JP2007164642A - Resource management device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a resource management device and method which achieve more efficient use of resources. <P>SOLUTION: The resource management device has: a first table for expressing the resources, the processing content, processing assignment start time and processing assignment end time while making them correspond to one another; a second table for expressing the resources and resource providers while making them correspond to each other; a third table for expressing the processing content and required processing time while making them correspond to each other and a fourth table for expressing the resource providers, action classifications of the resource providers and action time of day in the action classifications while making them correspond to one another and controls processing by the resources provided by the resource providers by using the fourth table. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a resource management apparatus and a resource management method for managing resources.

A grid system that performs processing using resources (computers) not used by resource providers (computer management users) is used.
Here, the start and end of provision of resources are usually performed by voluntary actions from resource providers. That is, the resource provider starts providing resources by allowing their resources to participate in the grid system on time. In addition, the resource provider stops the resource provision by causing the resource provider to leave the grid system.
In a distributed system that is distributedly managed by a plurality of resource management apparatuses, a technique for avoiding continuous failure of reservation when executing a job for reserving resources in advance has been disclosed (Patent Document 1). reference).
JP 2004-302748 A

When the resource provider controls the start and end of the provision of resources, the following inconvenience arises in terms of efficient use of the computer. That is, resource provision is started and stopped at the will of the resource provider, and scheduling of resource provision is difficult. For example, there is a possibility that the resource provider suddenly stops providing the resource. In addition, there is a possibility that a time lag will occur when job allocation processing is performed on a computer on which the OS is not activated.
In view of the above, an object of the present invention is to provide a resource management apparatus and a resource management method that enable more efficient use of resources.

A resource management device according to an aspect of the present invention includes a first table that represents resources, processing contents, processing allocation start time, and processing allocation end time, a resource, a resource provider, Corresponding to the second table, the third table corresponding to the processing content and the processing time required, the resource provider, the action type of the resource provider, and the action type An execution request information receiving unit for receiving execution request information including a fourth table corresponding to an action time zone, a processing content, a processing start time, and a processing end time limit;
Based on the first, second, third, and fourth tables, and the received execution request information, a table updating unit that updates the first table, and the updated first table And a process execution unit that causes the resource to execute the process.

  A resource management method according to an aspect of the present invention includes a step of receiving execution request information including a process content, a process start time, and a process end time limit, a resource, a process content, and a process allocation start time. A first table for storing process allocation end time correspondingly, a second table for storing resource and resource provider correspondingly, a processing content, and a processing time required correspondingly The third table, the resource provider, the action type of the resource provider, the action time zone in the action type, and the received execution request information. And updating the first table, and causing the resource to execute a process based on the updated first table.

  According to the present invention, it is possible to provide a resource management apparatus and a resource management method that enable more efficient use of resources.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(First embodiment)
FIG. 1 is a system configuration diagram showing a grid system according to the first embodiment of the present invention. FIG. 2 is a functional configuration diagram showing details of the functional configuration of the master host 12 according to the present embodiment.
The grid system includes a client 11, a master host 12, a providing computer 13, a resource provider schedule storage unit 14, and a data server 15.

  The client 11 is a functional unit that requests the master host 12 to process a job, for example, a computer. The client 11 has an input unit for inputting data, an output unit for outputting data, and a display unit for displaying data. Execution request information is input from the input unit and transmitted to the master host 12 to request job processing. The job execution result transmitted from the master host 12 is output and displayed by the output unit and the display unit.

FIG. 3 is a schematic diagram illustrating an example of execution request information.
The execution request information includes a job name, a job start time, a time limit, and a used OS name.
“Job name” is the name of a job and is a type of information for identifying a job executed by the providing computer 13.
The “job start time” is the time when job execution is started.
“Time limit” is the time limit for the end of job execution.
The “used OS name” is a name of an OS (operating system) used for executing a job, and is a kind of information for identifying the OS.

The master host 12 is a system unit, for example, a computer, which controls the grid system and effectively uses the resources of the providing computer 13. Details of this will be described later.
The provided computer 13 (13 (1), 13 (2)) is a resource provided to the grid system, for example, a computer (computer). Here, the resource refers to processing capability such as calculation and information processing. The provided computer 13 is divided into a provided computer main body 131 and an auxiliary storage device 132 such as a hard disk (HD). The provision computer main body 131 includes an input unit, an output unit, a display unit, and a storage unit. Here, as an example, two provided computers 13 are shown, but this number may be one, or three or more.

The resource provider schedule storage unit 14 stores an action schedule (schedule) of a person (resource provider) who manages and uses the provided computer 13. The resource provider's schedule (electronic schedule table) is electronically read and written in the resource provider schedule storage unit 14 by schedule management software (for example, Cybozu).
As long as the resource provider schedule storage unit 14 can be read from the master host 12, various configurations can be adopted. For example, the resource provider schedule storage unit 14 can be a computer independent of the master host 12.

The resource provider schedule storage unit 14 stores information on resource provider name, action type, action start time, and action end time. The stored contents of the resource provider schedule storage unit 14 correspond to a fourth table that represents the resource provider, the action type of the resource provider, and the action time zone in the action type.
The “resource provider name” is information for identifying a management user of the providing computer 13.
The “behavior type” is information for identifying the behavior of the resource provider, for example, work (meeting, experiment, etc.), going out, vacation, outside working hours (going home). Of these, types other than work, such as going out, vacation, and working hours outside (returning home) indicate that the resources of the providing computer 13 can be provided to the grid system.
“Action start time” is the time when the resource provider starts the action of the action type.
“Action end time” is the time when the resource provider ends the action of the action type.

  The data server 15 is, for example, a computer having an auxiliary storage device 151 that stores an OS (operating system) and application programs used by the providing computer 13.

  The master host 12 includes a data recording unit 21, a CPU name acquisition unit 22, a computer registration information acquisition unit 23, an allocation information acquisition unit 24, a computer schedule setting unit 25, an allocation determination unit 26, a resource provision time zone acquisition unit 27, and an execution process. Part 28.

  The data recording unit 21 stores data, and stores a computer registration table T1, a job execution time table T2, a computer schedule table T3, and an allocation candidate table T4. A “table” is a collection of information in which a plurality of pieces of information are generally associated, but a single piece of information may be referred to as a table. The same applies to the later tables T5 to T8.

In the computer registration table T1, information of a computer ID, computer performance, and resource provider name is shown correspondingly. The computer registration table T1 corresponds to a second table that represents resources and resource providers correspondingly.
FIG. 4 is a schematic diagram illustrating an example of the computer registration table T1.
“Computer ID” is information (IDentifier) for identifying the provided computer 13, and a computer name is used in this embodiment. The “computer name” is a name of a computer and is a kind of information for identifying the provided computer 13 itself.
“Computer performance” is information for identifying the performance of the providing computer 13, and a CPU name is used in this embodiment. This is because the processing speed varies depending on the CPU. The “CPU name” is the name of the CPU of the providing computer 13 and is a kind of information for identifying the CPU.
The “resource provider name” is information for identifying a management user of the providing computer 13.

In the job execution time table T2, information on job name, execution environment, and execution time is shown correspondingly. The job execution time table T2 corresponds to a third table that represents the processing contents and the processing time required.
FIG. 5 is a schematic diagram illustrating an example of the job execution time table T2.
“Job name” is the name of a job and is a type of information for identifying a job executed by the providing computer 13.
“Execution environment” is information for identifying an environment in which a job is executed (a factor that affects a job processing speed), and is represented by an OS name and a CPU name in the present embodiment. The “OS name” is information for identifying the OS of the providing computer 13 and functions as operating system identification information.
“Execution time” is information representing the time required for execution processing of each job in a predetermined execution environment.

The computer schedule table T3 shows job assignment information to be processed. The computer schedule table T3 corresponds to a first table that shows resources, process contents, process allocation start time, and process allocation end time in association with each other.
FIG. 6 is a schematic diagram illustrating an example of the computer schedule table T3.
A computer name, a job name, a job allocation start time, a job allocation end time, an OS name, and an OS activation flag are represented in correspondence.
“Job allocation start time” represents the time at which job allocation to the provided computer 13 starts.
“Job allocation end time” represents the time when the job allocation to the providing computer 13 ends.
The “OS activation flag” indicates whether or not the OS is required to be started by the providing computer 13 and functions as activation necessity information indicating whether or not the resource activation is necessary. When the “OS activation flag” is ON, the OS is activated. When the “OS activation flag” is not ON (OFF), the OS is not activated.

In the allocation candidate table T4, temporary information used when a job is allocated to the providing computer 13 is represented.
FIG. 7 is a schematic diagram illustrating an example of the allocation candidate table T4. A computer name, a job start time, a job execution expected end time, and an OS activation flag are represented correspondingly.
“Job start time” represents the time at which the execution of the job starts in the provided computer 13 (assignable time zone of the candidate computer).
“Estimated job execution end time” represents the expected end time of job execution in the providing computer 13.

The CPU name acquisition unit 22 acquires a CPU name corresponding to the execution request information from the client 11. Specifically, the CPU name in the job execution time table T2 is acquired in association with the job name and the used OS name included in the execution request information.
The computer registration information acquisition unit 23 acquires the computer name corresponding to the CPU name acquired by the CPU name acquisition unit 22 and its resource provider name from the computer registration table T1.
The allocation information acquisition unit 24 acquires allocation information from the job start time to the time limit from the computer schedule table T3.
The computer schedule setting unit 25 records job assignment information in the computer schedule table T3, and corresponds to an additional registration unit.

The assignment determination unit 26 determines the possibility of job assignment and writes the result in the assignment candidate table T4. Specifically, the job assignment possibility is determined using the job assignment information related to the computer acquired from the computer schedule table T3 and the resource provision time zone obtained from the resource provider schedule storage unit 14.
The allocation determination unit 26 is a time zone calculation unit that calculates an allocatable time zone for processing, a scheduled end time calculation unit that calculates a scheduled processing end time, an allocation availability determination unit that determines whether processing content can be allocated, and a resource It functions as an activation necessity determination unit that determines whether activation is necessary.

  The resource provision time zone acquisition unit 27 acquires the resource provision time from the resource provider schedule storage unit 14 and sends it to the allocation determination unit 26. The resource provision time zone acquisition unit 27 acquires the resource provision time related to the specified resource provider name from the resource provider schedule storage unit 14. As the resource provision time, information on going out, vacation, and outside working hours (while going home) among the action types set in the resource provider schedule storage unit 14 is acquired.

FIG. 8 is a schematic diagram illustrating an example of resource provider schedule information that the resource provision time zone acquisition unit 27 acquires from the resource provider schedule storage unit 14.
Information of resource provider name, action type, action start time, and action end time is represented correspondingly.
The “resource provider name” is information for identifying the administrator or user of the providing computer 13.
The “behavior type” is information for identifying the behavior of the resource provider, for example, going out, vacation, or outside working hours (going home). An action type (other than work near the providing computer 13) indicating that the resources of the providing computer 13 can be provided to the grid system is represented.
“Action start time” is the time when the resource provider starts the action of the action type.
“Action end time” is the time when the resource provider ends the action of the action type.
The execution processing unit 28 acquires job assignment information from the computer schedule table T3, and controls OS activation and job execution processing in the providing computer 13.

  The master host 12 functions as a detachment control unit that controls detachment of the auxiliary storage device 132 from the providing computer main body 131 and an activation control unit that controls activation of the providing computer main body 131.

(Operation of the grid system in the first embodiment)
An operation procedure of the grid system will be described.
FIG. 9 is a flowchart showing an example of a basic operation procedure of the grid system according to the present embodiment. FIG. 10 is a flowchart showing details of step S17 in FIG. FIG. 11 is a flowchart showing details of step S19 in FIG.

A. Setting of master host 12 (step S11)
The following pre-setting is performed on the master host 12.
Register information on the computer name, CPU name, and resource provider name of the providing computer 13 in the computer registration table T1 of the data recording unit 21 (see FIG. 4).
For a job that may be handled as a grid system, the job name, OS name, CPU name, and execution time are registered in the job execution time table T2 of the data recording unit 21 in the master host 12 (see FIG. 5).

B. Input to resource provider schedule storage unit 14 (step S12)
It is assumed that the resource provider who provides the providing computer 13 sequentially inputs his / her schedule to the resource provider schedule storage unit 14.

C. Receiving execution request information from the client 11 (step S13)
A job execution request is made by sending execution request information from the client 11 to the master host 12. For example, the execution request information shown in FIG. 3 is received.

D. Acquisition of CPU name (step S14)
In the master host 12 that has received the execution request information from the client 11, the CPU name acquisition unit 22 acquires the CPU name and sends it to the computer registration information acquisition unit 23. The CPU name (pentium3, pentium4, pentium5) corresponding to the job execution time table T2 is acquired based on the job name (JOB3) and the OS name (Windows 2000 (registered trademark)) in the execution request information.

E. Acquisition of computer name and resource provider name (step S15)
The computer registration information acquisition unit 23 acquires the computer name (machine A, machine B) and the resource provider name (provider A, provider B) corresponding to the CPU name from the computer registration table T1, and the computer name is assigned information. The resource provider name is sent to the acquisition unit 24 to the resource provision time zone acquisition unit 27, respectively.

F. Acquisition of computer allocation information and resource provision time zone information (step S16)
(1) The allocation information acquisition unit 24 acquires computer allocation information from the computer schedule table T3 (see FIG. 6) and sends it to the allocation determination unit 26. That is, the computer allocation information from the job start time to the time limit corresponding to the computer name sent from the computer registration information acquisition unit 23 is acquired from the computer schedule table T3. Here, the computer assignment information represents job assignment in the providing computer 13.
FIG. 12 is a schematic diagram showing the computer schedule of FIG. 6 in time series, and the situation in which the requested job 3 is assigned is indicated by a dotted line.

(2) The resource provision time zone acquisition unit 27 acquires the resource provision time zone information (see FIG. 8) and sends it to the allocation determination unit 26. FIG. 13 is a schematic diagram showing the resource provider schedule of FIG. 8 in time series.
Based on the resource provider name (provider A, provider B) sent from the computer registration information acquisition unit 23, the resource provider schedule storage unit 14 responds to the job from the job start time to the deadline. Get resource provision time zone information that is possible. The availability of resources is indicated by the fact that the action type in the resource provider schedule storage unit 14 is, for example, going out, vacation, or not working hours (returning home).
Resource provision time zone from the acquired action start time to action end time ((1) Provider A: Outing: Start 2005/02/01/07: 30, End 2005/02/01/14: 45, (2) Provided Person A: Going out: Start 2005/02/01/15: 45, End 2005/02/01/21: 00, ---) is sent to the assignment determination unit 26.

G. Allocation determination (step S17)
The allocation determination unit 26 performs allocation determination for each candidate computer 13 and for each assignable time zone (steps S21 to S34 and steps S23 to S33).
For this determination, job allocation information from the allocation information acquisition unit 24, resource provision time zone from the resource provision time zone acquisition unit 27, and execution request information from the client 11 (job name, deadline, OS name used, job) Start time) is used.

(1) Calculation of assignable time zone (step S22)
An allocatable time zone is calculated. That is, in the time range from the job start time to the deadline, the resource provision time in the time zone in which the job assignment that is the inversion of the assignment information sent from the assignment information acquisition unit 24 is not performed, and in the range from the job start time to the deadline A time zone in which two time zones of the resource provision time zone sent from the acquisition unit overlap is extracted. The extracted time zone (assignable time zone) is, for example, assignable time zone A1 (start 2005/02/01/08: 05, end 2005/02/01/12: 00), A2 (start 2005/02 / 01/15: 00, End 2005/02/01/17: 00), ---.

(2) Necessity determination of OS activation (step S24)
Using the determination J1 and the determination J2 defined as follows, it is determined whether the OS needs to be started for each assignable time zone.
[Judgment J1]: The resource provider plans to use immediately before the allocatable time zone. That is, it is a case where the use as the providing computer 13 is not continuously performed. If this determination J1 is “Yes”, it is necessary to start the OS.
[Judgment J2]: The OS name of the allocation information immediately before the allocatable time zone is the same as the used OS name of the execution request information. If this determination J2 is “No”, it is necessary to start the OS.
Except for the case where the judgment J1 = “No” and the judgment J2 = “Yes”, the OS needs to be activated.

(3) Determination of OS startup flag / allocation required time (steps S25 to S28)
The OS startup flag and the allocation required time are determined.
(A) When J1 = “Yes”, J1 = “No”, and J2 = “No” In this case, it is necessary to start the OS.
“Allocation required time = job execution time + OS activation time”. That is, the OS startup time is added to the execution time. The OS startup time is determined as appropriate. Also, the OS activation flag is determined to be ON.

(B) When J1 = “No” and J2 = “Yes” In this case, it is not necessary to start the OS.
“Allocation required time = job execution time”. The OS startup flag is determined to remain OFF.

(4) Determining whether or not allocation is possible (steps S29 to S32)
The job end time is calculated from the job start time and the required allocation time.
If the job end time is earlier than the end time of the assignable time zone, it is determined that assignment is possible. If the allocation is possible, information on the job start time, job end time, and OS activation flag is recorded in the allocation candidate table T4.
The above processing is performed for each assignable time zone of the computer. When one computer is completed, the same processing is performed for each extracted computer.

H. Assignment setting (step S18)
Registration in the computer schedule table T3 is performed based on the allocation candidate table T4.
The computer schedule setting unit 25 registers the job assignment information in the computer schedule table T3 according to the assignment setting (for example, the one to be executed most recently) (for example, computer name: machine A, job name: JOB3, Job assignment start time 2005/02/01/09: 00, job assignment end time 2005/02/01/10: 48, OS name: Windows2000, OS flag: none).
Note that FIG. 12 described above represents a situation in which the requested job 3 is assigned by a dotted line.

I. Execution process (step S19)
First, a case where the auxiliary storage device 132 is not disconnected will be described.
The job is executed according to the computer schedule table T3. The execution processing unit 28 causes the providing computer 13 to execute the job at the time (2005/02/01/09: 00) at which the job is assigned in the computer schedule table T3. Here, if the OS start flag is ON in the computer schedule table T3, the OS is started prior to the execution process.
When the job execution process is completed, the execution processing unit 28 transmits the job execution result to the client 11. Further, the execution processing unit 28 updates the information in the computer schedule table T3 through the computer schedule setting unit 25.

Next, a case where the auxiliary storage device 132 is disconnected will be described.
FIG. 5 is a flowchart showing the execution procedure of the auxiliary storage device 132 detachment process by the data server 15 in step S19 of FIG.
(1) HDD disconnection processing (step S41)
When the time when the job is assigned in the computer schedule table T3, the execution processing unit 28 performs the HDD disconnection processing of the provision computer 13 scheduled to be used. This disconnection means disconnection of the electrical connection between the providing computer main body 131 and the auxiliary storage device 132. When the providing computer main body 131 is activated, the auxiliary storage device 132 is disconnected after the providing computer main body 131 is stopped.

(2) OS network boot (steps S42 and S43)
When the OS flag of the corresponding computer in the computer schedule table T3 is ON, the execution processing unit 28 performs network booting of the OS to the providing computer 13. The OS is acquired from the data server 15. PXE (Preboot eXecution Environment) can be used for network boot. PXE is a kind of network boot standard.

Network boot is executed in the following procedure.
The power supply of the providing computer main body 131 from which the auxiliary storage device 132 is disconnected is turned on. As a result, the operating system and applications in the auxiliary storage device 151 of the data server 15 are expanded on the memory of the auxiliary storage device 132 by remote boot, and the providing computer main body 131 is activated.

(3) Job execution (step S44)
When the booting of the OS is completed, the target computer grid control unit causes the providing computer 13 to acquire the application from the data server 15 and execute the job.

(4) Transmission of job results and update of computer schedule table T3 (steps S45 and S46)
When the job execution is completed, the execution processing unit 28 transmits the job execution result to the client 11. Further, the execution processing unit 28 updates the information in the computer schedule table T3 through the computer schedule setting unit 25.

In this embodiment, the resources of the providing computer 13 are effectively utilized by assigning jobs as follows.
(1) Allocate jobs in consideration of the resource provider's schedule.
(2) Allocate jobs in consideration of the OS startup time.
(3) If the OS used to execute the job has already been started on the providing computer 13 (when the job that was submitted immediately before was used by the providing computer 13), the OS Allocating jobs in consideration of the fact that startup is not required.
(4) By using the network boot method, when resources return to the provider, the difficulty in terms of time and work is greatly reduced. The content of the auxiliary storage device 132 is not rewritten, and the burden of subsequent use on the resource providing side is reduced.

(Second Embodiment)
FIG. 14 is a system configuration diagram showing a grid system according to the second embodiment of the present invention.
In the present embodiment, a GPS information transmission unit 16 is added to the grid system. Further, a GPS information acquisition unit 31 and a resource provider information correction unit 32 are added to the master host 12A, and a latitude / longitude correspondence table T5 and a GPS information table T6 are added to the data recording unit 21A. That is, in this embodiment, in addition to the resource provider schedule storage unit 14, it is possible to grasp the resource provision time based on GPS information. A system for receiving and interpreting data from GPS information is added in the master host 12.
In addition, as a system which can grasp | ascertain the action schedule of a resource provider, there exist the position observation system GPS, the entrance / exit system installed in the entrance, etc., Here, GPS is taken as an example.

The GPS information transmission unit 16 is an external position information distribution unit, and is, for example, a portable device having position measurement means such as GPS and transmission means. The resource provider carries this portable device, and the portable device transmits the location information of the resource provider. The GPS information acquisition unit 31 records the information from the GPS transmitted from the GPS information transmission unit 16 in the GPS information table T6. It is a functional part to do. The GPS information acquisition unit 31 functions as a location information receiving unit that receives location information of the resource provider.
The resource provider information correction unit 32 considers the information from the resource provider time zone acquisition unit 27 and the location information of the resource provider from the GPS information table T6 and the time information from the latitude / longitude correspondence table T5. This is a functional unit that sends schedule information to the allocation determining unit 26. The resource provider information correction unit 32 functions as a prediction unit that predicts the process allocation start time.
In the latitude / longitude correspondence table T5, the latitude, the longitude, and the information of the longitude, the required time from the latitude to the target point are represented in correspondence.
Information obtained from GPS is represented in the GPS information table T6.

(Operation of Grid System in Second Embodiment)
The operation contents of the grid system in the second embodiment will be described below.
(1) Reception of data from GPS The location information of the resource provider is transmitted from the GPS information transmission unit 16 and received by the master host 12. The GPS information acquisition unit 31 in the master host 12 acquires the location information of the resource provider from external GPS information. The information is recorded in the GPS information table T6 in the master host 12.

(2) Prediction of resource provider behavior The master host 12 predicts the resource provider behavior based on the location information of the resource provider received from the GPS information transmitter 16.
For example, the resource provider information correction unit 32 determines how many minutes from the latitude / longitude information of the resource provider's GPS information table T6 and the latitude / longitude correspondence table T5 to the target point (return target return position) (scheduled return time). Is calculated. That is, before the resource provider returns to the company office, the resource provider information correction unit 32 predicts a new scheduled return time from the scheduled time.

(3) Correction of resource provision time The resource provision time obtained from the resource provider schedule storage unit 14 is corrected. For example, the calculated return time is overwritten on the numerical value obtained from the resource provision time zone acquisition unit 27 and sent to the allocation determination unit 26.
By assigning a job using the corrected resource provision time, it is possible to cope with a situation of returning earlier or later than the time registered in the resource provider schedule storage unit 14.

Since the processing in the allocation determination unit 26 is the same as that in the first embodiment, the description thereof is omitted.
Note that assignment with higher accuracy can be achieved by performing assignment determination again for those already set in the computer schedule table T3. Job allocation processing is performed using more accurate resource provision time by using means such as GPS that can predict the behavior of resource providers.

  In the present embodiment, job allocation processing can be performed more efficiently by using means such as GPS that can predict the behavior of a resource provider.

(Third embodiment)
FIG. 15 is a system configuration diagram showing a grid system according to the third embodiment of the present invention.
In the present embodiment, the execution request information setting unit 41, the OS preliminary preparation number setting unit 42, and the OS preliminary preparation state setting unit 43 are included in the master host 12B, and the execution request information table T7 and the OS preliminary preparation number are included in the data recording unit 21B. A table T8 and an OS advance preparation status table T9 are added. In other words, it is not registered yet in future scheduling, but the OS is activated in advance.

The execution request information setting unit 41 is a functional unit that records execution environment information necessary for a job requested from the client 11 in the execution request information table T7 from the client 11.
The OS pre-preparation number setting unit 42 is a functional unit that records setting values obtained based on information in the execution request information table T7 from the client 11 in the OS pre-preparation number table T8. The OS preliminary preparation number setting unit 42 functions as a priority determination unit that determines the priority of the operating system.
The OS advance preparation status setting unit 43 is a functional unit that records information on the OS started in advance from the computer schedule table T3 in the OS advance preparation status table T9.

In the execution request information table T7, information on the execution environment (OS, CPU) included in the execution request information is represented. In other words, the execution request information table T7 stores information on the OS requested as the OS to be used, and can be used to calculate the ratio.
In the OS pre-preparation number table T8, information on setting values of the type and number of OSs to be started in advance from the contents requested for job execution from the client 11, although not set yet, is represented. The
In the OS pre-preparation status table T9, information on the OS that has been activated in advance and its machine name are displayed correspondingly.

(Operation of the grid system in the third embodiment)
The operation contents of the grid system in the third embodiment will be described below.
In the present embodiment, an OS that is determined not to be allocated by the master host 12 as an OS that has not been prepared for startup even though the usage frequency is high (or an OS that is not prepared) is a provisioning computer 13 that is not scheduled to be allocated. Start.

(1) Setting of OS Advance Preparation Number The OS advance preparation number setting unit 42 derives the set number and records it in the OS advance preparation number table T8.
Based on the information in the execution request information table T7 from the client 11 set by the execution request information setting unit 41 from the request information from the client 11, the set number is derived from the following introduction method. It should be noted that the calculation of the set value is preferably reviewed at regular time intervals so as to be in line with the current status.
Number of advance preparations for each OS =
(Each OS request frequency ratio) × {(Total number of registered machines) × (Machine number ratio constant to be started)}
The “OS request frequency ratio” represents the ratio of the OS (usage OS) requested at the time of execution request, and is calculated from the information of the execution environment accumulated in the execution request information table T7.
“Total number of registered machines” is the total number of provided computers 13.
“Machine number ratio constant to be activated” represents the ratio of the number of machines to be constantly activated with respect to the total number of the provided computers 13. A predetermined value can be used for this ratio constant.

(Example)
A job request from the client 11
Windows2000 is 50% (0.5) of the total
Windows 98 is 30% (0.3) of the total
Linux9 is 20% of the total (0.2)
Machine number ratio constant to be started = 0.3
If it was.

-Number to start Windows2000 beforehand =
0.5 (required ratio of Windows2000) x {50 units x 0.3} = 8

-Number to start Windows 98 in advance 0.3 (Required ratio of Windows 98) x {50 units x 0.3} = 5

-Number of Linux 9 to start in advance 0.2 (Requirement ratio of Linux 9) x {50 units x 0.3} = 3

(2) Setting for Advance Startup The OS advance preparation status setting unit 43 sets advance startup. That is, based on the information in the computer schedule table T3 and the OS preliminary preparation number table T8, the information on the machine to be started and the OS to be started is recorded in the OS preliminary preparation status table T9.
In this setting, a high-performance CPU may be preferentially activated. That is, a high-performance CPU among the available CPUs is targeted, and the performance is gradually improved.

(3) Execution of Advance Startup The execution processing unit 28 sequentially starts up the set OS for the machine to be started up in advance from the information in the OS advance preparation status table T9.
In the allocation process, the allocation determination unit 26 performs the allocation process based on information in the OS preliminary preparation status table T9 (considering allocation including a machine on which the OS has already been started).

  In this embodiment, the OS startup time can be saved, and the providing computer 13 can be used more effectively. An OS with a small number of currently prepared ones is started up in an available computer 13 in advance.

(Other embodiments)
Embodiments of the present invention are not limited to the above-described embodiments, and can be expanded and modified. The expanded and modified embodiments are also included in the technical scope of the present invention.
For example, it is possible to realize a grid system having both the functions of the second and third embodiments (the resource provision time correction function and the OS preparation function).
Further, the tables T1 to T9 and the like can be appropriately divided and integrated. For example, some or all of the contents of the tables T1 to T9 and the resource provider schedule storage unit 14 can be handled as a single table.
Moreover, the information represented in the tables T1 to T9 is an example, and all of the information is not necessarily essential. For example, if the execution environment is uniform, information on the execution environment (OS, CPU) can be omitted from the table.

It is a system configuration figure showing the grid system concerning a 1st embodiment of the present invention. FIG. 3 is a functional configuration diagram illustrating details of a functional configuration of a master host according to the first embodiment. It is a schematic diagram showing an example of execution request information. It is a schematic diagram showing an example of a computer registration table. It is a schematic diagram showing an example of a job execution time table. It is a schematic diagram showing an example of a computer schedule table. It is a schematic diagram showing an example of an allocation candidate table. It is a schematic diagram showing an example of a resource provider schedule. It is a flowchart showing an example of the basic operation | movement procedure of the grid system which concerns on 1st Embodiment. It is a flowchart showing the detail of step S17 of FIG. It is a flowchart showing the detail of step S19 of FIG. FIG. 7 is a schematic diagram showing the computer schedule of FIG. 6 in time series. It is a schematic diagram showing the resource provider schedule of FIG. 8 in time series. It is a system configuration figure showing the grid system concerning a 2nd embodiment of the present invention. It is a system configuration figure showing the grid system concerning a 3rd embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 ... Client, 12 ... Master host, 13 ... Provision computer, 131 ... Provision computer main body, 132 ... Auxiliary storage device, 14 ... Resource provider schedule storage part, 15 ... Data server, 16 ... GPS information transmission part, 21 ... Data Recording unit 22 ... CPU name acquisition unit 23 ... Computer registration information acquisition unit 24 ... Allocation information acquisition unit 25 ... Computer schedule setting unit 26 ... Allocation determination unit 27 ... Resource provision time zone acquisition unit 28 ... Execution Processing unit 31 ... GPS information acquisition unit 32 ... Resource provider information correction unit 41 ... Execution request information setting unit 42 ... Preliminary preparation number setting unit 43 ... Preliminary preparation status setting unit T1 ... Computer registration table T2 Job execution time table, T3 Computer schedule table, T4 Assignment candidate table, T5 Latitude / longitude correspondence table, T6 GP Information table, T7 ... execution request information table, T8 ... advance preparation number table, T9 ... advance preparation status table

Claims (8)

A first table representing resources, processing details, processing allocation start time, and processing allocation end time,
A second table correspondingly representing resources and resource providers;
A third table representing the processing contents and the processing time required,
A fourth table that represents the resource provider, the action type of the resource provider, and the action time zone in the action type;
An execution request information receiving unit that receives execution request information including a processing content, a processing start time, and a processing end time limit;
A table updating unit that updates the first table based on the first, second, third, and fourth tables and the received execution request information;
A process execution unit that causes a resource to execute a process based on the updated first table;
A resource management device comprising: The table updating unit
A time zone calculation unit that calculates an assignable time zone for processing based on the first and fourth tables and the received execution request information;
Based on the processing start time and the required processing time, a scheduled end time calculation unit that calculates a scheduled processing end time;
An assignability determination unit that determines whether or not to assign process contents based on the calculated estimated process end time and the calculated assignable time zone;
The resource management apparatus according to claim 1, further comprising: an additional registration unit that additionally registers, in the first table, processing contents that the allocation determination unit determines that allocation is possible. The first table further includes activation necessity information indicating necessity of activation of resources,
The table update unit further includes an activation necessity determination unit that determines whether the resource activation is necessary based on the calculated allocatable time zone and the fourth table,
The scheduled termination time calculation unit calculates a scheduled processing termination time based on the result of the necessity determination of activation,
The resource management apparatus according to claim 2, wherein the additional registration unit additionally registers activation necessity information in the first table based on a result of the necessity determination for activation. The first and third tables and the execution request information further include operating system identification information,
4. The resource management apparatus according to claim 3, wherein the activation necessity determination unit determines whether the resource activation is necessary based on an operating system correspondence relationship in the first and third tables. . The resource has a main body and an auxiliary storage device;
2. The resource according to claim 1, wherein the resource management device further includes a disconnection control unit that controls disconnection of the auxiliary storage device from the main body, and an activation control unit that controls activation of the main body. Management device. A location information receiving unit for receiving the location information of the resource provider;
A prediction unit that predicts a process allocation start time based on the received position information, and
The table updating unit updates the first table based on the first, second, third, and fourth tables, the received execution request information, and the predicted process allocation start time. The resource management apparatus according to claim 1. The first and third tables and the execution request information further include operating system identification information,
An operating system identification information accumulating unit for accumulating operating system identification information included in the execution request information;
A priority determining unit that determines the priority of the operating system based on the accumulated operating system identification information;
The resource management apparatus according to claim 1, further comprising an activation control unit that controls activation of resources based on the determined priority. A step of receiving execution request information including a processing content, a processing start time, and a processing end time limit;
A first table that stores resources, processing contents, processing allocation start time, and processing allocation end time correspondingly, a second table that stores resources and resource providers correspondingly, and processing A third table for storing the contents and the required processing time correspondingly, the resource provider, the action type of the resource provider, and the action time zone for the action type Updating the first table based on the table of 4 and the received execution request information;
Causing the resource to perform processing based on the updated first table;
A resource management method comprising:

Images