JP6676400B2 - Information processing device control device, information processing device control method, and program - Google Patents

Description

  The present invention relates to a control device for an information processing device, a control method for the information processing device, and a program.

  In order to cope with an increase in the amount of arithmetic processing, a method of performing arithmetic processing by a plurality of servers is generally known. In this case, for example, when the amount of calculation processing in a certain server, in other words, the value of a CPU (central processing unit) load exceeds a predetermined value, control is performed to start another server and distribute the calculation processing to the started server. ing.

  However, starting the server involves steps such as starting the hardware of the server, starting an operating system (hereinafter also referred to as “OS”), and starting an application (hereinafter also referred to as “APP”). Therefore, a certain amount of time is required from when the server is determined to be started to when the server can be actually operated. In other words, since a time lag occurs, a period in which the CPU load on the server that was performing the arithmetic processing becomes excessive may occur, and the arithmetic processing may be delayed.

  In order to solve this problem, by comparing the current value of the CPU load with a relatively short-term predicted value and a relatively long-term predicted value, the number of application server startups is determined, and the above-described time lag and the like are obtained. There has been proposed a technique for suppressing the occurrence of phenomena (for example, see Patent Document 1).

JP 2011-232968 A

  As described above, in the technology described in the conventional patent document 1, the accuracy of prediction is insufficient, and there is a possibility that the number of activated servers obtained may be excessive or insufficient. For example, if the obtained number of activated servers is excessive, the amount of calculation processing with respect to the number of activated servers is small, so that it was necessary to continue the operation in an inefficient state or stop the activated servers again. .

  If the server is started and stopped frequently as described above, thermal stress is repeatedly applied to the electronic components mounted on the server, and there is a possibility that a problem such as a crack may occur. That is, when the server is started and the arithmetic processing is performed, the electronic components expand due to the heat generated by the electronic components as the temperature rises. On the other hand, when the server is stopped, the generation of heat in the electronic components and the like also stops, and the expanded electronic components shrink as the temperature decreases. Since the degree of the expansion and shrinkage varies depending on the properties of the material, a place where a tensile force or a compressive force acts to concentrate stress (thermal stress) occurs in the electronic component. When this thermal stress is repeatedly applied, cracks and the like may be generated in the above-described locations.

  Further, if the number of activated servers is insufficient, the amount of arithmetic processing with respect to the number of activated servers is large. Therefore, as described above, there is a problem that the CPU load of the server becomes excessive and the arithmetic processing may be delayed. there were.

  The present invention has been made in order to solve the above-described problems, and a control device of an information processing device, a control method of an information processing device, which can suppress occurrence of a defect by improving prediction accuracy, And to provide the program.

In order to achieve the above object, the present invention provides the following means.
A control device for an information processing device according to a first aspect of the present invention is a control device for an information processing device that controls the operating number of a plurality of information processing devices connected to a plurality of terminal devices via a network, A first prediction method based on an acquisition unit that acquires target data that changes in a time series from the information processing device and stores the target data in a predetermined storage unit, and the target data within a predetermined past time extracted from the storage unit; A first prediction unit that calculates a first prediction value of the target data by using a second prediction method different from the first prediction method based on the target data within the past predetermined time. A second prediction unit for calculating a second prediction value of the target data, and comparing with a predetermined threshold to determine whether to use one of the first prediction value and the second prediction value. The first schedule determined Using one of the value and the second predicted value, and the value of the target data at the time of calculation, calculate the required number of operating information processing apparatuses after a certain period of time, or calculate the number of starting or stopping information processing apparatuses. And an estimating unit for performing the calculation.

  A method for controlling an information processing device according to a second aspect of the present invention is a method for controlling an information processing device that controls the number of operating information processing devices connected to a plurality of terminal devices via a network, An acquisition step of acquiring target data that changes in time series from the information processing device and storing the target data in a predetermined storage unit, and a first prediction method based on the target data within a predetermined past time extracted from the storage unit A first prediction step of calculating a first prediction value of the target data by using a second prediction method different from the first prediction method based on the target data within the past predetermined time. A second prediction step of calculating a second prediction value of the target data, and comparing with a predetermined threshold value to determine whether to use one of the first prediction value and the second prediction value. , Judge Using one of the obtained first predicted value and the second predicted value, and the value of the target data at the time of calculation, the required number of operating information processing devices after a certain period of time is calculated, or Estimating the number of units to be started or stopped.

  A program according to a third aspect of the present invention is a program that causes a computer to function as a control device of an information processing device that controls the number of operating information processing devices connected to a plurality of terminal devices via a network. An acquisition function of acquiring target data that changes in time series from the information processing device to the computer and storing the target data in a predetermined storage unit, based on the target data within a predetermined time in the past extracted from the storage unit, A first prediction function of calculating a first prediction value of the target data using a first prediction method, and a second prediction function different from the first prediction method based on the target data within the past predetermined time. A second prediction function of calculating a second prediction value of the target data using the prediction method of the above, and comparing the first prediction value and the second prediction value by comparing with a predetermined threshold value A determination is made as to whether to use one of them, and using one of the determined first predicted value and the second predicted value and the value of the target data at the time of calculation, after a predetermined time of the information processing apparatus, And an estimating function of calculating the required number of units to be operated or the number of units to be started or stopped.

  According to the first mode, the second mode, and the third mode of the present invention, when calculating the required number of operating information processing apparatuses after a certain period of time, or calculating the number of starting or stopping information processing apparatuses, , One of the first predicted value and the second predicted value determined by comparing with the predetermined threshold value, and the value of the target data at the time of calculation are used. By doing so, a prediction value that can easily improve the prediction accuracy can be selected from the first prediction value and the second prediction value, based on the determination compared with the predetermined threshold value. Therefore, compared to the calculation method described in Patent Literature 1 in which the current value of the CPU load is compared with the relatively short-term predicted value and the relatively long-term predicted value, the prediction accuracy can be easily improved.

  In the first, second, and third aspects of the invention, the first predicted value is a predicted value of the target data after a first predetermined time, and the second predicted value Is a predicted value of the target data after a second predetermined time different from the first predetermined time, and the predetermined threshold is a time required for starting the information processing apparatus (hereinafter referred to as “start time”). The estimating unit compares a difference between the activation time and the first predetermined time and a difference between the activation time and the second predetermined time, and calculates the first predicted value and the It is preferable that it is determined that a predicted value having a small difference from the activation time among the second predicted values is used.

  As described above, in the determination compared with the predetermined threshold value, the difference between the startup time and the first predetermined time and the difference between the startup time and the second predetermined time are compared, and it is determined that the prediction value having the small difference is used. By doing so, it becomes easy to select a predicted value from which the prediction accuracy is easily improved from the first predicted value and the second predicted value.

In the first aspect, the second aspect, and the third aspect of the invention, it is preferable that the activation time differs according to the type of the information processing apparatus.
By making the length of the start-up time different according to the type of the information processing device, the first prediction value and the second prediction value are selected according to the type of the information processing device at the time of selection. This makes it easier to select a predicted value that easily improves the prediction accuracy.

  In the first aspect, the second aspect, and the third aspect of the invention, the activation time is a time required for the activation corresponding to at least one of an OS and an APP mounted on the information processing apparatus. Preferably they are different.

  In this way, by setting the length of the startup time to be different depending on at least one of the OS and the APP installed in the information processing apparatus, the selection of the first predicted value and the second predicted value makes the OS And APP can be selected according to the APP, and a prediction value that can easily improve the prediction accuracy can be further easily selected.

  In the first aspect, the second aspect, and the third aspect of the invention, the activation time is a time required for the activation corresponding to a communication speed in the network connected to the information processing apparatus. Preferably they are different.

  In this way, by making the length of the startup time different according to the communication speed in the network connected to the information processing apparatus, the communication speed in the network can be reduced when selecting the first predicted value and the second predicted value. Selection can be made in accordance with this, and it becomes easier to select a predicted value that can easily improve the prediction accuracy.

  In the first aspect, the second aspect, and the third aspect of the invention, the activation time may be a time zone in which the estimation unit makes a determination, and may use the information processing device at the time of the determination. It is preferable that the time required for the activation be different depending on at least one of the persons having the possibility.

  As described above, the first predicted value and the second predicted value are set differently by making at least one of the time period in which the length of the activation time is determined and the number of persons who may use the information processing apparatus at the time of the determination different. The time zone in which the determination is made when selecting the predicted value of the and the number of people who may use the information processing device at the time of the determination can be selected in accordance with the number of predicted values that can easily improve the prediction accuracy. It is easier to select.

  According to the control device for the information processing device of the first aspect of the present invention, the control method for the information processing device of the second aspect, and the program of the third aspect, the first device determined by comparing with the predetermined threshold value Using one of the predicted value and the second predicted value, and the value of the target data at the time of calculation, the required number of operating information processing apparatuses after a certain period of time is calculated, or the number of information processing apparatuses to be started or stopped is calculated. As a result, there is an effect that the occurrence of a defect can be suppressed by improving the prediction accuracy.

FIG. 1 is a diagram illustrating a schematic configuration of a power supply system according to a first embodiment of the present invention. FIG. 2 is a schematic diagram illustrating a state where distributed operation is performed in the power supply system of FIG. 1. It is a flow chart explaining control at the time of performing degeneration operation. It is a flow chart explaining control at the time of performing degeneration operation. FIG. 2 is a schematic diagram illustrating a state in which a degenerate operation is performed in the power supply system of FIG. 1. It is a flow chart explaining control at the time of shifting from degeneration operation to decentralized operation.

  Hereinafter, a control device, a control method, and a program of an information processing apparatus according to an embodiment of the present invention will be described with reference to FIGS. In the present embodiment, the control device of the information processing apparatus according to the present invention includes an ICT device (information processing device) 25 that performs processes such as information processing and information communication, and a power supply device 10, an air conditioner 35, and a lighting device 45. A description will be given by applying to an example of an integrated control device (control device) 60 in the information processing system 1 including related devices attached to the ICT device 25.

  As shown in FIG. 1, the information processing system 1 includes a power supply device 10, a current distribution device 15, a power supply control device 20, an ICT device 25, an ICT control device 30, an air conditioner 35, and an air conditioner control device 40. , A lighting device 45, a lighting control device 50, an integrated control device 60, and a monitoring control device 70.

  The power supply device 10 supplies power to the ICT device 25, and includes a plurality of power supply units 11 that receive a supply of commercial power and output a DC current. The power supply unit 11 is configured to control the output / stop of the DC current based on a command input from the power supply control device 20. The power supply device 10 has a function of acquiring operation information of the plurality of power supply units 11 and outputting the acquired information to a power supply control device 20 described later.

  Note that a method of converting commercial power in the power supply unit 11 into a DC current, a specific control method of output / stop of the DC current in the power supply unit 11, a method of connecting a plurality of power supply units 11 in the power supply device 10, and the like are described. A known method can be used, and there is no particular limitation. In the present embodiment, the power supply unit 11 is applied to an example in which commercial power, which is an alternating current, is converted into a direct current having a desired voltage used for the operation of the ICT device 25. An AC current having a desired voltage may be supplied instead of the DC current depending on the characteristics of the load device, and the mode of the current is not particularly limited.

  The current distribution device 15 distributes the DC current supplied from the power supply device 10 to a plurality of ICT racks 26 in which the ICT devices 25 are stored. The current distribution device 15 is provided with a switch 16 that performs DC current supply / non-supply control (hereinafter also referred to as “ON / OFF control”) for each post in accordance with a command from the power supply control device 20. . The switch 16 also has a function of measuring the power consumption of the ICT device 25 stored in the ICT rack 26 connected to each post. The power consumption measured by the switch 16 is output to the power supply control device 20. The switch 16 may be of a known type, and is not particularly limited.

  Further, the current distribution device 15 is provided with a protection element (not shown) such as a fuse or a breaker for each post in order to prevent an overcurrent due to a short circuit accident or the like. By installing this protection element, the influence of the system including the ICT device 25 in which the short-circuit accident has occurred (hereinafter also referred to as “accident system”) includes the ICT device 25 which continues other normal operations. It can be prevented from spreading to the system (hereinafter also referred to as “healthy system”).

  The power supply control device 20 controls the power supply ON / OFF for each post of the current distribution device 15 and the command to control the power supply ON / OFF for each outlet of the outlet bar 27 based on the command from the integrated control device 60. It outputs a command to control the operation / stop of the power supply unit 11 of the power supply device 10 and a command to control the operation / stop of the power supply device 10.

  The power supply control device 20 is a computer system having a CPU (central processing unit), a ROM, a RAM, a hard disk, an input / output interface, and the like. The control program stored in the ROM or the like causes the CPU to function at least as the arithmetic processing unit 21 and causes the input / output interface or the like to function at least as the acquisition unit 22.

  The arithmetic processing unit 21 performs arithmetic processing for controlling the operation of the power supply device 10, the switch 16 of the current distribution device 15, and the switch 27A provided inside the outlet bar 27, and generates a command for controlling the operation. It is also a thing. Details of the arithmetic processing and the like in the arithmetic processing unit 21 will be described later.

  The acquisition unit 22 outputs the operation information of the plurality of power supply units 11 output from the power supply device 10, the power consumption information of each post output from the current distribution device 15, and the power consumption of each outlet bar 27. This is an interface to which power information is input. The acquisition unit 22 is also an interface that outputs various commands generated by the arithmetic processing unit 21.

  The ICT device 25 performs information processing such as calculation, communication processing such as reception / transmission of information, and the like. Operations such as processing in the ICT device 25 are controlled based on a command input from the ICT control device 30. One or more ICT devices 25 are mounted on an ICT rack 26, and a DC current is supplied from an outlet bar 27 provided on the ICT rack 26. The above-described information processing, communication processing, and the like correspond to the workload in the claims.

  The outlet bar 27 is connected to the wiring distributed by the current distributor 15. The switch 27A inside the outlet bar 27 has a function of controlling ON / OFF of power supply to the ICT device 25 for each outlet based on a command input from the power supply control device 20. Further, a function of measuring current consumption and power is provided for each outlet, and information on the measured current consumption and power is output to the power supply control device 20. As described above, the outlet bar 27 uses what functions as a so-called intelligent outlet.

  As shown in FIG. 2, the plurality of ICT racks 26 are arranged in a row to form each of the rows 28A, 28B, 28C, 28D, 28E, and 28F. Further, the rows 28A, 28B, 28C, 28D, 28E, 28F are arranged in parallel.

  The ICT control device 30 is based on a command from the integrated control device 60, based on the command from the integrated control device 60, a degeneration command to instruct the ICT device 25 to perform the collective operation, a restoration command to instruct the distributed operation, a stop command to instruct the ICT device 25 to stop, It outputs a start command instructing the start of the operation.

  As shown in FIG. 1, the ICT control device 30 is a computer system having a CPU (Central Processing Unit), a ROM, a RAM, a hard disk, an input / output interface, and the like. The control program stored in the ROM or the like causes the CPU to function at least as the arithmetic processing unit 31, and causes the input / output interface or the like to function at least as the acquisition unit 32.

  The arithmetic processing unit 31 performs arithmetic processing for controlling the operation of the ICT device 25 and also generates a command for controlling the operation. Details of the arithmetic processing and the like in the arithmetic processing unit 31 will be described later.

  The acquisition unit 32 is an interface to which operation information such as current consumption and power information for each outlet output from the outlet bar 27, business processing amount output from the ICT device 25, internal device temperature, and power consumption is input. It is. The acquisition unit 32 is also an interface that outputs various commands generated by the arithmetic processing unit 31.

  As shown in FIG. 2, the air conditioner 35 cools the ICT device 25 through room air in a server room SR in which the ICT device 25 is arranged. That is, the heat generated in the ICT device 25 is released to the room air in the server room SR passing through the inside of the ICT device 25, and the temperature of the room air in the server room SR increases. The air conditioner 935 takes in the room air whose temperature has risen to take in heat, cools it down to a desired temperature, and sends it out into the server room SR. The heat taken from the indoor air is released to the outside in an outdoor unit (not shown) arranged outside the server room SR. The operation of the air conditioner 35 is controlled based on a command output from the air conditioning controller 40.

  In the present embodiment, in the server room SR in which the rows 28A, 28B, 28C, 28D, 28E, 28F are arranged in parallel at the center, the ends of the rows 28A, 28B, 28C, 28D, 28E, 28F. A description will be given by applying to an example in which the air conditioners 35 are arranged side by side in the vicinity. In addition, as a format for cooling the indoor air in the air conditioner 35 and a format for sending the cooled air into the server room SR, a known format can be used and is not particularly limited.

  Furthermore, in the present embodiment, for a plurality of air conditioners 35 arranged in the server room SR, an outdoor unit or the like which is arranged outdoors and emits heat to the outside air is provided in a one-to-one correspondence. Although described by applying to an example of a system air conditioner, a central heat source system air conditioner in which a plurality of air conditioners 35 are provided with one cooling tower or the like that is a means for releasing heat to the outside air may be used. There is no particular limitation.

  The air-conditioning control device 40 outputs a command for controlling activation of the air-conditioning device 35, a command for controlling stop of the air-conditioning device 35, and a command for controlling operating conditions of the air-conditioning device 35, based on a command from the integrated control device 60. It is.

  The air conditioning controller 40 is a computer system having a CPU (Central Processing Unit), a ROM, a RAM, a hard disk, an input / output interface, and the like, as shown in FIG. The control program stored in the ROM or the like causes the CPU to function at least as the arithmetic processing unit 41, and causes the input / output interface or the like to function at least as the acquisition unit 42.

  The arithmetic processing unit 41 performs arithmetic processing for controlling the operation of the air conditioner 35 and also generates a command for controlling the operation. Details of the arithmetic processing and the like in the arithmetic processing unit 41 will be described later. The acquisition unit 42 is an interface to which operation information indicating an operation state output from the air conditioner 35 is input.

  The lighting device 45 illuminates the server room SR where the ICT device 25 and the like are arranged. The lighting device 45 is provided with a human sensor (not shown) for detecting the presence of a person. The lighting device 45 illuminates when the human sensor detects a person, such as when a person enters the server room SR. The control of lighting / non-lighting in the air conditioner 35 is also performed based on a command (operation signal) output from the air conditioner control device 40. In addition, as the lighting device 45, a known device can be used, and there is no particular limitation.

  The lighting control device 50 is configured to output a command for controlling turning on / off of the lighting device 45 and a command for controlling operating conditions of the lighting device 45 based on a command from the integrated control device 60.

  The lighting control device 50 is a computer system having a CPU (Central Processing Unit), ROM, RAM, hard disk, input / output interface, and the like. The control program stored in the ROM or the like causes the CPU to function at least as the arithmetic processing unit 51, and causes the input / output interface or the like to function at least as the acquisition unit 52.

  The arithmetic processing unit 51 performs an arithmetic process for controlling the operation of the lighting device 45 and also generates a command for controlling the operation. Details of the arithmetic processing and the like in the arithmetic processing unit 51 will be described later. The acquisition unit 52 is an interface to which driving information indicating an operation state output from the lighting device 45 is input, and is also an interface to which measurement information of a human sensor attached to the lighting device 45 is input.

  The above-described power supply control device 20, ICT control device 30, air conditioning control device 40, and lighting control device 50 are output when the communication with the integrated control device 60 is interrupted and no command is input, or when the command is output from the integrated control device 60. If the received command is an abnormal value, it is determined that the integrated control device 60 has failed, and each control device transitions to an independent operation state.

  The integrated control device 60 is input from failure information and operation abnormality information input from the monitoring control device 70, external EMS (energy management system) such as BEMS and BAS, DCIM (data center infrastructure management), and the like. Based on the information and the command, the processing for determining the control content for the power supply control device 20, the ICT control device 30, the air conditioning control device 40, and the lighting control device 50, and the process for generating a command to be output are performed.

  The integrated control device 60 is a computer system having a CPU (central processing unit), a ROM, a RAM, a hard disk, an input / output interface, and the like. The control program stored in the ROM or the like causes the CPU to function as at least an arithmetic processing unit (estimating unit) 61 and a prediction processing unit (first prediction unit, second prediction unit) 63. And the like function at least as the acquisition unit 64.

  The acquisition unit 64 includes a boot time database (storage unit) 66 (hereinafter, also referred to as a “boot time DB 66”) in which a boot time (a predetermined threshold), which is a time required to boot the ICT device 25, is stored, and the ICT device. 25 is an interface for acquiring information from a CPU load database 67 (hereinafter, also referred to as “CPU load DB 67”) in which information on CPU load (target data) is stored.

  The acquisition unit 64 also stores information output from the power supply control device 20, the ICT control device 30, the air conditioning control device 40, the lighting control device 50, and the integrated control device 60, the arrangement position of the ICT rack 26, and the ICT in the ICT rack 26. This is an interface for acquiring information on the mounting position of the device and command information input from the outside. The acquisition unit 64 is also an external interface that can issue information and commands to an external EMS system, DCIM, and the like.

  The startup time DB 66 stores a startup time, which is a time required for starting up the ICT device 25. The activation time includes the type of the ICT device 25, the OS and APP installed in the ICT device 25, the communication speed in a network connected to the ICT device 25, the time zone in which the ICT device 25 is activated, and the activation time. It depends on the number of people who may use the ICT device 25 at the time of the operation. Therefore, the activation time DB 66 stores a plurality of activation times associated with these.

  The CPU load DB 67 stores information on the CPU load in the ICT device 25. The information on the CPU load is stored in association with the acquired time, and the information on the CPU load in the time zone designated by the integrated control device 60 is stored so as to be output.

  The arithmetic processing unit 61 and the prediction processing unit 63 both calculate the required number of operating ICT devices 25 after a certain period of time, or calculate the number of starting or stopping ICT devices 25. The specific contents of the arithmetic processing in the arithmetic processing unit 61 and the prediction processing unit 63 will be described later.

  The monitoring control device 70 is a computer system having a CPU (Central Processing Unit), ROM, RAM, hard disk, input / output interface, and the like. The control program stored in the ROM or the like causes the CPU to function at least as the arithmetic processing unit 71, and causes the input / output interface or the like to function at least as the acquisition unit 72.

  The acquisition unit 72 is an interface that acquires information on a failure state and an operation abnormality output from devices configuring the information processing system 1 such as the power supply device 10, the current distribution device 15, the ICT device 25, and the air conditioning control device 40. It is. The arithmetic processing unit 71 monitors the failure state and the operation abnormality of the devices constituting the information processing system 1 based on the information on the failure state and the operation abnormality acquired by the acquisition unit 72.

Next, control in the information processing system 1 having the above configuration will be described. First, the operation of the information processing system 1 based on a schedule will be described.
First, a normal operation state in which the ICT device 25 is performing a distributed operation will be described with reference to FIG. The distributed operation is a state in which arithmetic processing such as information processing is performed in all the ICT devices 25 in the information processing system 1 or in most of the ICT devices 25.

  In FIG. 2, the percentage described in the ICT device 25 (ICT rack 26) indicates the load factor of the corresponding ICT device. The hatching in the server room SR indicates the temperature of the indoor air in the server room SR, and indicates that the temperature increases as the hatching increases.

  The power supply device 10, the air conditioner 35, and the ICT device 25 are arranged in the server room SR. The breakdown of the configuration of the ICT device 25 based on the power consumption base is as follows: 80% of server devices and 20% of network devices such as switches and routers. In the present embodiment, a description will be given by applying to an example in which the total maximum power consumption of the ICT device 25 is 300 kW.

  FIG. 2 shows an example in which the average operation rate of the ICT device 25 is about 20%. The analysis result of the power consumption in this case is that the power consumption of the ICT device 25 is 163 kW, the air conditioning power is 62 kW, the power supply loss is 21 kW, and the total is 246 kW.

  Next, a description will be given of a case where the integrated linkage control according to the present invention is performed on the information processing system 1 having the above configuration. In the present description, the integrated control device 60 issues a command to perform a degenerate operation for consolidating the load on the ICT device 25 during a time period when the load on the ICT device 25 is reduced, such as on a weekday night, Saturday, Sunday, or holiday. Is output. When the load of the ICT device 25 increases during the degenerate operation, the integrated control device 60 outputs a command to release the degenerate operation in advance, and then outputs a command to return to the normal distributed operation.

  Here, the control processing when executing the degeneration operation will be described with reference to the flowcharts of FIGS. 3 and 4. First, as shown in FIG. 3, the integrated control device 60 performs a process (S10) of estimating the number of ICT devices 25 to be operated in the information processing system 1 by the degenerate operation.

Specifically, as shown in FIG. 4, the acquisition unit 64 of the integrated control unit 60 performs processing for acquiring information _{P t} CPU load mounted on ICT device 25 (S11: acquisition step). Information P _t of the CPU load is information that changes in time series, for example, performs a process of acquiring every minute. Integrated control unit 60 performs a process of storing information _{P t} of the acquired CPU load on the CPU load DB 67. The time interval for acquiring the information P _t of the CPU load, may be the 1 minute as described above, may be longer or shorter than 1 minute.

Subsequently, the integrated control unit 60 performs a process of aligning the information P _t CPU load within the scope of going back only in the past predetermined time from the present time to perform the calculation processing (n min) (S12). Specifically, a predetermined time (n min) from the current time from the CPU load DB67 only obtains information _{P t} CPU load within the scope of going back in the past, when data sequence linked to sequence _{(P t-1} to P _n ) Is performed.

Thereafter, the prediction processing unit 63 of the integrated control device 60 determines the first time, which is the activation time of the ICT device 25 after the first predetermined time, based on the data sequence (Pt _{-1 to Pn} ) connected in time series. A process of calculating the predicted value _{Pt + 5} is performed (S13: first prediction step). In the present embodiment, the first predetermined time is described as applied to an example in which the time is 5 minutes, but may be longer or shorter than 5 minutes, and is not particularly limited.

Next, the prediction processing unit 63 of the integrated control device 60 determines the second time that is the activation time of the ICT device 25 after the second predetermined time based on the data sequence (P _{t−1 to} P _n ) connected in time series. A process of calculating the predicted value _{Pt + 10} is performed (S14: second prediction step). In the present embodiment, a description will be given by applying to an example in which the second predetermined time is 10 minutes. However, the specific value is not limited as long as it is longer than the first predetermined time. Further, an example in which the first predicted value Pt _{+ 5} and the second predicted value Pt _{+ 10} are calculated using chaos prediction processing (first prediction method and second prediction method) disclosed in Japanese Patent Laid-Open No. 7-270200. However, other known methods may be used.

After calculating the first predicted value _{Pt + 5} and the second predicted value _{Pt + 10} , the arithmetic processing unit 61 of the integrated control device 60 determines the start time of the ICT device 25, the first predicted value _{Pt + 5,} and the second predicted value _{Pt + 5} . A process of determining a predicted value to be used in the following estimation process based on the value _{Pt + 10} is performed (S15: estimation step). First, the acquisition unit 64 performs a process of acquiring a start time from the start time DB 66. At this time, the type of the ICT device 25, the OS and APP installed in the ICT device 25, the communication speed in the network connected to the ICT device 25, the time period during which the ICT device 25 is started, and the start are performed. On the basis of information on the number of persons who may use the ICT device 25 at the time of starting, the start time corresponding to the information is acquired. Note that, as information on the number of persons who may use the ICT device 25 when the activation is performed, a value obtained by arithmetic processing by the arithmetic processing unit 61 or the like of the integrated control device 60 is used.

Next, the arithmetic processing unit 61 sets a first time difference that is a time difference between the activation time acquired from the activation time DB 66 and the first predetermined time used when obtaining the first predicted value Pt _{+ 5} , and the activation time DB 66 And a second time difference that is a time difference between the startup time acquired from the second time and the second predetermined time used in obtaining the second predicted value _{Pt + 10} . Thereafter, a process of comparing the absolute value of the first time difference and the absolute value of the second time difference to determine a predicted value used for the estimation process is performed.

Specifically, when it is determined that (the absolute value of the first time difference) <(the absolute value of the second time difference), the arithmetic processing unit 61 determines the first predicted value _{Pt + 5} as the predicted value used for the estimation processing. Yes (in the case of YES). On the other hand, when it is determined that (absolute value of the first time difference)> (absolute value of the second time difference), the arithmetic processing unit 61 determines the second predicted value _{Pt + 10} as the predicted value used for the estimation processing. (If NO).

When (absolute value of the first time difference) = (absolute value of the second time difference), the process returns to S11, and the above process may be repeated, or the first predicted value _{Pt + 5} may be used for the estimation process. It may be determined as a predicted value to be used, or the second predicted value _{Pt + 10} may be determined as a predicted value used in the estimation processing.

If the answer is YES in S15, the arithmetic processing unit 61 includes information _{P t} CPU load in the arithmetic processing time point, the operation of the first predicted value _{P t + 5} and ICT device 25 required after a certain time by comparing Calculation processing for calculating the number is performed (S16: estimation step). The arithmetic processing unit 61 determines the number of ICT devices 25 to be started or the number of ICT devices 25 to be stopped based on the calculated required number of operating devices and the number of operating ICT devices 25 actually operating at the time of the arithmetic processing. Is calculated.

If NO is determined in S15, the arithmetic processing unit 61 compares the information P _t of the CPU load at the time of the arithmetic processing with the second predicted value P _{t + 10} and operates the ICT device 25 necessary after a certain period of time. Calculation processing for calculating the number is performed (S17: estimation step). A known method can be used as a method for calculating the required number of operating ICT devices 25 used in S16 and S17, and is not particularly limited.

  Returning to FIG. 4, the integrated control device 60 outputs a command to execute the aggregation process for shifting to the degenerate operation to the target ICT device 25 (S21). When the aggregation process is performed by virtualization, the degeneracy source ICT device 25 performs live migration to the degeneration destination ICT device 25. After the transition is completed, the degeneration source ICT device 25 transitions to the sleep state. If the aggregation process is a suspension of the operation of the surplus ICT device 25, the target device is shifted to a suspension state.

  When selecting the degeneration source ICT device 25 that shifts to the sleep state and the degeneration destination ICT device 25 that increases the load, a degeneration pattern that is a combination of the ICT devices 25 to be operated and stored in advance is used.

  In the present embodiment, as shown in FIG. 5, the ICT devices 25 mounted on the rows 28A, 28B, 28E, and 28F are degenerated ICT devices 25, and the ICT devices 25 mounted on the rows 28C and 28D are degenerated. A description will be given by applying to an example in which the ICT device 25 is used.

  Next, returning to the flowchart of FIG. 3, a process of determining whether or not the movement of the ICT resource (move of the Guest machine) is successful (S22). Specifically, the determination is made based on whether the following five conditions are satisfied. First, no hardware failure (hereinafter, referred to as “H / W failure”) has occurred in the degeneration target ICT device 25. Second, no hardware failure has occurred in the degenerate source ICT device. Third, the virtual guest machine of the degeneration source is normal on the virtualization management software. Fourth, the virtual guest machine to be moved is normal on the degeneration source virtualization software. Fifth, there is room for the ICT resources at the degeneration destination. More specifically, even if the operation processing of the degeneration source ICT device 25 is shifted to the degeneration destination ICT device 25, the CPU usage rate in the degeneration destination ICT device 25 is less than a predetermined value (for example, 80%), The usage rate is less than a predetermined value (for example, 80% of the maximum capacity), and the usage rate of the hard disk (HDD) is less than a predetermined value (for example, 80% of the maximum capacity).

  If it is determined that the movement of the ICT resource has failed (in the case of NO), the process of determining whether or not the degenerate operation is possible is stopped. When it is determined that the movement of the ICT resource is successful (in the case of YES), the integrated control device 60 issues a stop instruction to the ICT device 25 of the degeneration source to the ICT control device, and performs a stop process for the ICT device 25. A process is performed to determine whether or not was successful (S23). Specifically, the determination is made based on whether the following four conditions are satisfied. First, the CPU usage rate of the suspension target ICT device 25 is equal to or less than a predetermined value (for example, 5%), and the memory usage rate is equal to or less than a predetermined value (for example, 5%). Second, there is no virtual Guest machine on the ICT device 25 to be suspended on the virtualization management software. Third, there is no virtual Guest machine on the virtualization software of the ICT device 25 to be suspended. Fourth, the current consumption of the outlets (Outlets) of the switches 16 and 27A connected to the ICT device 25 to be paused is not more than a predetermined value (for example, 30% of the rated current).

  When it is determined that the suspension of the ICT device 25 has failed (in the case of NO), the process of determining whether or not the degeneration operation is possible is stopped. When it is determined that the halt of the ICT device 25 is successful (in the case of YES), the integrated control device 60 checks the halt state information of the ICT device 25 output from the ICT control device 30. After the confirmation, a stop (shutdown) command is issued to the ICT device 25 that has been suspended, to the ICT control device 30. After confirming the stop of the ICT device, the integrated control device outputs a command to the power supply control device to turn off the power supply to the outlet bar 27 on the power supply route or the switch 16 or the switch 27A of the current distribution device 15 ( S31).

  Next, a process is performed to determine whether the power is turned off at the switch 16 or the switch 27A corresponding to the stopped ICT device 25 (S32). Specifically, a process of determining whether the following two conditions are satisfied is performed. The first is that the connected ICT device 25 is stopped. More specifically, the following three contents must be satisfied. When there is no response even if a pin (Ping) is connected to the target ICT device 25, it cannot be connected to the virtualization software of the target ICT device 25, and the target ICT device 25 is connected to the target ICT device 25 via a management network (NW), and Is turned off (OFF). Second, the current consumption of the outlet (Outlet) of the switch 16 or the switch 27A connected to the target ICT device 25 is equal to or less than a predetermined value (for example, 10% of the rated current).

When it is determined that the power supply has not been turned off (in the case of NO), the process of determining whether or not the degeneration operation is possible is stopped.
Next, when it is determined that the power supply is turned off (in the case of YES), the integrated control device 60 outputs a command to stop the power supply unit 11 corresponding to the stopped ICT device 25 (S41). In other words, since the operation load factor of the power supply device 10 decreases with the stop of the ICT device 25, a command to stop the surplus power supply unit 11 is output to the power supply control device 20.

  Specifically, the number of power supply units 11 mounted on power supply device 10 is compared with the output current value, and a command to stop the number of power supply units 11 determined by the following equation is output. The number of the power supply units 11 to be stopped may be a number in consideration of a failure of the power supply unit 11 or a sudden increase in load, and the number of the necessary power supply units 11 in consideration of redundancy.

[(Number of power supply units 11 mounted on power supply device 10) × (Capacity of power supply unit 11 alone) − {(Output current) + (Capacity of power supply unit 11 alone)}] ÷ (Capacity of power supply unit 11 alone) ≧ Number of power supply units 11 that can be stopped Next, processing is performed to determine whether or not the power supply units 11 corresponding to the stopped ICT devices 25 are stopped (S42). When it is determined that the power supply unit 11 is not stopped (in the case of NO), the process of determining whether the degenerate operation is possible is stopped.

  Next, when it is determined that the power supply unit 11 is stopped (in the case of YES), the integrated control device 60 responds to the stopped ICT device 25 with a margin for the temperature of the stopped ICT device 25 to decrease. A command to stop the air conditioner 35 is output to the air conditioning controller 40 (S51). The air conditioning controller 40 outputs to the air conditioner 35 a command to stop the air conditioner 35 corresponding to the stopped ICT device 25.

  In the present embodiment, as shown in FIG. 5, description will be made by applying to an example in which the air conditioners 35 corresponding to the rows 28A, 28B, 28E, and 28F are stopped. In FIG. 5, power supply OFF is indicated by diagonal lines in the corresponding air conditioner 35.

  If the temperature of the air-conditioning zone to which the ICT device 25 of the degenerate destination belongs rises by stopping the air-conditioning device 35 corresponding to the stopped ICT device 25, the air-conditioning control device 40 A command to restart the operation of the device 35 may be output.

  Next, returning to the flowchart of FIG. 3, a process of determining whether or not the air conditioner 35 corresponding to the suspended ICT device 25 is stopped (S52). Specifically, a process of determining whether the following four conditions are satisfied is performed. First, the air conditioner 35 corresponding to the ICT device 25 other than the one to be stopped is normal (in other words, no failure has occurred). Second, the air conditioner 35 corresponding to the ICT device 25 other than the stop target has sufficient cooling capacity and is less than a predetermined value (90%) of the rated capacity. Third, the temperature in the vicinity of the suction part of the ICT device 25 (or a part of the cold aisle (Cold Aisle) which is a part between rows) of the air conditioner 35 other than the stop target is lower than a predetermined temperature. . Fourth, the ICT device 25 controlled by the air conditioner 35 to be stopped is not operating. More specifically, the ICT device 25 is stopped, and the outlet of the switch 16 or the switch 27A connected to the ICT device 25 is OFF.

  When it is determined that the air conditioner 35 is not stopped (in the case of NO), the process of determining whether the degeneration operation is possible is stopped. When it is determined that the air conditioner 35 is stopped (in the case of YES), the integrated control device 60 performs a process of outputting a command to turn off the lighting device 45 (S61). In other words, a turn-off command is output to the lighting control device 50 in order to turn off the unnecessary lighting device 45. After confirming the absence of the worker as necessary, the lighting control device 50 turns off the corresponding portion.

  Next, a process of determining whether or not the lighting device 45 has been turned off is performed (S62). Specifically, a process of determining whether the following three conditions are satisfied is performed. First, the ICT device 25 has stopped normally. Second, all the corresponding switches 16 or 27A are OFF. Third, the corresponding air conditioner 35 is stopped.

  When it is determined that the lighting device 45 has not been turned off (in the case of NO), the process of determining whether or not the degraded operation is possible is stopped. When it is determined that the lighting device 45 is turned off (in the case of YES), the process of determining whether or not the degraded operation is possible is ended.

  In executing the transition process to the degenerate operation as described above, network control and load control such as a software defined network (SDN) technology, a network function virtualization (NFV) technology, and a load balancer may be performed as necessary. .

Next, a control process when executing the restoration from the degenerate operation to the distributed operation will be described with reference to the flowchart of FIG.
First, the integrated control device 60 performs a process of activating a schedule that determines the timing of the transition determination to the distributed operation, and performs a process of determining whether the schedule has been activated (S111). Here, the timing of the shift determination may be, for example, the end of a time zone in which a decrease in information processing or communication processing is expected, such as a weekday night, Saturday, Sunday, or holiday. If the activation of the schedule is determined to be unsuccessful (NO), the process of determining whether or not migration is possible is stopped.

  When it is determined that the activation of the schedule is successful (in the case of YES), the integrated control device 60 determines whether or not a failure has occurred in the ICT device 25 or the like or whether or not operation is hindered. Is performed (S112). Specifically, based on whether or not the monitoring control device 70 has received information indicating that an abnormality has occurred in the distribution source ICT device 25 and the distribution destination ICT device 25, the air conditioner 35, and the power supply device 10. , The above-described determination processing is performed. If it is determined that a failure has occurred or operation has been hindered (in the case of NO), the process of determining whether or not migration is possible is stopped.

Note that the distribution source ICT device 25 is the above-described degeneracy destination ICT device 25, and the distribution destination ICT device 25 is the above-described degeneration source ICT device 25.
When it is determined that no failure has occurred (in the case of YES), the integrated control device 60 performs a process of determining whether or not the power supply device 10 has a sufficient capacity (S113). That is, the integrated control device 60 determines whether the current consumption current and the rated current of the ICT device 25 to be activated do not exceed the outputable current value of the power supply device 10. When it is determined that there is no room for the capacity (in the case of NO), the process of determining whether or not migration is possible is stopped.

  When it is determined that there is enough capacity (in the case of YES), a process of outputting a command to start the air conditioner 35 related to the start target is performed (S121). In other words, a command to start an operation for pre-cooling is output to the air conditioning control device 40 that controls the air conditioner 35 in the air conditioning zone of the ICT device 25 that resumes operation. The air conditioning controller 40 outputs a start command to the corresponding air conditioner 35.

  Next, a process of determining whether or not the air conditioner 35 relating to the activation target has been activated is performed (S122). When it is determined that the air conditioner 35 relating to the activation target has not been activated (in the case of NO), the process of determining whether or not the transition is possible is stopped.

  When it is determined that the air conditioner 35 related to the activation target has been activated (in the case of YES), the integrated control device 60 performs a process of determining whether the temperature around the ICT device 25 is equal to or lower than a predetermined temperature. (S123). As the temperature around the ICT device 25, for example, the air temperature around the ICT device 25 or the suction air temperature in the air conditioner 35 can be exemplified. In this determination, specifically, it is determined whether the following two conditions are satisfied. First, no abnormality has occurred in the activated air conditioner 35. Second, the temperature near the suction part of the ICT device 25 (or a part of the cold aisle (Cold Aisle), which is a part between rows) of the ICT device 25 under the control of the activated air conditioner 35 is equal to or lower than a predetermined temperature. thing.

  If it is determined that the temperature around the ICT device 25 is not lower than the predetermined temperature (NO), the determination in S123 is performed again. When it is determined that the temperature around the ICT device 25 is lower than or equal to the predetermined temperature (in the case of YES), the integrated control device 60 activates the power supply unit 11 that supplies power to the ICT device 25 that resumes operation. A process of outputting a command is performed (S131). Specifically, it outputs a command to activate the power supply unit 11 to the power supply control device 20. The power supply control device 20 activates the power supply unit 11.

  Next, the integrated control device 60 performs a process of determining whether or not the power supply unit 11 has been started (S132). The determination is made based on whether the following three conditions are satisfied. First, no abnormality has occurred in the power supply device 10. Second, the power supply unit 11 has started normally. Third, no abnormality has occurred in the power supply unit 11 even after startup. When it is determined that the power supply unit 11 has not been activated (NO), a process of outputting an alert is performed.

  When it is determined that the power supply unit 11 is activated (in the case of YES), a process of outputting a command to turn on the switch 16 and the switch 27A is performed (S141). Specifically, the power supply control device 20 outputs a command to shift from the OFF state to the ON state to the switch 16 and the switch 27A located on the power supply path to the ICT device 25 that resumes operation.

  Next, the integrated control device 60 performs a process of determining whether or not the switches 16 and 27A have transitioned to the ON state (S142). The determination is made based on whether the following three conditions are satisfied. First, no abnormality has occurred in the switches 16 and 27A. Second, the outlets of the switch 16 and the switch 27A are normally turned on. No abnormality has occurred even after the outlet is turned on. When it is determined that the switch 16 or the switch 27A remains in the OFF state (in the case of NO), a process of outputting an alert is performed.

  When it is determined that the switches 16 and 27A have shifted to the ON state (in the case of YES), the integrated control device 60 performs a process of outputting a start command to the ICT device 25 (S151). The ICT control device 30 to which the command has been input performs a process of starting the ICT device 25 according to a start procedure according to the specifications of the ICT device 25. The integrated control device 60 subsequently outputs an instruction to start the restoration process to the ICT control device 30 to shift from the degenerate operation to the distributed operation. The ICT control device 30 performs a process of shifting from the degenerate operation to the distributed operation state by performing a procedure necessary for the distributed operation.

  Next, the integrated control device 60 performs a process of determining whether or not the ICT device 25 (Host machine) has started (S152). The determination is made based on whether the following three conditions are satisfied. First, it can be connected to the target ICT device 25 by the management NW, and no abnormality has occurred. Second, it can be connected to the target ICT device 25 with a management NW and can be turned on. Third, start up without any abnormality even after the power is turned on. When it is determined that the ICT device 25 has not been activated (in the case of NO), a process of outputting an alert is performed.

  When it is determined that the ICT device 25 has been activated (in the case of YES), the integrated control device 60 performs a process of determining whether or not ICT resources have been moved (Guest machines have been moved) (S153). . The determination is made based on whether the following five conditions are satisfied. First, no H / W failure has occurred in the distribution destination ICT device 25 which is the switching (moving) destination ICT device 25. Second, no hardware failure has occurred in the distribution source ICT device 25 that is the switching (moving) source ICT device 25. Third, the virtual guest machine at the distribution source is normal on the virtualization management software. Fourth, the virtual guest machine to be moved on the distribution source virtualization software is normal. Fifth, there is enough ICT resources at the distribution destination. Specifically, the CPU usage rate is less than a predetermined value (for example, 80%), the memory usage rate is less than a predetermined value (for example, 80%), and the hard disk (HDD) usage rate is less than a predetermined value (for example, 80%). For example, less than 80%).

  When it is determined that the ICT resource has not been moved (in the case of NO), a process of outputting an alert is performed. When it is determined that the ICT resource has been moved (in the case of YES), the integrated control device 60 performs a process of outputting a command to turn on the lighting device 45 (S161).

  Next, the integrated control device 60 performs a process of determining whether or not the lighting device 45 is turned on (S162). The determination is made based on whether or not the power supply device 10, the ICT device 25, and the air conditioner 35 have started normally and no abnormality has occurred.

  When it is determined that the lighting device 45 is not turned on (in the case of NO), a process of outputting an alert is performed. When it is determined that the lighting device 45 is turned on (in the case of YES), the process of executing the restoration from the degenerate operation to the distributed operation ends. As described above, the power consumption when the information processing system 1 shifts to the distributed operation state returns to the power consumption before the degenerate operation illustrated in FIG.

  In the restoration processing for shifting from the degenerate operation to the decentralized operation, network control and load control such as a software defined network (SDN) technology, a network function virtualization (NFV) technology, and a load balancer may be performed as necessary. If an unnecessary alarm is generated during the transition process, it is possible to take measures to temporarily mask the alarm.

According to the above configuration, when calculating the required number of operating ICT devices 25 after a certain period of time, or calculating the number of starting or stopping ICT devices 25, the first predicted value _{Pt + 5} determined by comparing with the starting time. and one of the second predicted value _{P t + 10,} and the information _{P t} CPU load in the arithmetic processing time point is used. By doing so, a prediction value that can easily improve the prediction accuracy can be selected from the first prediction value _{Pt + 5} and the second prediction value _{Pt + 10 based} on the determination compared with the activation time. Therefore, compared to the calculation method described in Patent Literature 1 in which the current value of the CPU load is compared with the relatively short-term predicted value and the relatively long-term predicted value, the prediction accuracy can be easily improved.

In the determination process of S15 in comparison with the startup time, a first time difference between the startup time and the first predetermined time, and a second time difference between the startup time and the second predetermined time are compared, and a predicted value having a small time difference is used. Makes it easier to select a predicted value that can easily improve the prediction accuracy from the first predicted value _{Pt + 5} and the second predicted value _{Pt + 10} .

By making the length of the start-up time different according to the type of the ICT device 25, the selection according to the type of the ICT device 25 can be made when selecting the first predicted value _{Pt + 5} and the second predicted value _{Pt + 10.} This makes it easier to select a predicted value that can easily improve the prediction accuracy.

By making the length of the startup time different according to at least one of the OS and the APP mounted on the ICT device 25, when selecting the first predicted value _{Pt + 5} and the second predicted value _{Pt + 10} , Selection according to the OS and APP can be performed, and it becomes easier to select a predicted value that can easily improve the prediction accuracy.

By making the length of the start-up time different according to the communication speed in the network connected to the ICT device 25, the communication speed in the network is selected when the first predicted value _{Pt + 5} and the second predicted value _{Pt + 10} are selected. Can be selected, and it becomes easier to select a predicted value that can easily improve the prediction accuracy.

The first predicted value _{Pt + 5} and the second predicted value are determined by making the length of the start-up time different from at least one of the time zone in which the determination is performed and the number of persons who may use the information processing apparatus at the time of the determination. It is possible to make a selection according to the time zone in which the determination is made when selecting the predicted value _{Pt + 10} and the number of persons who may use the ICT device 25 at the time of the determination, and the predicted value is likely to improve the prediction accuracy. Becomes easier to select.

The technical scope of the present invention is not limited to the above embodiment, and various changes can be made without departing from the spirit of the present invention. For example, the above embodiment is applied to an example in which the process of calculating the first predicted value _{Pt + 5} (S13) and then the process of calculating the second predicted value _{Pt + 10} (S14) are performed. As described above, the processing of S14 may be performed first and the processing of S13 may be performed later, or the processing of S13 and the processing of S14 may be performed simultaneously.

  Further, in the above embodiment, the processing of S13 and the processing of S14 are performed, and then the processing of determining the predicted value to be used for the estimation processing (S15) has been described as applied to the example in which the processing of S15 is performed. Only the process (S13 or S14) that is performed first and obtains the determined predicted value may be performed.

  25: ICT device (information processing device), 60: integrated control device (control device), 61: arithmetic processing unit (estimating unit), 63: prediction processing unit (first prediction unit, second prediction unit), 64 ... acquisition Unit, 66: startup time database (storage unit), S11: acquisition step, S13: first prediction step, S14: second prediction step, S15: estimation step, S16: estimation step, S17: estimation step

Claims (15)

A control device for an information processing device that controls the number of operating information processing devices connected to the plurality of terminal devices via a network,
An acquisition unit that acquires target data that changes in chronological order from the information processing device and stores the data in a predetermined storage unit;
A first prediction unit that calculates a first prediction value of the target data using a first prediction method based on the target data within a predetermined time in the past extracted from the storage unit;
A second prediction unit that calculates a second prediction value of the target data using a second prediction method different from the first prediction method, based on the target data within the past predetermined time;
By comparing with a predetermined threshold value, it is determined whether to use one of the first predicted value and the second predicted value, and one of the determined first predicted value and the second predicted value is determined. And, using the value of the target data at the time of calculation, calculate the required number of operating the information processing apparatus after a certain period of time, or, or an estimating unit that calculates the number of starting or stopping,
Is provided ,
The first predicted value is a predicted value of the target data after a first predetermined time,
The second predicted value is a predicted value of the target data after a second predetermined time different from the first predetermined time,
The predetermined threshold is a time required for starting the information processing apparatus (hereinafter, referred to as “start time”),
The estimating unit compares a difference between the activation time and the first predetermined time, a difference between the activation time and the second predetermined time, and compares the difference between the first predicted value and the second predicted value. The control device for an information processing device, wherein the control unit determines that a predicted value having a small difference from the activation time is used . The start time, the control device of the information processing apparatus according to claim 1, wherein a time required for the activation in response to the type of the information processing apparatus is different. The start time, the control device of the information processing apparatus according to claim 1, wherein a time required for the activation in response to at least one of the operating systems and applications are installed in the information processing apparatus is different from . 2. The control device according to claim 1, wherein the activation time is different in a time required for the activation depending on a communication speed in the network connected to the information processing device. 3. The activation time differs according to at least one of a time zone in which the estimation unit makes a determination and a number of persons who may use the information processing apparatus at the time of the determination. The control device for an information processing device according to any one of claims 1 to 4 , wherein: A method of controlling an information processing device that controls the number of operating information processing devices connected to a plurality of terminal devices via a network,
An acquisition step of acquiring target data that changes in chronological order from the information processing apparatus and storing it in a predetermined storage unit;
A first prediction step of calculating a first prediction value of the target data using a first prediction method based on the target data within a predetermined time in the past taken out from the storage unit;
A second prediction step of calculating a second predicted value of the target data using a second prediction method different from the first prediction method based on the target data within the past predetermined time;
By comparing with a predetermined threshold value, it is determined whether to use one of the first predicted value and the second predicted value, and one of the determined first predicted value and the second predicted value is determined. And, using the value of the target data at the time of calculation, calculate the required number of operating the information processing apparatus after a certain period of time, or an estimation step of calculating the number of starting or stopping,
Have a,
The first predicted value is a predicted value of the target data after a first predetermined time,
The second predicted value is a predicted value of the target data after a second predetermined time different from the first predetermined time,
The predetermined threshold is a time required for starting the information processing apparatus (hereinafter, referred to as “start time”),
In the estimating step, a difference between the activation time and the first predetermined time, and a difference between the activation time and the second predetermined time are compared, and a difference between the first predicted value and the second predicted value is calculated. A method for controlling an information processing apparatus, characterized in that it is determined that a predicted value having a small difference from the activation time is used. 7. The control method for an information processing apparatus according to claim 6, wherein the activation time is different in a time required for the activation depending on a type of the information processing apparatus. 7. The control method according to claim 6, wherein the activation time differs according to at least one of an operating system and an application installed in the information processing apparatus. . 7. The method according to claim 6, wherein the activation time is different in time required for the activation according to a communication speed in the network connected to the information processing device. The activation time differs according to at least one of a time zone in which the determination in the estimation step is performed and a number of persons who may use the information processing apparatus at the time of the determination. control method for an information processing apparatus according to any one of claims 6 9, characterized in that. A program that causes a computer to function as a control device of an information processing device that controls the number of operating information processing devices connected to a plurality of terminal devices via a network,
An acquisition function of acquiring target data that changes in chronological order from the information processing device to the computer and storing the acquired target data in a predetermined storage unit;
A first prediction function of calculating a first prediction value of the target data using a first prediction method based on the target data within a predetermined time in the past extracted from the storage unit;
A second prediction function for calculating a second predicted value of the target data using a second prediction method different from the first prediction method based on the target data within the past predetermined time;
By comparing with a predetermined threshold value, it is determined whether to use one of the first predicted value and the second predicted value, and one of the determined first predicted value and the second predicted value is determined. And, using the value of the target data at the time of calculation, calculate the required number of operating the information processing apparatus after a certain period of time, or implement an estimation function of calculating the number of starting or stopping ,
The first predicted value is a predicted value of the target data after a first predetermined time,
The second predicted value is a predicted value of the target data after a second predetermined time different from the first predetermined time,
The predetermined threshold is a time required for starting the information processing apparatus (hereinafter, referred to as “start time”),
The estimating function compares a difference between the activation time and the first predetermined time, and a difference between the activation time and the second predetermined time, and compares the difference between the first predicted value and the second predicted value. The program is characterized in that it is determined that a predicted value having a small difference from the activation time is used . 12. The program according to claim 11, wherein the startup time is different from the time required for the startup according to the type of the information processing apparatus. 12. The program according to claim 11, wherein the startup time differs according to at least one of an operating system and an application installed in the information processing apparatus. 12. The program according to claim 11, wherein the start-up time differs according to a communication speed in the network connected to the information processing apparatus. The activation time differs according to at least one of a time zone in which the determination in the estimation function is performed and a number of persons who may use the information processing apparatus at the time of the determination. The program according to any one of claims 11 to 14 , wherein: