KR20080071177A - Information processing apparatus, computer, resource assigning method and resource assigning program - Google Patents

Abstract

Where a plurality of partitions are dived in a computer and a resource assigned to the partition is used for the execution of data processing, a novel resource assigning technique is automatically provided for each partition to use a suitable quantity of resource in response to a load. In consideration of many cases in which resource using states may be found in advance, a memorizing means is prepared to memorize schedule information which describes an assigning quantity of resource to a timing slot about a term or a time. A resource quantity memorized corresponding to a timing slot to which the present time belongs is acquired from the memorizing means and each partition is controlled to use resources in the acquired resource quantity for the execution of data processing.

Description

Information processing unit, calculator, resource allocation method and resource allocation program {INFORMATION PROCESSING APPARATUS, COMPUTER, RESOURCE ASSIGNING METHOD AND RESOURCE ASSIGNING PROGRAM}

The present invention provides an information processing apparatus for executing information processing, a calculator for dividing the calculator into a plurality of partitions, and performing data processing using resources allocated to each partition, and the information processing apparatus or the calculator. A resource allocation method and a resource allocation program used for realizing the resource allocation method.

Server devices capable of operating a plurality of OSs (operating systems) on one server device have been used.

In such a server device, a plurality of CPUs are mounted, for example, OS1 performs processing as a mail server using 10 CPUs, OS2 performs processing as a business server using 5 CPUs, and OS3 uses 3 CPUs. As in the case where the processing is performed as a batch server, the server device is divided into a plurality of partitions, and data processing is executed using resources such as a CPU to which each partition is allocated.

Conventionally, in such a server device, the administrator always monitors the use state of the resource of the server device, and according to the use state, the distribution of the resource is performed by others' hands.

In addition, a method is provided in which a pool is formed without allocating any amount of resources to any task, and in the case of an emergency, the resources that form the pool are allocated to a task that requires it. In other words, resources are allocated by others.

However, in a task such as a network business, a sudden increase in load occurs. In such a case, if the distribution of resources is to be changed by the hands of others, there is a problem that a large load is imposed on the manager who performs the change of the distribution.

Therefore, recently, as an example of using a CPU as a resource, the CPU utilization rate is monitored, and when the monitored CPU utilization rate exceeds a certain threshold, the CPU is issued by issuing a command to incorporate a new CPU. When the utilization rate of is large, the technique of automatically incorporating a new CPU has been used (see, for example, Patent Document 1).

In addition, as a background art of the present invention, Patent Document 2 described below discloses a server system including a computer for receiving a service and a plurality of servers, wherein the service provided by each server is changed in accordance with a time zone, thereby reducing the quality of service. An invention is described which makes it possible to prevent it.

As a background art of the present invention, in Patent Document 3, a plurality of nodes are connected by a network to form a node group called a cluster, and clustering is performed by dividing the processing contents and assigning them to each node for distributed processing. A method for realizing a clustering service, comprising: providing a virtual server for a hosting service and a virtual server for a clustering service to each of a plurality of physical servers, connecting each clustering service virtual server through a network to form a cluster, and An invention is disclosed in which a time scheduling function is used to allocate only surplus time of a CPU to a virtual server for a clustering service, thereby realizing the clustering service without degrading the performance of the hosting service.

In addition, as a background art of the present invention, Patent Document 4 below discloses the CPU usage time required for each execution on the basis of operation statistics information for a plurality of transaction processing messages injected from a plurality of terminals. An invention is described which achieves a leveling of CPU utilization efficiency of each server by calculating the predicted value and distributing these transaction processing messages to the plurality of servers based on the calculated predicted value and the current CPU utilization rate of the plurality of servers. have.

In addition, as a background art of the present invention, Patent Document 5 below discloses that a resource management server collects a state of use of a resource of a virtual calculator, predicts a state of use of a resource based thereon, and also executes a past virtual calculator. Based on the history, the correlation of the resource usage status of each virtual calculator is calculated, and the resource allocation amount of each virtual calculator is calculated based on the predicted value and the correlation coefficient, so that resource allocation optimization can be realized. The invention to be described is described.

Patent Document 1: Japanese Patent Laid-Open No. 2003-150571

Patent Document 2: Japanese Patent Laid-Open No. 2002-150000

Patent Document 3: Japanese Patent Laid-Open No. 2002-318862

Patent Document 4: Japanese Patent Application Laid-Open No. 2003-296289

Patent Document 5: Japanese Patent Application Laid-Open No. 2004-199561

However, if the administrator is to monitor the usage of the resource and follows the conventional technique of changing the distribution of resources such as the CPU by others' hands, the administrator should always monitor the usage of the resource. There is a problem that there is a problem, and there is a problem that the distribution of resources must be performed by others.

The use of the server device by the user is often increased or decreased at a specific time period or a specific time period, and in such a case, the resource use status can be grasped in advance. Very inefficient.

In addition, if the conventional technology of monitoring the CPU utilization rate and automatically incorporating a new CPU when the monitored CPU utilization rate exceeds a certain threshold value is reached, Since resources are allocated, there is a problem that the resources cannot be allocated to these tasks even if the load of other tasks is increased.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a new resource allocation technique that enables the processing function to reliably use an appropriate amount of resources in accordance with the load.

In addition, in the case of adopting a configuration in which the inside of the calculator is divided into a plurality of partitions and each partition performs data processing using the resources allocated to each partition, in consideration of the fact that the use state of the resources is often found in advance, By pre-scheduling the allocation of resource amounts, in such a case, it is aimed at providing a new resource allocation technique that enables each partition to reliably use the appropriate amount of resources according to the load, without going through others' hands.

[1] Configuration of the information processing apparatus of the present invention

[1-1] First configuration

In order to achieve the above object, the information processing apparatus of the present invention is to be assigned to a processing function according to (1) a processing unit that executes information processing, (2) timekeeping means, and (3) time. A storage means for storing information indicating the resource amount, and (4) a process for determining the amount of resources to be allocated at the time by referring to the storage means in accordance with the time-keeping result by the timekeeping means, and corresponding processing of the determined amount of resources. It is configured to have control means assigned to the function.

[1-2] Second Configuration

Further, in order to achieve the above object, the information processing apparatus of the present invention corresponds to (1) a plurality of processing units executing different processing functions and (2) the amount of resources to be allocated to each of the processing units in accordance with time. And storage means (3) and control means for allocating an amount of resources to be allocated to a certain processing part at the present time with reference to the storage means.

In this configuration, the storage means may store the amount of resources to be allocated to each processing unit for each time zone.

In addition, the information processing apparatus of this invention may be comprised by including the 1st apparatus which has a some process part at least, and one or some terminal apparatus connected to a 1st apparatus.

[2] configuration of a resource allocation method of the present invention

In order to achieve the above object, in the resource allocation method of the present invention, in order to allocate resources required for processing to a processing unit that performs information processing, (1) schedule information indicating an amount of resources to be allocated to the processing unit; (2) determine the amount of resources to be allocated to the processing unit at the current time based on the schedule information, and (3) process to allocate the determined amount of resources to the processing unit.

The resource allocation method of the present invention operates in the information processing apparatus of the present invention employing the above configuration, so that one or more processing functions included in the information processing apparatus of the present invention can reliably use an appropriate amount of resources according to the load. It becomes possible.

[3] Configuration of the calculator of the present invention

In order to achieve the above object, the calculator provided with the present invention divides the inside of the calculator into a plurality of partitions, and adopts a configuration in which data processing is executed using resources allocated to each partition. A storage means for storing schedule information describing that the partition allocates this amount of resources to the time range for this period or time, and (2) the storage means corresponding to the time range to which the current time belongs. The acquisition means which acquires the amount of resources stored, and (3) the control part which controls each partition to perform data processing using the resource of the resource amount acquired by the acquisition means are employ | adopted.

In adopting this configuration, there may be a case where it is provided with judging means for judging whether or not the temporal range to which the current time belongs is changed from the one judged from the previous time. When the judging means is provided, the obtaining means When the judging means judges the change of the temporal range, the resource amount is acquired from the storage means.

Here, each processing means other than the storage means can also be realized by a computer program, which is provided by being recorded on a suitable computer readable recording medium or provided via a network, and installed when the present invention is implemented. The present invention is realized by operating on a control means such as a CPU.

In the calculator having the present invention configured as described above, when the resource amount stored in correspondence with the temporal range to which the current time belongs is obtained from the storage means, each partition uses the resource of the acquired resource amount to perform data processing. Control to run

That is, for a partition using less resource amount than the resource amount acquired from the storage means, it is controlled to newly allocate a resource corresponding to the difference amount, while using more resource amount than the resource amount acquired from the storage means. The partition is controlled to deallocate resources corresponding to the difference amount.

Here, in the case of releasing this allocation, if the resource to be released is being used, control is made to release the allocation of the resource after the use is finished.

In this way, the calculator according to the present invention divides the inside of the calculator into a plurality of partitions, and in advance adopts a configuration in which data processing is executed using the resources allocated to each partition, so as to grasp in advance the state of use of the resources. In consideration of many cases, the allocation of resource amount is scheduled in advance, and the allocation of resources to each partition is controlled in accordance with the scheduling, so that the appropriate amount of each partition is automatically adjusted without load. It is possible to realize that the resources of the system can be used reliably.

Effect of the Invention

According to the present invention, one or more processing functions included in the information processing apparatus can reliably use an appropriate amount of resources according to the load.

From this, according to the present invention, the information processing apparatus which processes information by using one or a plurality of processing functions can efficiently execute the information processing.

Further, according to the present invention, when adopting a configuration in which the inside of the calculator is divided into a plurality of partitions, and each partition executes data processing using the allocated resources, each partition is automatically made without anyone's hands. You will be able to use the right amount of resources to meet your load.

From this, according to the present invention, in the case of adopting a configuration in which the inside of the calculator is divided into a plurality of partitions and each partition executes data processing using the allocated resources, the data processing can be efficiently executed. do.

1 is an example of one embodiment of a server device of the present invention.

2 is an example of schedule information managed by a schedule information table.

3 is an explanatory diagram of schedule information managed by a schedule information table.

4 is an example of a processing flow executed by a schedule information creating unit.

5 is an example of a processing flow executed by a schedule information creating unit.

6 is an example of a processing flow executed by the setting reflecting unit.

7 is an example of the processing flow executed by the setting reflecting unit.

8 is an explanatory diagram of a schedule information creation screen.

9 is an explanatory diagram of an OS selection screen.

10 is an explanatory diagram of a CPU number setting screen.

11 is an explanatory diagram of a CPU number setting end screen.

12 is another example of the processing flow executed by the setting reflecting unit.

13 is another example of the processing flow executed by the setting reflecting unit.

14 shows another embodiment of the server device of the present invention.

15 is another example of the processing flow executed by the setting reflecting unit.

16 is another example of the processing flow executed by the setting reflecting unit.

<Description of the code>

1: server device 2: server management device

3: network 4: network

5: user terminal 10: partition

11: system board 12: service processor

20: schedule information creating unit 21: schedule information transmitting unit

100: operating system 101: application program

102: setting reflecting unit 103: time zone definition information storage unit

104: previous time zone information storage unit 120: schedule information table

121: schedule information register

EMBODIMENT OF THE INVENTION Hereinafter, this invention is demonstrated in detail according to embodiment.

Fig. 1 shows an example of one embodiment of the server device 1 of the present invention.

As shown in this figure, the server apparatus 1 of this embodiment is connected with the server management apparatus 2 which manages its own apparatus through the network 3, and is connected via the network 4 The user terminal 5 is subjected to a process of providing a plurality of server functions, for example, as a function as a mail server, a function as a work server, and a function as a deployment server.

In order to realize the provision of these server functions, the server apparatus 1 of this embodiment implements the operating system 100-i (i = 1 to 3) and the application program 101- which are provided for realizing each server function. i) According to (i = 1 to 3), the device is divided into a plurality of partitions 10-i (i = 1 to 3), and each partition 10-i has a hardware allocated to the CPU. Predetermined server functions according to the operating system 100-i (i = 1 to 3) and the application programs 101-i (i = 1 to 3) to be used as resources and deployed in the magnetic partition 10-i. Will run

Here, reference numerals 11-i (i = 1 to 4) shown in the drawings are system boards included in the server device 1 of the present embodiment, and are assigned to each operating system 100-i (a round of the drawings). Display) and memory. In addition, the code | symbol 12 is a service processor with which the server apparatus 1 of this embodiment is equipped, and cooperates with the server management apparatus 2, and manages the server apparatus 1 of this embodiment.

In this configuration, in order to realize the present embodiment in the server device 1 of the present embodiment, the service processor 12 schedules information on the number of CPUs allocated to the respective operating systems 100-i. And a schedule information register 121 for receiving schedule information transmitted from the server management device 2 and registering the schedule information to the schedule information table 120. On the other hand, each partition 10-i is a setting reflecting unit 102-i (i) for setting the number of CPUs used in its partition 10-i according to the schedule information managed by the schedule information table 120. = 1 to 3).

In addition, the server management device 2 includes a schedule information creating unit 20 for creating schedule information registered in the schedule information table 120 in order to execute management of the server device 1 of the present embodiment; A schedule information transmitter 21 is provided to transmit the schedule information created by the schedule information generator 20 to the service processor 12 and register the schedule information in the schedule information table 120.

2 shows an example of schedule information managed by the schedule information table 120.

As shown in Fig. 2, the schedule information table 120 corresponds to each set time zone, and in this time zone, the number of CPUs allocated to each operating system 100-i, and each operating system 100-i. ) Stores information that manages the number of CPUs that are not allocated and form a pool.

In the example of the schedule information shown in FIG. 2, the server device 1 of this embodiment is comprised of three partitions, namely, the mail server OS1, the work server OS2, and the deployment server OS3. In addition, when the server apparatus 1 has 24 CPUs as a whole,

Mail server (OS1) AM 9:00 AM to 10:00 AM: High load 10 CPU

 PM 12:00-PM 3:00: High load 8 CPU

 Other: low load 5 CPU

Business Server (OS2) 10:00 AM to PM 5:00 PM: High Load 10 CPU

 Other: low load 6 CPU

Deployment Server (OS3) AM 1:00 AM to 3:00 AM: High Load 10 CPU

 Other: low load 2 CPU

It is assumed that the CPU allocation is performed.

That is, in a time zone in which a high load is expected for a certain function, the CPU number is scheduled to reduce the load on the operating system 100-i corresponding to the high-loading function by allocating more CPUs for the function. . On the other hand, in a time zone in which a low load is expected for a function, the number of CPUs can be allocated to the other high load operating system 100-i by reducing the number of CPU allocations for that function. To schedule.

Here, whether or not the operating system 100-i is a high load or a low load in a certain time period can be determined using the utilization rate of the CPU in that time period.

In the example of the schedule information shown in Fig. 2, although the time zone is set in units of one hour, it is also possible to set the time zone in units of arbitrary time, such as 40 minutes. In addition, in the example of the schedule information shown in FIG. 2, it is assumed that the time zone common to the three operating systems 100-i is set. However, each time zone · for each function / operating system 100-i is set. It is also possible to set each time division.

3 is a diagram showing the schedule shown in FIG. 2 on the horizontal axis of time and the number of CPUs allocated on the vertical axis. The number of CPUs to be allocated or to be allocated to each time zone can be easily recognized by displaying the format as shown in FIG. 3.

In this way, the schedule information table 120 manages schedule information describing that each operating system 100-i allocates a certain number of CPUs in a certain time period.

4 and 5 show an example of a processing flow executed by the schedule information creating unit 20 included in the server management apparatus 2, and settings provided by the server apparatus 1 in FIGS. 6 and 7. An example of the processing flow executed by the reflecting unit 102-i is shown.

Next, the processing executed by the server device 1 of the present embodiment shown in FIG. 1 according to these processing flows will be described in detail.

First, according to the processing flow of FIG. 4 and FIG. 5, the process which the schedule information preparation part 20 with which the server management apparatus 2 comprises is demonstrated.

If the schedule information creation unit 20 included in the server management device 2 has a request to create schedule information registered in the schedule information table 120 from an administrator who operates the server management device 2, first, in step 10, The schedule information currently managed from the schedule information table 120 is read. The read schedule information is displayed in the form of a graph display etc. on the display which omit illustration in FIG. 1 as needed.

8 is a diagram illustrating an example of a schedule information creation screen displayed on a display. That is, the schedule information preparation unit 20 displays the schedule information creation screen as shown in FIG. 8 on the display, and displays the schedule information read from the schedule information table 120 in a graphical form. It is. The graph displayed on the schedule information creation screen of FIG. 8 displays schedule information created using the schedule information creation screen in correspondence with the time and the number of CPUs, and is equivalent to the graph shown in FIG.

Here, the schedule information creation screen shown in FIG. 8 is operated at the time of setting request of OS selection button operated at the time of the selection request of the operating system 100-i, and the number of CPUs allocated to the operating system 100-i. It has a CPU number setting button.

In addition, although the schedule information creation screen shown in FIG. 8 shows the example which does not display a graph about the number of CPUs (the number of CPUs which make up a pool) which are not allocated to the operating system 100-i, this CPU is shown. It is also possible to graph the numbers.

Subsequently, in step 11, it is detected whether the OS selection button displayed on the schedule information creation screen has been operated. When it is determined that the OS selection button has been operated, the selection request of the operating system 100-i is issued, so that the schedule information creating unit 20 waits for the selection request of the operating system 100-i to be issued. If a selection request of the operating system 100-i is issued, the flow advances to step 12 to display an OS selection screen for use in selecting the operating system 100-i, and accepts an input thereon for the operating system to be processed. Select (100-i).

That is, as shown in FIG. 9, the OS selection screen used for the selection of the operating system 100-i is displayed, and the input of the operating system 100-i is selected by selecting the operating system 100-i. In the OS selection screen, since the OS to be selected is displayed in a list, the operator instructs the OS to be selected on the screen and operates the selection button.

Subsequently, after the OS is selected, it is detected in step 13 whether the CPU number setting button displayed on the schedule information creation screen has been operated. When the CPU number setting button is operated, a CPU number setting request is issued. The schedule information preparation unit 20 waits for a request for setting the number of CPUs to the operating system 100-i to be processed, and if a request for setting the number of CPUs is issued, proceeds to step 14 to set the number of CPUs. Displays the CPU number setting screen used for. Then, by accepting the input from the CPU number setting screen, the correspondence relationship between the time zone and the CPU number for the operating system 100-i to be processed is input. When there is an input request for setting the corresponding relationship, the schedule information creating unit 20 updates the schedule information read from the schedule information table 120 in response to the setting request.

10 is a diagram illustrating a display example of a CPU number setting screen. That is, as shown in FIG. 10, the schedule information preparation part 20 displays the CPU number setting screen used for setting of the CPU number with respect to the operating system 100-i of processing object, and accepts the input about it. do. The CPU number setting screen has an input field for setting a time zone and an input field for inputting the number of CPUs to be set. The operator inputs a correspondence between the time zone and the number of CPUs for the operating system 100-i to be processed from these input fields, and operates a setting button displayed on the CPU number setting screen. The schedule information creation unit 20 regards the operation of the setting button as a setting request of the correspondence input by the operator, and updates the schedule information read from the schedule information table 120 with the input information in response to the setting request. It is.

Subsequently, in step 15, the schedule information preparation unit 20 determines whether the end button displayed on the CPU number setting screen has been operated, that is, whether or not a request for ending the CPU number setting has been issued. When the end button is not operated, that is, when the setting button of the CPU number setting screen is operated, the process returns to step 14 to continue updating the schedule information.

On the other hand, when the schedule information preparation unit 20 detects that the end button of the CPU number setting screen is operated in step 15 and a termination request for the CPU number setting has been issued, the schedule information preparation unit 20 proceeds to step 16 to request processing continuing. By displaying the CPU number setting end screen used for the acceptance or the acceptance of the process termination request, and accepting an input thereto, it is determined whether there is a request for processing continuation or a request for processing termination.

11 is a diagram illustrating an example of the CPU number setting end screen displayed on the screen. As shown in Fig. 11, a CPU number setting end screen for use in accepting a process continuation request or a process end request is displayed, and by accepting an input therefor, there is a request for processing continuation or a request for end of processing. The schedule information preparation unit 20 determines whether or not there is.

When it is determined that the request for processing continuation has been issued in accordance with the determination process of step 16, the schedule information creating unit 20 returns to step 11 to continue updating the schedule information.

On the other hand, when it is judged that the request for the end of processing has been issued in accordance with the determination process of step 16, the schedule information creating unit 20 proceeds to step 17 to step schedule information used for displaying a graph on the schedule information creation screen. The process is changed to the one updated in the process of step 14, and the process of graph-displaying the newly created schedule information is performed.

Subsequently, at step 18, the schedule information preparation unit 20 obtains the total number of CPUs assigned to each operating system 100-i for each time period, and it is larger than the total number of CPUs mounted on the system board 11. Determine whether it is small. The schedule information preparation unit 20 checks whether there is a time zone in which the CPU does not form a pool by this determination.

Subsequently, in step 19, the schedule information preparation unit 20 determines whether or not there is a time zone in which there is no CPU forming the pool, in accordance with the processing in step 18. If it is determined that there is a time zone that does not have a CPU, the process proceeds to step 20. For the schedule information graphed on the schedule information creation screen, the time zone without a CPU forming the pool is explicitly displayed, and then the process returns to Step 11. Goes. This instructs the administrator who creates the schedule information to reset the number of CPUs so that there is no time zone without the CPUs forming the pool.

On the other hand, when it is determined in step 19 that there is no time zone without the CPU forming the pool in accordance with the process of step 18, the process proceeds to step 21, and the schedule information creating unit 20 provides the schedule information transmitting unit 21 with respect to the schedule information transmitting unit 21. The transmission of the newly created schedule information is instructed, and the process ends.

In response to this transmission instruction, the schedule information transmitter 21 transmits schedule information with a transmission instruction to the service processor 12. In response to the transmission of the schedule information, the schedule information registration unit 121 included in the service processor 12 receives the schedule information transmitted from the schedule information transmission unit 21 and registers it in the schedule information table 120.

In this way, the schedule information as shown in FIG. 2 is registered in the schedule information table 120.

Next, the process which the setting reflecting part 102-i with which the server apparatus 1 comprises is performed according to the process flow of FIG. 6 and FIG. 7 is demonstrated.

When the setting reflecting unit 102-i is activated by reaching a control period set to, for example, a 5-minute period, first, in step 30, the present time is obtained by referring to a clock (not shown), and in the following step 31, the schedule information table The schedule information is read from 120.

Subsequently, according to the schedule information read in step 32, the operating system 100-i deployed in the self-operating system 100-i (partition 10-i to which it belongs) assigned to the time zone to which the current time belongs. The number of set CPUs of i)) is extracted, and in the following step 33, the number of CPUs currently used by the self-operating system 100-i is acquired.

Subsequently, in step 34, the extracted set number of CPUs is compared with the acquired number of used CPUs, and in step 35, based on the comparison result, whether the set CPU number and the number of used CPUs are the same value, in other words, the set number of CPUs. It is determined whether the number is larger than the number of used CPUs or whether the number of set CPUs is smaller than the number of CPUs used.

When the comparison result that the set number of CPUs and the number of used CPUs is the same value is obtained in accordance with the determination process of step 35, the setting reflecting unit 102-i does nothing and finishes the processing.

On the other hand, when the comparison result that the number of set CPUs is larger than the number of used CPUs is obtained in accordance with the judgment processing of step 35, the setting reflecting unit 102-i proceeds to step 36, where the CPU incorporation command (operating system ( Is issued to 100-i), and a CPU corresponding to the difference between the number of set CPUs and the number of CPUs used is incorporated into the self-operating system 100-i. (Assignment), the process ends.

On the other hand, when the number of set CPUs is smaller than the number of used CPUs according to the judgment processing of step 35, the setting reflecting unit 102-i proceeds to step 37 and the CPU assigned to the self-operating system 100-i. Determine if the job is running on the server.

When it is determined that the job is not being executed on the CPU in accordance with the judgment processing of step 37, since the number of CPUs corresponding to the difference between the number of used CPUs and the set CPUs can be separated, the process proceeds to step 41. By issuing a CPU disconnection command (a command included in the operating system 100-i), the number of CPUs corresponding to the difference between the number of used CPUs and the set CPUs is separated from the self-operating system 100-i. It accumulates in the pool, and the process ends.

On the other hand, when it is judged that the job is being executed on the CPU in accordance with the judgment processing of step 37, the setting reflecting unit 102-i proceeds to step 38 to determine whether or not there is a CPU that has finished executing the job. .

When there is a CPU that has finished execution of the job in accordance with the determination process of step 38, the setting reflecting unit 102-i proceeds to step 39, where a number corresponding to the difference between the number of used CPUs and the number of set CPUs is obtained. By issuing a CPU disconnection command until the CPU is detached, the CPU that is not executing a task from the self-operating system 100-i is separated and accumulated in the pool.

In step 40, it is determined whether or not the CPU corresponding to the difference between the number of used CPUs and the set CPUs has been separated. If not, the process returns to step 38 to execute the job. Wait for the CPU to exit. On the other hand, when the number of CPUs corresponding to the difference between the number of used CPUs and the set CPUs is separated, the setting reflecting unit 102-i ends the processing.

Thus, in this embodiment, when the several operating system 100-i operates in the server apparatus 1, when providing several server function, the CPU which each operating system 100-i uses is used. The allocation of the number is scheduled in advance, and control is made to allocate a CPU to each operating system 100-i in accordance with the scheduling.

Since there are many things which can be grasped | ascertained beforehand about the temporal fluctuation | variation of the load of each operating system 100-i, according to the structure of this embodiment, each operating system 100-i is automatically made through no hand. You can use the appropriate number of CPUs to match the load.

12 and 13 show another example of the processing flow executed by the setting reflecting unit 102-i. Here, the steps of executing the same processing as the processing flow of Figs. 6 and 7 are described with the same step numbers.

6 and 7, when the setting reflecting unit 102-i acquires the current time in step 30, the setting reflecting unit 102-i immediately proceeds to step 31 to read schedule information from the schedule information table 120. FIG. I'm trying to.

In contrast, in the case of executing the processing flows of FIGS. 12 and 13, when the setting reflecting unit 102-i acquires the current time in step 30, in the subsequent step 100, the time zone from the time zone processed in the previous control cycle is obtained. Is determined, and the process proceeds to step 31 only when the time zone is changed to read schedule information from the schedule information table 120.

For example, in the case of providing a protocol for setting a time zone every hour, it is possible to judge whether or not the time zone has been changed from the time zone processed in the previous control cycle by simply looking at the current time.

In this case, if the time zone is not changed, there is no change in the number of set CPUs. Therefore, it is not necessary to proceed to the process after step 31.

Therefore, when executing the processing flows of FIGS. 12 and 13, when the setting reflecting unit 102-i acquires the current time in step 30, the time zone is changed from the time zone processed in the previous control cycle in the subsequent step 100. When it is judged whether or not it has been changed, and when it is determined that the time zone has not changed, the processing is performed so as to immediately terminate the processing without proceeding to the processing after step 31.

By executing this processing flow of Figs. 12 and 13, the setting reflecting unit 102-i is finished without executing unnecessary processing.

On the other hand, instead of establishing a protocol for setting a time zone every hour, when allowing the administrator who creates schedule information to set the time zone freely, the process is performed in the previous control cycle only by viewing the current time. It is not possible to determine whether the time zone has changed from one time zone.

In this case, as shown in Fig. 14, each partition 10-i includes a time zone definition information storage unit for copying and storing information on a time zone defined by schedule information managed by the schedule information table 120 ( 103-i) (i = 1 to 3) and a previous time zone information storage unit 104-i (i = 1 to 3) for storing information on the time zone processed in the previous control cycle.

The setting reflecting unit 102-i included in the server device 1 executes the processing flow of FIGS. 15 and 16. Here, the steps of executing the same processing as the processing flow of Figs. 6 and 7 are described with the same step numbers.

That is, when allowing the administrator who creates schedule information to freely set the time zone, the setting reflecting unit 102-i displays the current time in step 30, as shown in the processing flow of FIGS. 15 and 16. Next, subsequent to this, in step 200, the information of the time zone processed in the previous control cycle is acquired from the previous time zone information storage unit 104-i. Subsequently, in step 201, a time zone to which the current time belongs is specified by referring to the time zone definition information storage unit 103-i.

Subsequently, in step 202, on the basis of the information of the time zone acquired in step 200 and the information of the time zone specified in step 201, it is determined whether or not the time zone has changed from the time zone processed in the previous control cycle, and only when the time zone has changed. Proceeding to step 31, the schedule information is read from the schedule information table 120.

Then, before ending the process, in step 203, the time zone information storage unit 104-i records the time zone information (information of the time zone specified in step 201) processed in the current control cycle, The process ends.

In this way, when allowing the administrator who creates schedule information to freely set the time zone, the setting reflecting unit 102-i performs the processing flow in the previous control cycle by executing the processing flows of FIGS. 15 and 16. It is determined whether or not the time zone is changed from one time zone. If the time zone is not changed, the processing is performed to immediately end the processing without proceeding to the processing after step 31.

By executing this processing flow of Figs. 15 and 16, the setting reflecting unit 102-i is finished without executing useless processing.

Although it demonstrated according to embodiment shown, this invention is not limited to this. For example, the present invention is realized by scheduling an allocation of resources for the processing function in the form described above, with an information processing apparatus adopting a configuration of processing information using one or a plurality of processing functions. Even if it is, it can be applied as it is.

In the above-described embodiment, the application to the server device has been described as a specific example. However, the present invention is not only applicable to the server device, but can be applied to a general calculator as it is.

In addition, in the embodiment described above, the form in which the functions are separated into the server apparatus and the server management apparatus has been described. However, the functions and the processes executed in these apparatuses do not necessarily need to be separated. All may be completed in a single device. On the other hand, even if each function is isolate | separated between a some apparatus, it is also possible to make the distribution into the form different from the said embodiment suitably.

The present invention can be applied to an information processing apparatus employing a configuration in which information is processed using one or a plurality of processing functions, and the processing functions can be reliably used in an appropriate amount of resources according to the load. To realize.

The present invention can be applied to a calculator employing a configuration in which the inside of the calculator is divided into a plurality of partitions, and each partition executes data processing using resources allocated to each partition. It ensures that partitions can use the right amount of resources to meet their load.

Claims (11)

A processing unit that performs information processing; Means of revelation, Storage means for storing information indicating the amount of resources to be allocated to the processing function according to the time; Control means for determining the amount of resources to be allocated to the time with reference to the storage means and allocating the determined amount of resources to the corresponding processing function in accordance with the result of time counting by the time means; An information processing apparatus comprising: a. A plurality of processing units each executing a different processing function; Storage means for storing the amount of resources to be allocated to each of the processing units in correspondence with time; An information processing apparatus comprising: a control means for allocating an amount of resources to be allocated to a processing unit at a current time with reference to the storage means. The information processing apparatus according to claim 2, wherein in said information processing apparatus, said storage means stores an amount of resources to be allocated to each processing unit for each time zone. The information processing apparatus according to claim 2, wherein the information processing apparatus includes at least a first apparatus having the plurality of processing units, and one to a plurality of terminal apparatuses connected to the first apparatus. In the method of allocating resources required for processing to a processing unit that performs information processing, Referring to schedule information indicating a resource amount to be allocated to the processing unit; Determining an amount of resources to be allocated to the processing unit at a current time, based on the schedule information; Allocating the determined amount of resources to the processing unit. Resource allocation method comprising the. A calculator that divides the calculator into a plurality of partitions, and executes data processing using resources allocated to each partition. Storage means for storing, for each partition, schedule information describing that this amount of resource is allocated to a time range for this period or time; Acquisition means for acquiring, from the storage means, the amount of resources stored in correspondence with the temporal range to which the current time belongs; Control means for controlling each partition to execute data processing using resources of the resource amount acquired by the acquisition means Calculator comprising: a. The resource amount acquired by the said acquiring means according to Claim 6, wherein the control means controls to newly allocate a resource corresponding to the difference amount to a partition using a resource amount smaller than the resource amount acquired by the acquiring means. A partition for controlling a partition that uses a larger amount of resources so as to deallocate a resource corresponding to the difference amount. 8. The calculator according to claim 7, wherein the control means controls to release the allocation of the resource after the use thereof, when the release target resource is used when the allocation is released. . The apparatus according to any one of claims 6 to 8, further comprising determining means for determining whether or not the temporal range to which the current time belongs belongs to a change from a previous judgment. And said acquiring means acquires a resource amount from said storage means when said determining means determines the change of temporal range. A resource allocation method executed in a calculator that divides a calculator into a plurality of partitions, and performs data processing using resources allocated to each partition. For each partition, the amount of resources stored in correspondence with the temporal range to which the current time belongs is stored from storage means for storing schedule information describing that this amount of resource is allocated to the temporal range for this period or time. Acquisition process, Controlling each partition to execute data processing using the resources of the acquired resource amount Resource allocation method comprising the. A resource allocation program that is implemented in a calculator that divides a calculator into a plurality of partitions and executes data processing using resources allocated to each partition. For each partition, the amount of resources stored in correspondence with the temporal range to which the current time belongs is stored from storage means for storing schedule information describing that this amount of resource is allocated to the temporal range for this period or time. Processing to get, A process of controlling each partition to execute data processing using the resources of the acquired resource amount Resource allocation program for running the program on a computer.

Images