
50-90 resources
100 PP information database
110 Job schedule execution part
120 Job execution part
130 Resource error information receiver
140 Resource error information decoding part
150 Resource usage status determination unit
160 Resource Survey Department
170 Resource deletion request section
180 Installation instruction section
190 Installation information transmitter
200 Job schedule definition change part
210 Resource error detector
220 Resource error notification section
230 Resource information notification part
240 Command receiving part
250 Resource deletion part
260 Installation information receiver
270 Installation execution part

Claims (6)

In the job execution system for executing a scheduled job group, the resource manager host includes a plurality of resource manager hosts that execute jobs with a plurality of types of resources, and an integrated manager host that manages job execution. Includes resource error detection means for detecting resource errors by monitoring the status of resources during job execution, and resource error notification means for reporting resource errors to the integrated manager host. Determine the migration destination resource manager host that can execute the job that was being performed on the resource manager host that reported the resource error from the resource status of the resource manager host other than the resource manager host that reported the error Job execution system, characterized in that it comprises means that. In the job execution control method that controls the execution of scheduled job groups, the resource manager host that received the resource error report from the resource manager host that executed the job with multiple types of resources, and that reported the resource error Job execution characterized by determining the migration-destination resource manager host that can execute the job performed on the resource manager host that reported the resource error from the resource status of other resource manager hosts Control method. 3. The job execution control method according to claim 2, wherein an application which is insufficient on the migration destination resource manager host is transmitted to the migration destination resource manager host for installation. 4. The job execution control according to claim 2, wherein when a migration destination resource manager host cannot be determined from the status of the resource, a job schedule is changed to determine a migration destination resource manager host. Method. 5. The job execution control method according to claim 2, wherein when there is unnecessary information in the resource of the resource manager host of the migration destination, the information is deleted. In a job execution control program that controls the execution of scheduled jobs, a processing step for receiving a resource error report from a resource manager host that executes a job with multiple types of resources and a resource error report were performed. A processing step for determining a migration destination resource manager host capable of executing a job performed on the resource manager host that has reported the resource error based on the resource status of the resource manager host other than the resource manager host. A job execution control program that controls execution of the job by executing each of the processing steps.

Images