JP6443123B2 - Program, driving method, driving device - Google Patents

Description

  The present invention relates to a program, an operation method, and an operation apparatus for operating a plurality of electronic devices by a computer.

  A so-called rack mount system is used in which a plurality of electronic devices are installed and used on a standardized shelf. A management system that associates an ID (Identification) for individual identification of an electronic device with a physical position of the electronic device installed on a shelf is disclosed (Patent Document 1).

JP 2011-165104 A

  However, Patent Document 1 does not describe a method of using information regarding the physical position of each electronic device installed in the rack mount system.

  In one aspect, an object of the present invention is to provide a program or the like that reduces the amount of power used in consideration of the continuity of operation of a system using an electronic device.

In one aspect, the program of the present invention obtains a limitation on the amount of power consumed by a plurality of electronic computers and a time zone for performing the limitation , and the relationship between the limitation and the CPU usage rate of the electronic computer and the power consumption. determine the number of the computer to stop on the basis of, calculated was based on the number and the location of installation of the computer to stop, computer management to stop the computer of the identified the site before the time slot To run.

  In one aspect, it is possible to reduce the amount of power used in consideration of the continuity of the operation of the system using electronic devices.

It is an apparatus block diagram which shows the hardware constitutions of a server system. It is explanatory drawing which shows the structure of a data center. It is explanatory drawing which shows the cooling method of a server system. It is explanatory drawing which shows the relationship between CPU utilization and power consumption. It is explanatory drawing which shows the relationship between the stop number of a managed server, power consumption, and CPU usage rate. It is explanatory drawing which shows the structure of the electronic device mounted in the shelf. It is explanatory drawing explaining the change of the temperature of the server system which stopped some managed servers. It is a flowchart which shows the flow of a process of a program. It is a flowchart which shows the flow of a process of stop number calculation. It is an example of the table which shows the location of the managed server to stop. It is a flowchart which shows the flow of a specific process for the managed server to stop. 10 is a flowchart illustrating a flow of processing for specifying a managed server to be stopped according to the second embodiment. 10 is a table showing an example of specifying a place where a managed server to be stopped according to the second embodiment is arranged. 10 is a flowchart showing a flow of processing for specifying a managed server to be stopped according to the third embodiment. 10 is a flowchart showing a flow of processing for specifying a managed server to be stopped according to the third embodiment. 10 is a table showing an example of specifying a place where a managed server to be stopped according to a third embodiment is arranged. 14 is a flowchart illustrating a processing flow of a program according to the fourth embodiment. 14 is a flowchart illustrating a flow of processing for calculating the number of stopped vehicles according to the fourth embodiment. 14 is a flowchart illustrating a flow of processing for calculating the number of stopped vehicles according to the fourth embodiment. FIG. 10 is an apparatus configuration diagram illustrating a hardware configuration of a server system according to a fifth embodiment. FIG. 10 is an explanatory diagram illustrating a configuration of an electronic device mounted on a shelf according to a fifth embodiment. FIG. 10 is an explanatory diagram illustrating a change in temperature when a part of managed servers according to the fifth embodiment is stopped. FIG. 20 is an explanatory diagram illustrating a comparative example of a change in temperature when a part of managed servers according to the fifth embodiment is stopped. 10 is a flowchart showing a flow of processing for specifying a managed server to be stopped according to the fifth embodiment. 18 is a flowchart showing a flow of processing in the sixth embodiment. 18 is a table showing an example of determining a storage location according to the sixth embodiment. 18 is an example of a table indicating a location of a managed server to be stopped according to the seventh embodiment. FIG. 20 is a device configuration diagram illustrating a hardware configuration of a server system according to an eighth embodiment.

[Embodiment 1]
The present embodiment relates to a program for operating a system in which a plurality of electronic devices that can be substituted for each other are installed on a shelf. When receiving a request for power saving, the program according to the present embodiment determines whether the power consumption of the system can be reduced according to the request, and executes power saving according to the request if possible. To do. The power saving is performed by stopping a part of electronic devices that can be substituted for each other during the operation of the electronic devices included in the system. An electronic device that can replace each other is, for example, a server on which a plurality of virtual machines operate. The system is a server system 10 including a plurality of servers on which a plurality of virtual machines operate and peripheral devices.

  FIG. 1 is an apparatus configuration diagram illustrating a hardware configuration of the server system 10. The server system 10 includes a shelf 11, a plurality of electronic devices installed on the shelf 11, a heat exchanger and a blower (not shown). The shelf 11 has a structure in which a plurality of electronic devices arranged in the height direction can be fixed and installed, for example, a 19-inch rack. The dimensions of the 19-inch rack and the electronic equipment installed in the 19-inch rack are defined by EIA (Electronic Industries Alliance) standards. In a 19-inch rack, one section having a reference height is called one unit. The height of one unit compartment is about 45 millimeters. The height of the electronic device used in combination with the 19-inch rack is designed to be installed in a section whose height is an integral multiple of one unit. That is, in a 19-inch rack, one unit is a unit section. Hereinafter, a case where a 42-unit 19-inch rack, that is, a 19-inch rack having 42 sections in the height direction is used for the shelf 11 will be described as an example.

  The electronic devices installed on the shelf 11 are, for example, the communication device 12, the management server 13, and a plurality of managed servers 14. All of these electronic devices have dimensions corresponding to the 19-inch rack standard, and are fixed to the shelf 11. The arrangement of the electronic device will be described later.

  The communication device 12 is, for example, a switching hub or a router. The communication device 12 performs communication between the management server 13 and the managed server 14 and communication between the server system 10 and the external network 31. The communication device 12 and the management server 13, the communication device 12 and the managed server 14, and the communication device 12 and the network 31 are connected by wire or wirelessly.

  The network 31 is a local area network (LAN) in the data center where the server system 10 is installed, a network that connects the data centers using a dedicated line, a network of the Internet line, or the like.

  The management server 13 manages the server system 10. The management server 13 includes a CPU (Central Processing Unit) 21, a main storage device 22, an auxiliary storage device 23, a communication interface 24, a clock 25, and a bus.

  The CPU 21 is an arithmetic control device that executes a program according to the present invention. As the CPU 21, one or a plurality of CPUs or a multi-core CPU is used. The CPU 21 is connected to each hardware part constituting the management server 13 via a bus.

  The main storage device 22 is a storage device such as an SRAM (Static Random Access Memory), a DRAM (Dynamic Random Access Memory), or a flash memory. The main storage device 22 temporarily stores information necessary during the processing performed by the CPU 21 and a program being executed by the CPU 21.

  The auxiliary storage device 23 is a storage device such as an SRAM, a flash memory, a hard disk, or a magnetic tape. The auxiliary storage device 23 stores a program to be executed by the CPU 21 and various data necessary for executing the program.

  The clock 25 outputs time. The clock 25 is preferably capable of adjusting the time with a standard time such as Japan standard time via the communication device 12 and the network 31.

  The communication interface 24 is an interface that performs communication between the management server 13 and the communication device 12.

  The managed server 14 performs various types of information processing. The managed server 14 includes a CPU, a main storage device, an auxiliary storage device, and a communication interface (not shown). The function of each component is the same as that of the management server 13.

  The managed server 14 is equipped with a large-capacity main storage device and can operate a plurality of virtual machines. Each virtual machine operates as a virtual computer and executes various application software. The management server 13 manages, via the communication device 12, which of the plurality of managed servers 14 the virtual machine that causes which managed server 14 to execute which application software is arranged.

  A virtual machine arranged in one managed server 14 can be moved to another managed server 14 during operation of the server system 10. The management server 13 moves the virtual machine appropriately to level the CPU and input / output loads of the managed server 14 and increase the processing efficiency of the entire server system 10. That is, the plurality of managed servers 14 can be substituted for each other.

  An operation terminal 15 is connected to the network 31. The operation terminal 15 is, for example, an information device such as a general-purpose personal computer, tablet, or smartphone installed in a data center management room.

  The operation terminal 15 includes a CPU (not shown), a main storage device, an auxiliary storage device, a communication interface, an input device, and a display device. Functions of a CPU, a main storage device, an auxiliary storage device, and a communication interface (not shown) are the same as those of the management server 13. An input device (not shown) is an input device such as a keyboard, a touch panel, a touch pen, and a microphone. A display device (not shown) is an output device such as a liquid crystal display, a speaker, or headphones.

  The operation terminal 15 receives an input from an input device (not shown) and transmits it to the network 31 and outputs it to a display device (not shown) in response to an instruction from the network 31. A heat exchanger and a blower (not shown) will be described later.

  A large number of server systems 10 (not shown) are connected to the network 31.

  FIG. 2 is an explanatory diagram showing the configuration of the data center. The data center is a facility that operates a large number of server systems 10, storage devices, communication facilities, and cooling devices that cool these electronic devices. FIG. 2 shows an example of a data center using an indirect water cooling type cooling device.

  As described above, electronic devices such as the communication device 12, the management server 13, and the managed server 14 are fixed to the shelf 11 of the server system 10. Each electronic device is installed in a row in the height direction of the shelf 11. Each electronic device is fixed to the shelf 11. The shelf 11 is provided with a blower 18 capable of blowing air toward the side surface of each electronic device. A heat exchanger 19 is installed adjacent to the blower 18. Operations of the blower 18 and the heat exchanger 19 will be described later.

  The cooling device includes a refrigerator 33, a buffer tank 34, a refrigerant pipe 35, a pump 36, a sensor 37, a valve 38, and a cooling control device (not shown).

  The refrigerant pipe 35 connects the refrigerator 33 and the buffer tank 34. Further, the refrigerant pipe 35 is branched into a refrigerant branch pipe 351 in the vicinity of each server system 10. The refrigerant branch 351 is connected to the heat exchanger 19. A plurality of heat exchangers 19 are connected in parallel to the refrigerator 33 and the buffer tank 34 by the refrigerant branch 351.

  The refrigerator 33 cools the refrigerant flowing in the refrigerant pipe 35. The buffer tank 34 temporarily stores the refrigerant before and after being cooled by the refrigerator 33 to prevent sudden fluctuations in the load of the refrigerator 33 and rapid fluctuations in the temperature of the refrigerant flowing in the refrigerant pipe 35. In addition, in the buffer tank 34, it isolate | separates so that the refrigerant | coolant before cooling and the refrigerant | coolant after cooling may not be mixed.

  The pump 36 causes the refrigerant in the refrigerant pipe 35 to flow in a predetermined direction. The refrigerant cooled by the refrigerator 33 enters the heat exchanger 19 through the refrigerant pipe 35 and the refrigerant branch pipe 351. The refrigerant is warmed in the heat exchanger 19. The heat transfer in the server system 10 will be described later.

  The refrigerant warmed by the heat exchanger 19 moves to the refrigerator 33 via the buffer tank 34 and is cooled. The refrigerant cooled by the refrigerator 33 is sent to the heat exchanger 19 through the buffer tank 34, the refrigerant pipe 35 and the refrigerant branch pipe 351. The refrigerant circulates between the refrigerator 33 and the heat exchanger 19 and is repeatedly used. In the present embodiment, water is used as the refrigerant.

  The sensor 37 measures the temperature of the refrigerant from the buffer tank 34 toward the server system 10. The valve 38 adjusts the flow rate of the refrigerant in the refrigerant pipe 35 and the refrigerant branch pipe 351. Although two valves 38 are illustrated in FIG. 2, the number of valves 38 is not limited to two. The valve 38 is appropriately installed at a place where the flow rate needs to be adjusted.

  The refrigerator 33, the pump 36, the sensor 37, and the valve 38 are connected to a cooling control device (not shown), and the operation of each part is controlled so as to keep the temperature measured by the sensor 37 constant. Although one sensor 37 is illustrated in FIG. 2, the number of sensors 37 is not limited to one. The sensor 37 is appropriately installed at a place where temperature measurement is necessary.

  FIG. 3 is an explanatory diagram showing a cooling method of the server system 10. FIG. 3 is a schematic view of the server system 10 cut along a TT section. That is, FIG. 3 shows a horizontal section when the shelf 11 is horizontally installed using a 19-inch rack.

  The TT cross section is a cross section passing through the managed server 14. The blower 18 creates an air flow indicated by a white arrow inside the server system 10. The heat transfer in the server system 10 will be described with reference to FIG.

  The air sent out from the blower 18 flows around the managed server 14 and cools the electronic components constituting the managed server 14. The air heated by the heat generated by the electronic components constituting the managed server 14 passes through the heat exchanger 19. A refrigerant branch 351 passes through the heat exchanger 19. The refrigerant that has entered the heat exchanger 19 through the refrigerant branch 351 is warmed by the air in the heat exchanger 19, thereby cooling the air in the heat exchanger 19. The air cooled by the heat exchanger 19 is sent out from the blower 18 toward the managed server 14 again.

  The refrigerant warmed in the heat exchanger 19 by the action of the pump 36 goes out of the heat exchanger 19 through the refrigerant branch pipe 351, so that the heat generated in the server system 10 is also carried out of the server system 10. . As described above, the refrigerant heated in the heat exchanger 19 is cooled by the refrigerator 33 via the buffer tank 34.

  FIG. 4 is an explanatory diagram showing the relationship between the CPU usage rate and the power consumption. FIG. 4 shows an example of the relationship between the usage rate of the CPU mounted on the managed server 14 and the power consumption. The horizontal axis in FIG. 4 indicates the CPU usage rate, and the unit is percent. The vertical axis in FIG. 4 indicates power consumption, and the unit is watts. Here, the CPU usage rate means a ratio of time during which the CPU mounted on the managed server 14 is performing some processing.

  The CPU usage rate changes from moment to moment depending on the execution status of the program. The server system 10 is often used in a range where the average value of CPU utilization of each managed server 14 is about 25% to 40%. In the following description, an average value of CPU usage rates within a certain time is referred to as CPU usage rate. Here, the fixed time is, for example, 10 minutes.

  In FIG. 4, the power consumption when the CPU usage rate is 0% is about 260 W. As the CPU usage rate increases, the power consumption increases. When the CPU usage rate is 35%, the power consumption is about 400W. In the managed server 14 having different specifications, the relationship between the CPU usage rate and the power consumption is different from that in FIG. 4, but the relationship that the power consumption increases when the CPU usage rate increases is the It is common regardless of the specifications of the server 14.

  FIG. 5 is an explanatory diagram showing a relationship between the number of managed servers 14 stopped, power consumption, and CPU usage rate. In the following, a case will be described in which the management server 13 manages the CPU usage rates of the managed servers 14 that are operating so that they are all equal. In reality, despite the management by the management server 13, there are some differences in the CPU usage rate between the managed servers 14. However, in general, the following explanations apply.

  FIG. 5 shows the power consumption and the operation of the managed server 14 in operation when some of the managed servers 14 are stopped while keeping the processing capacity constant in the server system 10 including 20 managed servers 14. Indicates the CPU usage rate.

  In FIG. 5, the initial CPU usage rate of the managed server 14 is 35%. Further, the processing capacity is made constant by moving virtual machines operating on some managed servers 14 to other managed servers 14 and stopping the managed servers 14 not operating virtual machines. keep.

  The horizontal axis in FIG. 5 indicates the number of managed servers 14 to be stopped. The unit of the horizontal axis in FIG. The vertical axis on the left side of FIG. 5 indicates the total power consumption of the managed server 14. The unit of the vertical axis on the left side of FIG. 5 is watts. The relationship between the number of managed servers 14 to be stopped and the total power consumption of the managed servers 14 is indicated by black diamonds. The vertical axis on the right side of FIG. 5 indicates the CPU usage rate of the managed server 14. The unit of the vertical axis on the right side of FIG. 5 is percentage. The relationship between the number of managed servers 14 stopped and the total power consumption of the managed servers 14 is indicated by a square.

  The power consumed by the managed server 14 eventually becomes heat. Therefore, the greater the power consumption, the greater the amount of heat generated and the temperature rises.

  When the number of managed servers 14 to stop increases, the total power consumption of the managed servers 14 decreases. In addition, when the number of managed servers 14 to be stopped increases, the CPU usage rate of the managed server 14 in operation increases. When the number of stops is 13 or more, the CPU usage rate is 100%. In other words, CPU operation time is not allocated to some virtual machines, and execution stops.

  FIG. 6 is an explanatory diagram showing the configuration of the electronic device installed on the shelf 11. As described above, in this embodiment, a 42-unit 19-inch rack is used. One unit of the communication device 12 is installed in the uppermost section, and one unit of the management server 13 is installed in the lower section. In the section below the management server 13, 20 units of two managed servers 14 from the first managed server 14 to the twentieth managed server 14 are installed. For the following explanation, the left side of FIG. 6 shows a number representing a section in the shelf 11 where the managed server 14 is installed. The number is assigned to each unit section.

  7A to 7C are explanatory diagrams for explaining a change in temperature of the server system 10 in which some of the managed servers 14 are stopped. FIG. 7A shows the server system 10 in a normal state. FIG. 7B shows the server system 10 of the present embodiment during power saving. FIG. 7C shows a server system 10 of a comparative example during power saving.

  SV indicates the managed server 14 that is operating, and X indicates the managed server 14 that is stopped. The white arrow indicates the air flow created by the blower 18. In FIG. 7, the left side shows the flow of air from the blower 18 toward the electronic device, and the right side shows the flow of air toward the heat exchanger 19, which is warmed by the heat generated by the electronic device. In the following description, air from the blower 18 toward the electronic device is referred to as intake air, and air from the electronic device to the heat exchanger 19 is referred to as exhaust air.

  FIG. 7A shows a case where the communication device 12, the management server 13, and the 20 managed servers 14 are all operating. 7B and 7C show a case where 11 managed servers 14 are stopped in order to respond to a request for power saving. FIG. 7B shows the case of this embodiment, and the locations of the managed servers 14 to be stopped are distributed. FIG. 7C shows the case of the comparative example, and the managed servers 14 to be stopped are concentrated on the upper side of the shelf 11.

  In the case of FIG. 7A, the average CPU usage rate of the 20 managed servers 14 operating is 35%, the average power consumption per unit is 400 W, and the total power consumption of the managed servers 14 is about 8000 W. . In the case of FIG. 7B and FIG. 7C, the CPU usage rate of the nine managed servers 14 that are operating averages 78%, the average power consumption per unit is about 600 W, and the total power consumption of the managed servers 14 is It is about 5400W. In either case, the intake and exhaust air flowing through the shelf 11 is 2500 cubic meters per hour and the ambient temperature is 25 degrees Celsius.

Table 1 shows an example of the intake air temperature and the exhaust gas maximum temperature in FIGS. 7A, 7B and 7C when the steady state is reached.

  In either case, the temperature of the exhaust gas is higher than the temperature of the intake air due to the heat generated by the managed server 14. In the case of FIG. 7B and FIG. 7C, the power consumption per one managed server 14 that operates is larger than that in the case of FIG. 7A, so the amount of heat generated per one managed server 14 that operates. Will also increase. However, in the case of FIG. 7B where the locations of the managed servers 14 that are operating are dispersed, the maximum temperature of the exhaust gas is lower than that in the case of FIG. 7A.

  On the other hand, in the case of FIG. 7C, which is a comparative example, since the locations of the managed servers 14 that are operating are concentrated, a sufficient cooling effect cannot be obtained, and the ambient temperature of the managed server 14 that is operating Rises. For this reason, the temperature of the intake air is higher than the ambient temperature. Further, since the temperature of the intake air is rising and the amount of heat generated by the managed server 14 that is operating is increasing, the maximum temperature of the exhaust gas is rising nearly 10 degrees compared to the case of FIG. 7A. .

  When the temperature rises, the electronic device is likely to fail. In addition, the CPU throttling function built in the CPU in order to suppress the heat generation amount and avoid the thermal runaway may operate, and the performance of the managed server 14 may deteriorate.

  As described above, when a part of the managed server 14 is stopped to save power, the temperature rise can be reduced by distributing the stopped managed servers 14 as shown in FIG. 7B. . By reducing the temperature rise, the power consumption of the cooling device can be reduced and power can be saved.

  FIG. 8 is a flowchart showing the flow of processing of the program. The flow of processing performed by the CPU 21 of this embodiment will be described using FIG. The CPU 21 determines whether or not a power saving request has been acquired (step S500).

  Here, the request for power saving is notified in advance from, for example, a power provider to a data center manager who is a power consumer. Alternatively, it is distributed to the manager of the power company data center as an electricity rate table that is set so that the fee increases when a certain amount of power is used in a specific time zone.

  The manager of the data center assigns to each server system 10 a time zone for limiting the amount of power used and an upper limit value of the amount of power that can be used in the limited time zone. The assignment may be determined by the administrator each time or may be automatically performed by the management system. This assignment is a request for power saving to the server system 10. The CPU 21 acquires an assignment input via the network 31.

  In the following, a case will be described as an example where the power consumption of the communication device 12 and the management server 13 is not changed, and the request for power saving is met by reducing the power consumption of the managed server 14.

  If the request for power saving has not been acquired (NO in step S500), the CPU 21 returns to step S500. If a request for power saving has been acquired (YES in step S500), the CPU 21 starts a subroutine for calculating the number of stopped vehicles (step S501). The subroutine for calculating the number of stops is a subroutine for calculating the number of managed servers 14 to be stopped in response to a request for power saving and notifying whether or not the requested power saving amount can be achieved. Processing in the subroutine will be described later.

  The CPU 21 determines whether or not the target indicated by the power saving request can be achieved by stopping a part of the managed server 14 (step S502). Specifically, it is determined whether or not the fact that the target can be achieved has been output from the subroutine for calculating the number of stops started in step S501. If the target cannot be achieved (NO in step S502), the CPU 21 ends the process.

  If the power consumption target indicated by the power saving request can be achieved (YES in step S502), the CPU 21 starts a specific subroutine for the managed server 14 to be stopped (step S503). The subroutine for specifying the managed server 14 to be stopped is a subroutine for specifying a section in the shelf 11 where the managed server 14 to be stopped is installed. Processing in the subroutine will be described later.

  The CPU 21 predicts the time required for stopping the managed server 14 (step S504). When stopping the managed server 14, the task and virtual machine operating on the CPU of the managed server 14 are moved to the managed server 14 that does not stop, and then the stop process is performed. In step S504, the CPU 21 predicts the time required for these processes.

  An example of a method for predicting the required time will be described. The CPU 21 predicts the required time by totaling approximate values such as the time for transferring the contents of the main memory in which the virtual machine is operating to another managed server 14 and the time for switching input / output to the virtual machine. . Here, the time for transferring the contents of the main memory is calculated from the amount of main memory used, and the time for switching input / output is calculated from the number and type of input / output used. The CPU 21 may calculate the time required for the stop operation from the product of the number of virtual machines to be moved to the other managed server 14 and the average time required for the movement of the virtual machines obtained in advance. Alternatively, the CPU 21 may predict that the required time is a certain time, for example, 5 minutes, regardless of the number of operating virtual machines.

  The CPU 21 calculates the stop work start time of the managed server 14 from the start time of the time zone specified in the power saving request and the required time predicted in step S504. The CPU 21 refers to the clock 25 to determine whether or not the work start time has been reached (step S505). If the work start time has not been reached (NO in step S505). The CPU 21 returns to step S505.

  When the work start time has been reached (YES in step S505), the CPU 21 moves the task and virtual machine that were operating on the managed server 14 to be stopped to the managed server 14 that is not to be stopped (step S506). Thereafter, the CPU 21 stops the managed server 14 (step S507). Thereafter, the server system 10 operates with an average power amount within the range of the power amount allocated in response to the request for power saving.

  The CPU 21 determines whether or not the power saving time period for which the request has been received has ended (step S508). If not completed (NO in step S508), the CPU 21 returns to step S508. If it has been completed (YES in step S508), the CPU 21 restarts the managed server 14 stopped in step S507 (step S509). The CPU 21 rearranges the operating task and the virtual machine (step S510). The rearrangement is performed, for example, so as to equalize the CPU usage rate of each managed server. Since the existing method can be used for the method of determining the placement of the virtual machine with respect to the managed server 14, the description is omitted. Thereafter, the CPU 21 ends the process.

  FIG. 9 is a flowchart showing the flow of processing for calculating the number of stopped vehicles. The flow of processing of the subroutine for calculating the number of stops will be described with reference to FIG.

  The CPU 21 sets the calculated number to the number of managed servers 14, that is, 20 (step S521). The CPU 21 acquires the upper limit value of the power usage based on the power saving request notification received in advance (step S522). Note that the upper limit value of power consumption in step S522 in the present embodiment is the upper limit value of power consumed by the managed server 14 included in the server system 10, and includes the communication device 12, the management server 13, and the blower. 18 and the heat consumed by the heat exchanger 19 are not included.

  The CPU 21 obtains a predicted value of the CPU usage rate of the managed server 14 in the time zone when the request for power saving is received (step S523). As the predicted value of the CPU usage rate, for example, a past statistical value of the CPU usage rate corresponding to the target time and day of the week is used. Statistical values of fluctuations of the CPU usage rate depending on the day of the week and time are stored in advance in the auxiliary storage device 23. Alternatively, the CPU 21 may acquire a statistical value of the variation of the CPU usage rate according to the day of the week and time from an external database via the communication device 12 and the network 31.

  Depending on the application of the virtual machine operating on the server system 10, the CPU 21 obtains a predicted value of the CPU usage rate in consideration of information other than time and day of the week, for example, information such as weather forecast, stock price or foreign exchange rate. May be. Further, in the case of the server system 10 in which the CPU usage rate of the managed server 14 does not have a particularly defined fluctuation pattern, the CPU 21 uses the CPU usage rate average value for several minutes before the execution of step S523. It may be used for the predicted value of rate.

  The CPU 21 calculates a predicted value of power usage based on the predicted value of CPU usage acquired in step S523 (step S524). In calculating the predicted value of power consumption, an example is shown in FIG. 4 on the assumption that the CPU usage rate is managed by the management server 13 so that all the managed servers 14 in operation are equal. The relationship between CPU usage rate and power consumption is used. The relationship between the CPU usage rate and the power consumption is stored in advance in the auxiliary storage device 23 as a mathematical expression for obtaining the power usage amount using the table or the CPU usage rate as an argument.

  The CPU 21 determines whether or not the predicted value of power usage calculated in step S524 exceeds the upper limit value acquired in step S522 (step S525). When the predicted value of the power consumption is equal to or less than the upper limit value (NO in step S525), the CPU 21 outputs that the target can be achieved (step S526). The output destination is, for example, an administrator who has assigned and notified a limit on the amount of power used. Thereafter, the CPU 21 ends the process.

  When the predicted value of the power consumption exceeds the upper limit value (YES in step S525), the CPU 21 reduces the number of calculations by one (step S527). The CPU 21 determines whether or not the calculated number is 1 or more (step S528). If the calculated number is 1 or more (YES in step S528), the CPU 21 calculates the CPU usage rate of the managed server 14 (step S529). The predicted value of the CPU usage rate of the managed server 14 is calculated by dividing the product of the predicted value of the CPU usage rate acquired in step S523 and the number of managed servers 14 by the calculated number.

  The CPU 21 determines whether or not the calculated predicted value of the CPU usage rate is equal to or less than the maximum usage rate of the CPU of the managed server 14 (step S530). Here, the maximum usage rate of the CPU is a value defined in advance, and is the maximum usage rate of the CPU at which the managed server 14 can operate stably. The maximum usage rate of the CPU is determined according to the usage of the server system 10 and the like. When the time change of the CPU usage rate of the managed server 14 is severe or when real-time processing is necessary, the maximum CPU usage rate is set to a value such as 60%. On the other hand, when the time change of the CPU usage rate of the managed server 14 is small, or when real-time processing is unnecessary and the CPU load is high, the maximum usage rate is set to 100% for applications where there is no problem even if the processing is delayed. May be. When the predicted value of the CPU usage rate of the managed server 14 is equal to or less than the maximum usage rate (YES in step S530), the CPU 21 returns to step S524.

  If the calculated number is less than 1 in step S528 (NO in step S528) and if the predicted value of the CPU usage rate of the managed server 14 exceeds the maximum usage rate in step S530 (NO in step S530), the CPU 21 achieves the target. The fact that it is impossible is output (step S531). Thereafter, the CPU 21 ends the process.

  At the end of the subroutine for calculating the number of stops, the CPU 21 stores in the main storage device 22 or the auxiliary storage device 23 information indicating which output of step S526 or step S527 was performed. Furthermore, when it is determined NO in step S525, the CPU 21 stores the value of the previous calculated number of units and the predicted value of the CPU usage rate in the main storage device 22 or the auxiliary storage device 23. When executing the calling program of this subroutine, the CPU 21 uses the stored value of the calculated number for the calculation result of the stopped number. Similarly, at the time of execution of the calling source program of this subroutine, the CPU 21 uses the stored predicted CPU usage rate as the predicted CPU usage rate when some managed servers are stopped.

  FIG. 10 is an example of a table indicating the location of the managed server 14 to be stopped. The contents of processing of a specific subroutine for the managed server 14 to be stopped will be described with reference to FIG.

  The table shown in FIG. 10 includes the number of managed servers 14 to be stopped, the predicted value of the CPU usage rate of the managed servers 14 after stopping some managed servers 14, and the shelf 11 of the managed servers 14 to be stopped. It is a table which shows the relationship with the number of the inside section. The leftmost column indicates the number of managed servers 14 to be stopped calculated by the subroutine for calculating the number of stops. The second column from the left indicates the predicted value of the CPU usage rate of the managed server 14 calculated in step S529. The second row from the top of the third and subsequent columns from the left indicates the partition numbers in the shelf 11. The partition numbers are indicated by the numbers on the left side of FIG. The location indicated by X in the third and lower rows from the top of the third and subsequent columns from the left indicates that the corresponding managed server 14 is stopped.

  More specifically, when four managed servers 14 are stopped and the predicted value of the CPU usage rate is set to 60%, they are installed in the sections corresponding to the numbers 9, 19, 29, and 39. The four managed servers 14 are stopped. That is, since the managed server 14 of 2 units is used in this embodiment, the 9th and 10th partitions, the 10th and 20th partitions, the 29th and 30th partitions, and the 39th and 40th partitions. The managed servers 14 respectively stored in the sections are stopped.

  FIG. 10 is created by obtaining the location of the managed server 14 to be stopped by using a method such as actual measurement or thermal simulation so that the maximum exhaust temperature can be kept low. In particular, by creating FIG. 10 by actual measurement after installing the server system 10 in the data center, individual differences of the managed servers 14 being used, the air conditioning status of the installation location of the server system 10, and between each electronic device The location of the managed server 14 to be stopped that can keep the maximum exhaust temperature low can be determined by combining various situations such as the state of the wiring to be connected.

  The table of FIG. 10 is stored in the auxiliary storage device 23. The CPU 21 reads the table shown in FIG. 10 from the auxiliary storage device 23 and uses it. Alternatively, the CPU 21 may acquire the table of FIG. 10 from an external storage device via the communication device 12 and the network 31.

  FIG. 11 is a flowchart showing a flow of processing for specifying a managed server to be stopped. The process flow of a specific subroutine for the managed server to be stopped will be described with reference to FIG.

  The CPU 21 acquires the number of managed servers to be stopped from the main storage device 22 or the auxiliary storage device 23 (step S701). As described above, the number of managed servers to be stopped is the value of the calculated number immediately before when NO is determined in step S525 of the stopped number calculation subroutine shown in FIG.

  The CPU 21 acquires a predicted value of the CPU usage rate from the main storage device 22 or the auxiliary storage device 23 (step S702). As described above, the predicted value of the CPU usage rate is the predicted value of the CPU usage rate immediately before it is determined NO in Step S525 of the stop number calculation subroutine shown in FIG.

  The CPU 21 acquires a table indicating the location of the managed server 14 to be stopped from the main storage device 22, the auxiliary storage device 23, or an external storage device (step S703). As described above, the table indicating the location of the managed server 14 to be stopped is, for example, the table shown in FIG.

  The CPU 21 refers to the table acquired in step S703 from the number of managed servers to be stopped and the predicted value of the CPU usage rate, and acquires the number of the partition where the managed server 14 to be stopped is installed (step S704). .

  The CPU 21 stores the acquired number in the main storage device 22 or the auxiliary storage device 23, and thereafter ends the processing.

  According to the present embodiment, when a request for power saving is received, it is determined whether or not the power consumption of the electronic device can be reduced. If possible, the server system 10 that performs power saving according to the request is provided. Can be provided.

  The method for cooling the server system 10 may be a method of cooling the entire room where the server system 10 is installed, or a method of introducing outside air into the room or the cooling pipe.

  The program of the present embodiment may be executed by a virtual machine that operates on any managed server 14.

  The shelf 11 may be divided into two or more. Further, the shelf 11 may have a dimension defined in Japanese Industrial Standards JIS (Japanese Industrial Standards) C6010 “General Electronic Equipment Racks and Unit Chassis Dimensions” or a dimension uniquely defined by the server system 10 manufacturer. . In these cases, the electronic equipment to be installed has specifications that can be installed on the shelf 11. The number of unit sections constituting the shelf 11 can be arbitrarily selected.

  The shelf 11 may be one in which sections for installing electronic devices are arranged in a horizontal direction.

  The managed servers 14 having different relationships between the CPU usage rate and the power consumption may be mixedly mounted on the shelf 11. In this case, in step S524, a predicted value of power consumption is calculated using the relationship between the CPU usage rate of each managed server 14 and the power consumption. Further, Table 10 is prepared in consideration of the heat generation amount of each managed server.

  The shelf 11 may have an empty section where no electronic device is installed. In that case, the table 10 in consideration of the empty section is prepared in advance. The communication device 12 or the management server 13 may be installed near the center of the shelf. In that case, Table 10 is prepared in advance considering the installation of the communication device or the management server.

  One unit of managed server 14 or three or more units of managed server 14 may be used. Two or more units of communication device 12 or two or more units of management server 13 may be used.

[Embodiment 2]
The present embodiment relates to a program or the like that determines a section where an electronic device that stops in response to a request for power saving is installed based on a calculation result represented by a mathematical expression. Also in this embodiment, a server system 10 in which 20 2 managed servers 14 are arranged in a 42 unit 19-inch rack will be described as an example. Description of portions common to the first embodiment is omitted.

  FIG. 12 is a flowchart illustrating a flow of specific processing for the managed server 14 to be stopped according to the second embodiment. In the present embodiment, the managed server 14 to be stopped is the same as that of the first embodiment except for a specific subroutine. The flow of specific processing for the managed server 14 to be stopped according to this embodiment will be described with reference to FIG.

  The CPU 21 sets the counter k to 1 which is an initial value (step S551). The CPU 21 calculates the total number j of unit partitions storing the managed server 14 by multiplying the number hsv of unit partitions constituting the partition in which one managed server 14 is installed by the number S of managed servers 14. (Step S552). Specifically, since 2 units of managed servers 14 are used, hsv is 2, and since 20 managed servers 14, S is 20. Therefore, the total number j of unit partitions storing the managed server 14 obtained by multiplying hsv by S is 40.

ｋは正の整数。
ｙ（ｋ）≦ｊ
ｙ（ｋ）は停止する被管理サーバ１４が設置されている単位区画の番号
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｎは停止する被管理サーバ１４の数。 CPU21 calculates y (k) based on Formula (1) (step S553).

k is a positive integer.
y (k) ≦ j
y (k) is the unit partition number where the managed server 14 to be stopped is installed. round (x) is a function that rounds x to obtain an integer.
n is the number of managed servers 14 to be stopped.

  Each variable and constant in the formula (1) will be described more specifically. y (k) is a number indicating a section in the shelf 11 shown on the left side of FIG. The numbers are assigned sequentially from 1 to the section where the managed server 14 is installed. n is the number of managed servers 14 to be stopped calculated based on the upper limit value of the electric energy and the predicted value of the CPU usage rate of the managed server 14. For example, n is calculated based on the flow of FIG. 9 described in the first embodiment.

  The CPU 21 determines whether or not y (k) calculated based on the formula (1) is equal to or less than the total number j of the electronic device or the unit section that stores the electronic device (step S554). If y (k) is equal to or less than j (YES in step S554), the CPU 21 sets the managed server 14 installed in the section y (k) as a managed server 14 that stops in response to a request for power saving. The data is stored in the main storage device 22 or the auxiliary storage device 23 (step S555). The CPU 21 adds 1 to the counter k (step S556). Thereafter, the CPU 21 returns to step S553.

  If y (k) is larger than j (NO in step S554), the CPU 21 ends the process.

  FIG. 13 is a table showing an example of specifying the location where the managed server 14 to be stopped according to the second embodiment is arranged. FIG. 13 is a table calculated from the equation (1). The usage of the value of y (k) obtained from the flowchart of FIG. 12 will be described with reference to FIG.

  The second line from the top shows the number n of managed servers 14 to be stopped calculated by the subroutine for calculating the number of stops. For example, when the number n of managed servers 14 to be stopped is 4, according to the flowchart of FIG. 12, y (1) is 9, y (2) is 19, y (3) is 29, and y (4) is 39. Is obtained. A blank field indicates a portion in which y (k) is not calculated when NO is determined in step S554 in the flowchart of FIG. 12 and the subroutine ends.

  As described above, in this embodiment, two units of the managed server 14 are used, and the section where the managed server 14 is installed is as shown in FIG. Since the fifth managed server 14 is installed at the first location y (1) = 9, the CPU 21 determines that the fifth managed server 14 is the managed server 14 to be stopped. Similarly, the CPU 21 determines that the tenth managed server 14, the fifteenth managed server 14, and the twentieth managed server 14 are the managed servers 14 to be stopped.

  According to the present embodiment, it is possible to appropriately distribute the sections where the electronic devices to be stopped are installed without creating a table indicating the location of the managed server 14 to be stopped in advance by actual measurement or simulation.

[Embodiment 3]
The present embodiment relates to a program or the like for randomly changing an electronic device to be stopped every time a request for power saving is made. Also in this embodiment, a server system 10 in which 20 2 managed servers 14 are arranged in a 42 unit 19-inch rack will be described as an example. A description of portions common to the second embodiment will be omitted.

  FIG. 14 and FIG. 15 are flowcharts showing the flow of specific processing for the managed server 14 to be stopped according to the third embodiment. In the present embodiment, the managed server 14 to be stopped is the same as the second embodiment except for a specific subroutine. The flow of processing for specifying the managed server 14 to be stopped according to this embodiment will be described with reference to FIGS.

  The CPU 21 sets the counter k to 1 which is an initial value (step S551). The CPU 21 calculates the total number j of unit partitions storing the managed server 14 by multiplying the number hsv of unit partitions constituting the partition in which one managed server 14 is installed by the number S of managed servers 14. (Step S552). Specifically, since 2 units of managed servers 14 are used, hsv is 2, and since 20 managed servers 14, S is 20. Therefore, the total number j of unit partitions storing the managed server 14 multiplied by hsv and S is 40.

ｋはｊ以下の整数。
Ｑ（ｋ）はｋの関数。
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｎは停止する被管理サーバ１４の数 CPU21 calculates Q (k) based on Formula (2) (step S572).

k is an integer of j or less.
Q (k) is a function of k.
round (x) is a function that rounds x to obtain an integer.
n is the number of managed servers 14 to stop

  Each variable and constant in Expression (2) will be described more specifically. Q (k) is a variable used during the processing of the present embodiment, and has no physical meaning. n is the number of managed servers 14 to be stopped calculated based on the upper limit value of the electric energy and the predicted value of the CPU usage rate of the managed server 14. For example, n is calculated based on the flow of FIG. 9 described in the first embodiment.

  The CPU 21 determines whether or not the counter k is equal to or less than the total number j of unit partitions that store electronic devices (step S573). If the counter k is equal to or less than j (YES in step S573), the CPU 21 stores Q (k) obtained in step S574 in the auxiliary storage device 23 (step S574). The CPU 21 adds 1 to the counter k (step S575). Thereafter, the CPU 21 returns to step S572.

  If the counter k is greater than j (NO in step S573), the CPU 21 extracts the minimum value from Q (k) stored in the auxiliary storage device 23 and substitutes it for the variable M (step S576). The CPU 21 generates an integer random number R between 0 and M (step S577).

The CPU 21 returns the counter k to 1 which is an initial value (step S578). CPU21 calculates y (k) based on Formula (3) (step S580).

  The CPU 21 determines whether or not y (k) calculated based on the formula (3) is equal to or less than the total number j of unit sections that store the electronic device (step S581). If y (k) is equal to or less than j (YES in step S581), the CPU 21 stops the managed server 14 installed in the section having the number y (k) in response to the request for power saving. 14 is stored in the auxiliary storage device 23 (step S582). Thereafter, the CPU 21 returns to step S578.

  If y (k) is larger than j (NO in step S581), the CPU 21 ends the process.

  FIG. 16 is a table showing an example in which the location of the managed server 14 to be stopped is specified according to the flowcharts shown in FIGS. 14 and 15. FIG. 16 shows an example in which y (k) is obtained by repeatedly executing the flowcharts shown in FIGS. 14 and 15 nine times when the number n of managed servers 14 to be stopped is three.

  The second random number line from the top indicates the random number R generated in step S577. Since M is 13 in step S576, the random number generated in step S577 is an integer from 0 to 13. The next three lines show the numbers of the sections in the shelf 11 where the managed server 14 to be stopped is installed. The division numbers are indicated by the numbers on the left side of FIG.

  More specifically, in response to the first reduction request, the three managed servers 14 installed in the sections corresponding to the numbers 11, 23, and 37 are stopped. That is, in this embodiment, since 2 units of the managed server 14 are used, they are stored in the 11th and 12th sections, the 23rd and 24th sections, and the 37th and 38th sections, respectively. The managed server 14 is stopped.

  In response to the second reduction request, the three managed servers 14 installed in the sections corresponding to the numbers 5, 17, and 31 are stopped. That is, the managed servers 14 stored in the 5th and 6th partitions, the 17th and 18th partitions, and the 31st and 32nd partitions are stopped.

  According to the present embodiment, the electronic device to be stopped is randomly changed each time a request for power saving is made, so that the usage status of the managed server 14 can be prevented from being shifted. Therefore, it is possible to prevent the secular change due to use and to extend the life of the server system 10.

  R may be an integer that changes from 0 to M in ascending or descending order each time a request for power saving is obtained. By doing so, it is possible to prevent a shift in the usage status of the managed server 14 without using the random number generation function.

[Embodiment 4]
The present embodiment relates to a task having a low priority, a program for stopping a virtual machine, and the like when necessary to satisfy a request for power saving. Description of portions common to the first embodiment is omitted.

  FIG. 17 is a flowchart showing a flow of processing of the program according to the fourth embodiment. 18 and 19 are flowcharts showing the flow of processing for calculating the number of stops according to the fourth embodiment. The flow of this embodiment will be described with reference to FIGS. Description of portions common to the first embodiment is omitted.

  The flow of processing performed by the CPU 21 of this embodiment will be described using FIG. The CPU 21 determines whether or not a power saving request has been acquired (step S500). If the request for power saving has not been acquired (NO in step S500), the CPU 21 returns to step S500.

  If a request for power saving has been acquired (YES in step S500), the CPU 21 starts a subroutine for calculating the number of stopped vehicles (step S601). The subroutine for calculating the number of stopped servers according to the present embodiment is a subroutine for calculating the number of managed servers 14 to be stopped in response to a request for power saving. This subroutine calculates a low-priority task and the number of stopped machines when the virtual machine is stopped when the requested power saving amount cannot be achieved. Processing in the subroutine will be described later.

  The CPU 21 determines whether or not the target indicated by the power saving request can be achieved by stopping a part of the managed server 14 (step S502). If the target cannot be achieved (NO in step S502), the CPU 21 ends the process.

  If the power consumption target indicated by the power saving request can be achieved (YES in step S502), the CPU 21 starts a specific subroutine for the managed server 14 to be stopped (step S503). The subroutine for specifying the managed server 14 to be stopped is a subroutine for specifying a section in the shelf 11 where the managed server 14 to be stopped is installed. Since this subroutine can use the same subroutine as the subroutine used in any of the first to third embodiments, the description thereof will be omitted.

  The CPU 21 predicts the time required for stopping the managed server 14 (step S504). The CPU 21 calculates the stop work start time of the managed server 14 from the start time of the time zone specified in the power saving request and the required time predicted in step S504. The CPU 21 refers to the clock 25 to determine whether or not the work start time has been reached (step S505). If the work start time has not been reached (NO in step S505), the CPU 21 returns to step S505.

  If the action start time has been reached (YES in step S505), the CPU 21 stops the low priority task and virtual machine (step S602). The CPU 21 moves the task and virtual machine that were operating on the managed server 14 to be stopped to the managed server 14 that is not to be stopped (step S506). Thereafter, the CPU 21 stops the managed server 14 (step S507). Thereafter, the server system 10 operates with an average power amount within the range of the power amount allocated in response to the request for power saving.

  CPU21 determines whether the time slot | zone designated by the request | requirement of a power saving was complete | finished (step S508). If not completed (NO in step S508), the CPU 21 returns to step S508. If it has been completed (YES in step S508), the CPU 21 restarts the managed server 14 stopped in step S507 (step S509). The CPU 21 rearranges the operating task and the virtual machine (step S510).

  The CPU 21 restarts the low priority task and the virtual machine that were stopped in step S602 (step S603). Thereafter, the CPU 21 ends the process. Note that after the task and virtual machine stopped in step S602 are restarted (step S603), the operating task and virtual machine may be rearranged (step S510).

  18 and 19 are flowcharts showing the flow of processing for calculating the number of stops according to the fourth embodiment. The processing flow of the subroutine for calculating the number of stops according to this embodiment will be described with reference to FIGS.

  The processes up to step S530 are the same as those in the subroutine for calculating the number of stops according to the first embodiment shown in FIG. If the calculated number is less than 1 in step S528 (NO in step S528) and if the predicted value of the CPU usage rate of the managed server 14 exceeds the maximum usage rate in step S530 (NO in step S530), the CPU 21 calculates The total number of managed servers 14 is set to 20, that is, 20 (step S611).

  The CPU 21 acquires a predicted value of the CPU usage rate of the managed server 14 by the high priority task and the virtual machine in the time zone in which the request for power saving is received (step S612). The priority may be described in advance in a program executed in each virtual machine, or an input of priority determination by a user who gives an instruction to execute each program may be accepted.

  The CPU 21 calculates a predicted value of power usage based on the predicted value acquired in step S612 (step S613). For the calculation of the predicted power consumption, the relationship between the CPU usage rate and the power consumption shown in FIG. 4 is used.

  The CPU 21 determines whether or not the predicted value of power usage calculated in step S613 is below the upper limit value acquired in step S522 (step S614). When the predicted value of the power consumption is equal to or lower than the upper limit value (NO in step S614), the CPU 21 outputs that the target can be achieved when the low priority task and the virtual machine are stopped (step S619). The output destination is, for example, an administrator who has assigned and notified a limit on the amount of power used. Thereafter, the CPU 21 ends the process.

  When the predicted value of the power consumption exceeds the upper limit value (YES in step S614), the CPU 21 reduces the calculated number by one (step S615). The CPU 21 determines whether or not the calculated number is 1 or more (step S616). When the calculated number is 1 or more (YES in step S616), the CPU 21 calculates a predicted value of the CPU usage rate of the managed server 14 (step S617). The predicted value of the CPU usage rate of the managed server 14 is calculated by dividing the product of the predicted value of the CPU usage rate acquired in step S612 and the number of managed servers 14 by the calculated number.

  The CPU 21 determines whether the calculated predicted usage rate is equal to or less than the maximum usage rate of the CPU 21 of the managed server 14 (step S618). When the predicted value of the CPU usage rate of the managed server 14 is equal to or lower than the maximum usage rate (YES in step S618), the CPU 21 returns to step S613.

  If the calculated number is less than 1 in step S616 (NO in step S616), and if the predicted value of the CPU usage rate of the managed server 14 exceeds the maximum usage rate in step S618 (NO in step S618), the CPU 21 sets the target. The fact that it cannot be achieved is output (step S531). Thereafter, the CPU 21 ends the process.

  At the end of the subroutine for calculating the number of stops, the CPU 21 stores in the main storage device 22 or the auxiliary storage device 23 information indicating which output of step S526, step S531, and step S610 has been performed. Furthermore, when it is determined NO in step S525, the CPU 21 stores the value of the previous calculated number of units and the predicted value of the CPU usage rate in the main storage device 22 or the auxiliary storage device 23. If NO is determined in step S614, the CPU 21 obtains the main storage device 22 or the auxiliary storage device with the value of the previous calculated number, the predicted value of the CPU usage rate, and information indicating the task and virtual machine used in step S612. 23.

  When executing the calling program of this subroutine, the CPU 21 uses the stored value of the calculated number for the calculation result of the stopped number. Similarly, when executing the calling source program of this subroutine, the CPU 21 uses the stored predicted CPU usage rate as the predicted CPU usage rate when some of the managed servers 14 are stopped. Further, the CPU 21 uses the information indicating the task and the virtual machine used in step S612 when determining the task and virtual machine whose operation is stopped when executing power saving.

  According to the present embodiment, when it is not possible to satisfy the power saving request with all tasks and virtual machines operating, it is possible to satisfy the power saving request by stopping the low priority tasks and virtual machines. .

  The priority of the task and virtual machine is divided into three or more levels in advance, and it is determined whether or not the request for power saving is satisfied when the lowest priority is stopped. It may be repeated to determine the case where the low one is stopped. In this way, even when a demand for power saving of a strict level is issued, it is possible to respond finely.

  Tasks and virtual machines with low priority may be moved to another server system 10 with a sufficient CPU utilization rate. In this case, the management server 13 communicates with the management server of the other server system 10 connected to the network 31 via the communication device 12, and moves low priority tasks and virtual machines if possible. Let

[Embodiment 5]
The present embodiment relates to a program of a system including a plurality of electronic devices that can be replaced with each other and electronic devices that cannot be replaced with each other. The program or the like of this embodiment determines whether or not the power consumption of the electronic device can be reduced when a request for power saving is received, and executes power saving according to the request if possible. An electronic device that can replace each other is, for example, a managed server 14 on which a virtual machine operates. An electronic device that cannot be substituted for each other is, for example, a storage that stores a large amount of data. The system is a server system 10 including, for example, a plurality of managed servers 14 on which virtual machines operate, a management server 13, a communication device 12, and a storage. Description of portions common to the first embodiment is omitted.

  FIG. 20 is a device configuration diagram illustrating a hardware configuration of the server system 10 according to the fifth embodiment. The server system 10 includes a shelf 11, a plurality of electronic devices installed on the shelf 11, a heat exchanger and a blower (not shown). The shelf 11 is, for example, a 19-inch rack of 42 units, that is, a 19-inch rack having 42 compartments in the height direction.

  The electronic devices installed on the shelf 11 are, for example, the communication device 12, the management server 13, the plurality of managed servers 14, and the storage 28. All of these electronic devices have dimensions corresponding to the 19-inch rack standard, and are fixed to the shelf 11. The installation location of each electronic device will be described later.

  Since the components other than the storage 28 are the same as those in the first embodiment, the description thereof is omitted. The storage 28 is a mass storage device equipped with, for example, several tens of hard disk drives. Even if the storages 28 have the same specifications, the stored data are different. When data stored in one storage 28 is moved or copied to another storage 28, a data transfer time corresponding to the data capacity is required. In response to a request for power saving, it is not realistic to transfer all data in the storage. Therefore, the storage 28 cannot be substituted for each other.

  FIG. 21 is an explanatory diagram illustrating a configuration of an electronic device installed on the shelf 11 according to the fifth embodiment. As described above, in this embodiment, a 42-unit 19-inch rack is used. One unit of the communication device 12 is installed in the uppermost section, and one unit of the management server 13 is installed in the lower section. In the lower section of the management server 13, there are two 2-unit managed servers 14, one 2-unit storage 28, one unit-managed server 14, one two-unit storage 28, It is installed sequentially. A total of 20 managed servers 14 and 8 storages 28 are installed on the shelf 11. For the following explanation, the left side of FIG. 21 shows a number representing the section where the managed server 14 and the storage 28 are installed in the shelf 11. The number is assigned to each unit section. Further, on the right side of FIG. 21, numbers assigned consecutively only to the sections in the shelf 11 where the managed servers 14 are installed are shown.

  22A and 22B are explanatory diagrams illustrating changes in temperature when some managed servers 14 according to the fifth embodiment are stopped. 22A and 22B have the same configuration as that of the server system 10 shown in FIG. SV indicates the managed server 14 that is operating, X indicates the managed server 14 that is stopped, and ST indicates the storage 28. The meanings of the white arrows in FIGS. 22A and 22B are the same as those in FIG.

  FIG. 22A shows the normal operation. All managed servers 14 are operating. FIG. 22B shows a state in which six managed servers 14 are stopped in response to a request for power saving. The managed servers 14 to be stopped are distributed in the shelf 11.

  FIG. 23A and FIG. 23B are explanatory diagrams illustrating a comparative example of a change in temperature when some managed servers of the fifth embodiment are stopped. The systems shown in FIGS. 23A and 23B include the same number of devices as the server system 10 shown in FIG. However, the sections where the devices are installed are different, and the managed server 14 is concentrated on the upper side of the shelf 11 and the storage 28 is concentrated on the lower side of the shelf 11. The meanings of the white arrows in FIGS. 23A and 23B are the same as those in FIG.

  FIG. 23A shows the normal operation. All managed servers 14 are operating. FIG. 23B shows a state in which six managed servers 14 are stopped in response to a request for power saving. The managed servers 14 to be stopped are concentrated in the central part of the shelf 11.

  In the case of FIG. 22A and FIG. 22B, the CPU usage rate of the 12 managed servers 14 operating is 35%, and the power consumption per unit is about 400W. The power consumption of the storage 28 is 400W. The total power consumption of the managed server 14 and the storage 28 is about 8000 W. The power consumption of the storage 28 does not change even in response to a power saving request. In the case of FIGS. 22B and 23B, the CPU usage rate of the six managed servers 14 operating is 70%, the power consumption per unit is about 550 W, and the power consumption of the managed server 14 and the storage 28 is The total is 6500W. In any case, the flow rate of the air flowing through the shelf 11 is 2500 cubic meters per hour, and the ambient temperature is 25 degrees Celsius.

Table 2 shows examples of the intake air temperature and the maximum exhaust gas temperature in FIGS. 7A, 7B, and 7C when the steady state is reached.

In either case, the temperature of the exhaust is higher than the temperature of the intake air due to heat generated by the electronic equipment.
In the case of FIG. 22B and FIG. 23B, since the power consumption per one managed server 14 that operates is greater than that in the case of FIG. 22A and FIG. 23A, one managed server 14 that operates. The amount of heat generated per hit also increases. For this reason, in both FIG. 22B and FIG. 23B, the periphery of the server system 10 becomes hot compared to the normal time, and the temperature of the intake air rises compared to the normal time.

  In the case of FIG. 22B, the storage 28, the managed server 14 that is operating, and the managed server 14 that is stopped are mixed. On the other hand, in the case of FIG. 23B, the storage 28, the managed server 14 that is operating, and the managed server 14 that is stopped are solidified. Therefore, the heat of the managed server 14 operating in FIG. 23B is more difficult to escape than in FIG. 22B, and the maximum exhaust temperature is higher.

  As described above, in the case of the server system 10 in which the managed server 14 and the storage 28 are mixedly mounted, the both are mixed in advance, and the managed servers 14 to be stopped are distributed when responding to the power saving request. The temperature rise can be reduced.

  FIG. 24 is a flowchart illustrating a flow of specific processing for the managed server 14 to be stopped according to the fifth embodiment. The processing flow of this embodiment will be described with reference to FIG.

  The CPU 21 sets the counter k to 1 which is an initial value (step S631). The CPU 21 calculates the total number j of unit partitions storing the managed server 14 by multiplying the number hsv of unit partitions constituting the partition in which one managed server 14 is installed by the number S of managed servers 14. (Step S632). Specifically, since 12 managed servers 14 of 2 units are used, hsv is 2 and S is 12. Therefore, the total number j of unit partitions for storing the managed server 14 multiplied by hsv and S is 24.

ｋは正の整数。
ｙ（ｋ）≦ｊ
ｙ（ｋ）は停止する被管理サーバ１４が設置されている単位区画の番号
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｎは停止する被管理サーバ１４の数 CPU21 calculates y (k) based on Formula (4) (step S633).

k is a positive integer.
y (k) ≦ j
y (k) is the unit partition number where the managed server 14 to be stopped is installed. round (x) is a function that rounds x to obtain an integer.
n is the number of managed servers 14 to stop

  Each variable and constant in Expression (4) will be described more specifically. y (k) is a number indicating the section where the managed server 14 is installed in the shelf 11 shown on the right side of FIG. The numbers are assigned sequentially from 1 to the section where the managed server 14 is installed. n is the number of managed servers 14 to be stopped calculated based on the upper limit value of the electric energy and the predicted value of the CPU usage rate of the managed server 14. For example, n is calculated based on the flow of FIG. 9 described in the first embodiment.

  The CPU 21 determines whether or not y (k) calculated based on the equation (4) is equal to or less than the total number j of unit partitions storing the managed server 14 (step S634). If y (k) is equal to or less than j (YES in step S634), the CPU 21 stops the managed server 14 installed in the section with the number y (k) in response to a request for power saving. 14 is stored (step S635). The CPU 21 adds 1 to the counter k (step S636). Thereafter, the CPU 21 returns to step S633.

  If y (k) is larger than j (NO in step S634), the CPU 21 ends the process.

  According to the present embodiment, even in the server system 10 in which the managed server 14 and the storage 28 are mixedly mounted, it is determined whether or not the power consumption of the electronic device can be reduced when a power saving request is received, When possible, power savings can be implemented on demand.

[Embodiment 6]
The present embodiment relates to an arrangement determination program for determining the arrangement of electronic devices of a system including a plurality of electronic devices that can be replaced with each other and electronic devices that cannot be replaced with each other. An electronic device that can replace each other is, for example, a managed server 14 on which a virtual machine operates. An electronic device that cannot be interchanged is, for example, a storage 28 that stores a large amount of data. The system includes, for example, a plurality of managed servers 14 on which virtual machines operate, management servers 13. The server system 10 includes a communication device 12 and a storage 28. Description of portions common to the first embodiment is omitted.

  FIG. 25 is a flowchart showing a process flow of the sixth embodiment. The processing flow of this embodiment will be described with reference to FIG.

  In the present embodiment, the managed server 14 and the storage 28 installed in the same number of unit units are used. In the following description, the height of both is indicated by the symbol hst.

  The CPU 21 sets the counter I to 1 which is an initial value (step S650). The CPU 21 multiplies hst by the sum of the number m of storages and the number S of managed servers 14 to calculate the total number j of unit partitions that store the storage and managed servers 14 (step S651).

Ｉは正の整数
Ｐ（Ｉ）≦ｊ
Ｐ（Ｉ）はストレージ２８を設置する単位区画の番号。
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。 CPU21 calculates P (I) based on Formula (5) (step S652).

I is a positive integer P (I) ≦ j
P (I) is the number of the unit section in which the storage 28 is installed.
round (x) is a function that rounds x to obtain an integer.

  The CPU 21 determines whether or not P (I) calculated based on Expression (5) is equal to or less than the total number j of unit partitions that store the storage 28 and the managed server 14 (step S653). If P (I) is equal to or less than j (YES in step S653), the CPU 21 stores P (I) in the auxiliary storage device 23 (step S654). The CPU 21 adds 1 to the counter I (step S655). Thereafter, the CPU 21 returns to step S652.

  If P (I) is larger than j (NO in step S653), the CPU 21 ends the process.

  FIG. 26 is a table showing an example of determining the location of the storage 28 according to the sixth embodiment. An example of the case where the total number j of the unit partitions storing the storage 28 and the managed server 141 is 20 and the number of unit partitions hst configuring the storage 28 is 2 is shown in FIG. ) Was calculated. FIG. 26 shows an example of calculation results when the number m of storages 28 is 3, 4, 5, and 6.

  For example, when there are four storage units, P (1) is 5, P (2) is 9, P (3) is 15, and P (4) is 19. In this case, when the server system 10 is installed, 2U storage is provided in the 5th and 6th, 9th and 10th, 15th and 16th, and 19th and 20th partitions, respectively. The managed server 14 is installed in the server. At the time of use, for example, the program shown in the fifth embodiment is used to respond to the request for power saving.

  According to the present embodiment, it is possible to determine the arrangement of electronic devices of the server system 10 that can effectively respond to a request for power saving.

  The arrangement determining program of the present embodiment may be executed by a CPU of a general-purpose personal computer.

  A storage may be installed in a partition having a number obtained by adding an integer of 0 or more and P (1) or less to P (I) calculated in step S652. Further, an integer from 0 to P (1) may be determined by a random number. When a large number of server systems 10 having the same configuration are used side by side, the section in which the storage is installed and the section in which the managed server 14 to be stopped in response to a power saving request is distributed. Therefore, since the distribution of the calorific value is made uniform, a local temperature rise can be prevented.

[Embodiment 7]
The present embodiment relates to a program of a system including a plurality of electronic devices that can be replaced with each other and electronic devices that cannot be replaced with each other. In the present embodiment, the heights of electronic devices that can be replaced and electronic devices that cannot be replaced are different. An electronic device that can replace each other is, for example, a managed server 14 on which a virtual machine operates. An electronic device that cannot be interchanged is, for example, a storage 28 that stores a large amount of data. The system is a server system 10 including, for example, a plurality of managed servers 14 on which virtual machines operate, a management server 13, a communication device 12, and a storage 28. Description of portions common to the first embodiment is omitted.

  FIG. 27 is an example of a table indicating the location of the managed server 14 to be stopped according to the seventh embodiment. The table in FIG. 27 is used instead of the flowchart of the fifth embodiment shown in FIG. 24, for example. With reference to FIG. 27, the contents of the process of specifying the section where the managed server 14 to be stopped according to the present embodiment will be described.

  In this embodiment, a case where a 22-unit 19-inch rack, that is, a 19-inch rack having 22 sections in the height direction is used as the shelf 11 will be described as an example. One unit of communication device 12 and one unit of management server 13 are installed, and one unit of managed server 14 and two units of storage 28 are installed in the remaining 20 units. 14 managed servers 14 and 3 storages are installed.

  The table shown in FIG. 27 shows the number of managed servers 14 to be stopped, the predicted value of the CPU usage rate of the managed servers 14 after stopping some managed servers 14, and the shelf 11 of the managed servers 14 to be stopped. It is a table which shows the number of the inside division. The leftmost column indicates the number of managed servers 14 to be stopped calculated by the subroutine for calculating the number of stopped units. The second column from the left indicates a predicted value of the CPU usage rate of the managed server 14 that is responding to the power saving request. The second row from the top of the third and subsequent columns from the left indicates the partition numbers in the shelf 11. The partition numbers are indicated by numbers on the left side of FIG.

  The locations described as ST in the third and lower rows from the top of the third and subsequent columns from the left indicate the sections where the storage 28 is installed. The location where the storage 28 is installed may be determined by the method described in the sixth embodiment, or may be determined by other methods. In the present embodiment, the storage 28 is installed in sections 3 and 4, 11 and 12, and 17 and 18.

  The managed server 14 is installed in a section where the storage 28 is not installed. The location indicated by X in the third and lower rows from the top of the third column on the left indicates that the managed server 14 installed in the corresponding section is to be stopped when responding to the request for power saving. .

  More specifically, when three managed servers 14 are stopped and the predicted value of the CPU usage rate is set to 60%, three servers installed in the sections corresponding to the numbers 7, 13, and 20 are used. The managed server 14 is stopped. That is, in this embodiment, since one unit of the managed server 14 is used, the managed servers 14 respectively installed in the No. 7, No. 13 and No. 20 sections are stopped.

  The table shown in FIG. 27 is created by determining the location of the managed server 14 to be stopped that can keep the maximum exhaust temperature low by using a method such as actual measurement or thermal simulation. In particular, by creating FIG. 10 by actual measurement after installing the server system 10 in the data center, individual differences of the installed managed servers 14, the air conditioning state of the installation location of the server system 10, and the connection between each electronic device It is possible to determine the location of the managed server to be stopped while keeping the maximum temperature of the exhaust gas low by combining various situations such as the wiring state.

  A flow of processing of a specific subroutine for the managed server 14 to be stopped will be described. In this subroutine, the section in which the managed server 14 to be stopped is located is specified based on the number of managed servers 14 to be stopped and the predicted value of the CPU usage rate of the managed server 14 and the table shown in FIG.

  According to the present embodiment, when a request for power saving is received, it is determined whether or not the power consumption of the electronic device can be reduced. If possible, the server system 10 that performs power saving according to the request is provided. Can be provided.

  The storage 28 such as 1U and 2U and the managed server 14 may be mixedly mounted. The table shown in FIG. 27 is created in advance according to the mixed situation. In this way, even when the server system 10 combined with any electronic device is used, the server system that examines whether or not it can respond to a request for power saving, and executes power saving according to the request if possible. 10 can be provided.

[Embodiment 8]
The eighth embodiment relates to a mode in which the server system 10 and the program 29 are operated in combination. Also in this embodiment, the server system 10 is a system in which a plurality of electronic devices that can be substituted for each other are installed on the shelf 11. The program determines whether the power consumption of the server system 10 can be reduced when a request for power saving is received, and executes power saving according to the request if possible. An electronic device that can replace each other is, for example, a managed server 14 on which a virtual machine operates. Below, it demonstrates using these specific examples. Description of portions common to the first embodiment is omitted.

  FIG. 28 is a device configuration diagram illustrating a hardware configuration according to the eighth embodiment. The configuration of the present embodiment will be described using FIG. Note that description of portions common to the first embodiment is omitted.

  The server system 10 includes a shelf 11, a plurality of electronic devices installed on the shelf 11, a heat exchanger and a blower (not shown). The shelf 11 is, for example, a 19-inch rack capable of fixing and installing a plurality of electronic devices arranged in the height direction.

  The electronic devices installed on the shelf 11 are, for example, the communication device 12, the management server 13, and a plurality of managed servers 14. The communication device 12 performs communication between the management server 13 and the managed server 14 and communication between the server system 10 and the external network 31.

  The management server 13 manages the server system 10. The management server 13 includes a CPU 21, a main storage device 22, an auxiliary storage device 23, a communication interface 24, a clock 25, a reading unit 26, and a bus.

  The reading unit 26 is a device that reads the portable recording medium 27, and specifically, for example, a micro SD card slot. Since the CPU 21, the main storage device 22, the auxiliary storage device 23, the communication interface 24, and the clock 25 are the same as those in the first embodiment, description thereof is omitted.

  The program 29 is recorded on the portable recording medium 27. The CPU 21 reads the program 29 via the reading unit 26 and stores it in the auxiliary storage device 23. The CPU 21 may read a program 29 stored in the semiconductor memory 30 such as a flash memory mounted in the management server 13. Further, the CPU 21 may download the program 29 from another server computer (not shown) connected via the communication interface 24 and save it in the auxiliary storage device 23.

  The CPU 21 reads the program 29 for executing the above-described various software processes from the portable recording medium 27 or the semiconductor memory 30 or downloads it from another server computer (not shown) via the communication interface 24. The program 29 is installed as a control program for the management server 13, loaded into the main storage device 22 and executed. Thereby, the server system 10 functions as the management apparatus described above as a whole.

The technical features (components) described in each embodiment can be combined with each other, and new technical features can be formed by combining them.
The embodiments disclosed herein are illustrative in all respects and should not be considered as restrictive. The scope of the present invention is defined not by the above-described meaning but by the scope of the claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of the claims.

  Regarding the above embodiment, the following additional notes are disclosed.

(Appendix 1)
Obtain a limit on the amount of power consumed by a plurality of electronic devices and a time zone for performing the limit,
Calculate the number of electronic equipment to be stopped based on the limit,
Based on the calculated number, specify the installation location of the electronic equipment to be stopped,
The program which makes a computer perform the process which stops the electronic device of the said installation place specified before the said time slot | zone.

(Appendix 2)
The electronic device includes a plurality of electronic devices that can be substituted for each other during operation,
The program according to claim 1, wherein the specifying is performed by selecting an installation location of an electronic device to be stopped from an installation location of the electronic device that can be replaced.

(Appendix 3)
The program according to appendix 2, wherein the replaceable electronic device is an electronic computer that operates a plurality of virtual machines.

(Appendix 4)
Prior to stopping the electronic computer based on the identification, a virtual machine operating on the electronic computer to be stopped is arranged in an electronic computer that does not stop,
Start the computer that was stopped after the time period ended,
The program according to appendix 3, which causes a computer to execute a process of rearranging an operating virtual machine to an operating electronic computer.

(Appendix 5)
The program according to any one of appendix 1 to appendix 4, wherein the specifying is performed with reference to a table in which the calculated number, the load of the electronic device, and the installation location of the electronic device to be stopped are associated with each other .

ｋは正の整数。
ｙ（ｋ）≦ｊ
ｙ（ｋ）は停止する電子機器が設置されている単位区画を示す番号
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｎは停止する電子機器の数
ｊは電子機器を設置する単位区画の総数。
ｈｓｖは、１台の電子機器が設置される区画が備える単位区画の数。 (Appendix 6)
The electronic device is installed in a place where it can be divided into unit sections having a reference length defined in advance in one direction,
The program according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the specifying is performed based on the calculated number and formula.

k is a positive integer.
y (k) ≦ j
y (k) is a number indicating the unit section in which the electronic device to be stopped is installed. round (x) is a function that rounds x to obtain an integer.
n is the number of electronic devices to stop j is the total number of unit sections in which electronic devices are installed.
hsv is the number of unit sections provided in a section in which one electronic device is installed.

ｋは正の整数。
ｙ（ｋ）≦ｊ
ｙ（ｋ）は停止する電子機器が設置されている単位区画を示す番号
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｎは停止する電子機器の数
ｊは電子機器を設置する単位区画の総数。
ｈｓｖは、１台の電子機器が設置される区画が備える単位区画の数。
Ｒは０以上ｙ（ｋ）の最小値以下の整数。 (Appendix 7)
The electronic device is installed in a place where it can be divided into unit sections having a reference length defined in advance in one direction,
The program according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the specifying is performed based on the calculated number and formula.

k is a positive integer.
y (k) ≦ j
y (k) is a number indicating the unit section in which the electronic device to be stopped is installed. round (x) is a function that rounds x to obtain an integer.
n is the number of electronic devices to stop j is the total number of unit sections in which electronic devices are installed.
hsv is the number of unit sections provided in a section in which one electronic device is installed.
R is an integer not less than 0 and not more than the minimum value of y (k).

(Appendix 8)
In the calculation, the predicted value of the amount of power consumed by the computer when a part of the computer is stopped and a virtual machine operating on the computer is transferred to another computer is based on the limit. Determine whether it is less than the amount,
Any one of appendix 3 to appendix 8 characterized in that when the predicted value is determined to be smaller than the amount of power based on the limit, the number of the computers is stopped. The listed program.

(Appendix 9)
The program according to any one of Supplementary Note 1 to Supplementary Note 8, wherein the electronic device is installed in a section arranged in a vertical direction.

Ｉは正の整数。
Ｐ（Ｉ）は記憶装置を配置する単位区画を示す番号。
Ｐ（Ｉ）≦ｊ
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｍは記憶装置の数
ｊは電子計算機および記憶装置を配置する単位区画の総数。
ｈｓｔは、１台の電子計算機または記憶装置を配置する区画を構成する単位区画の数。 (Appendix 10)
A section composed of the same number of unit sections can be divided into unit sections having a reference length defined in advance in one direction, and a serial number starting from 1 is assigned to the unit section. When placing a computer and storage device that can be placed in
P (I) that satisfies the equation based on the number of the storage devices to be arranged, the total number of the unit partitions in which the electronic computer and the storage device are arranged, and the number of unit partitions included in the partition in which the computer or the storage device is arranged )
A program for causing a computer to execute a process of determining a partition starting from a unit partition to which P (I) is assigned as a partition in which a storage device is arranged.

I is a positive integer.
P (I) is a number indicating a unit partition in which the storage device is arranged.
P (I) ≦ j
round (x) is a function that rounds x to obtain an integer.
m is the number of storage devices j is the total number of unit partitions in which the electronic computer and the storage device are arranged.
hst is the number of unit sections constituting a section in which one electronic computer or storage device is arranged.

(Appendix 11)
Obtain a limit on the amount of power consumed by a plurality of electronic devices and a time zone for performing the limit,
Calculate the number of electronic equipment to be stopped based on the limit,
Based on the calculated number, specify the installation location of the electronic equipment to be stopped,
An operation method for causing a computer to execute a process of stopping the electronic device at the installation location specified before the time period.

Ｉは正の整数。
Ｐ（Ｉ）は記憶装置を配置する単位区画を示す番号。
Ｐ（Ｉ）≦ｊ
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｍは記憶装置の数
ｊは電子計算機および記憶装置を配置する単位区画の総数。
ｈｓｔは、１台の電子計算機または記憶装置を配置する区画を構成する単位区画の数。 (Appendix 12)
A section composed of the same number of unit sections can be divided into unit sections having a reference length defined in advance in one direction, and a serial number starting from 1 is assigned to the unit section. When placing a computer and storage device that can be placed in
P (I) satisfying the formula based on the number of the storage devices to be arranged, the total number of the unit partitions in which the electronic computer and the storage device are arranged, and the number of unit partitions included in the partition in which the electronic computer or the storage device is arranged )
An arrangement determination method for causing a computer to execute a process of determining a partition starting from a unit partition to which P (I) is assigned as a partition in which a storage device is disposed.

I is a positive integer.
P (I) is a number indicating a unit partition in which the storage device is arranged.
P (I) ≦ j
round (x) is a function that rounds x to obtain an integer.
m is the number of storage devices j is the total number of unit partitions in which the electronic computer and the storage device are arranged.
hst is the number of unit sections constituting a section in which one electronic computer or storage device is arranged.

(Appendix 13)
An acquisition unit that acquires a restriction on the amount of power consumed by a plurality of electronic devices and a time zone for performing the restriction;
A calculation unit for calculating the number of electronic devices to be stopped based on the restriction;
A specifying unit for specifying an installation location of the electronic device to be stopped based on the calculated number;
A driving device comprising: a stop unit that stops the electronic device at the installation location specified before the time period.

Ｉは正の整数。
Ｐ（Ｉ）は記憶装置を配置する単位区画を示す番号。
Ｐ（Ｉ）≦ｊ
ｒｏｕｎｄ（ｘ）は、ｘを四捨五入して整数を得る関数。
ｍは記憶装置の数
ｊは電子計算機および記憶装置を配置する単位区画の総数。
ｈｓｔは、１台の電子計算機または記憶装置を配置する区画を構成する単位区画の数。 (Appendix 14)
A section composed of the same number of unit sections can be divided into unit sections having a reference length defined in advance in one direction, and a serial number starting from 1 is assigned to the unit section. When placing a computer and storage device that can be placed in
P (I) satisfying the formula based on the number of the storage devices to be arranged, the total number of the unit partitions in which the electronic computer and the storage device are arranged, and the number of unit partitions included in the partition in which the electronic computer or the storage device is arranged )
An arrangement determining apparatus, comprising: a determining unit that determines a section starting from a unit section to which P (I) is assigned as a section in which the storage device is disposed.

I is a positive integer.
P (I) is a number indicating a unit partition in which the storage device is arranged.
P (I) ≦ j
round (x) is a function that rounds x to obtain an integer.
m is the number of storage devices j is the total number of unit partitions in which the electronic computer and the storage device are arranged.
hst is the number of unit sections constituting a section in which one electronic computer or storage device is arranged.

DESCRIPTION OF SYMBOLS 10 Server system 11 Shelf 12 Communication equipment 13 Management server 14 Managed server 18 Blower 19 Heat exchanger 21 CPU
22 Main storage device 23 Auxiliary storage device 24 Communication interface 26 Reading unit 28 Storage 29 Program 31 Network 33 Refrigerator 34 Buffer tank 35 Refrigerant pipe

Claims (9)

Obtain a restriction on the amount of power consumed by a plurality of computers and a time zone for performing the restriction,
Calculate the number of computers to stop based on the limit and the relationship between the CPU usage rate and power consumption of the computer ,
Based on the calculated number, identify the installation location of the computer to be stopped,
A program for causing a computer to execute a process of stopping an electronic computer at the installation location specified before the time period. The electronic computer includes a plurality of electronic computers that can replace each other during operation,
The program according to claim 1, wherein the specifying is performed by selecting an installation location of the electronic computer to be stopped from an installation location of the electronic computer that can be replaced. 3. The program according to claim 2, wherein the replaceable electronic computer is an electronic computer that operates a plurality of virtual machines. Prior to stopping the electronic computer based on the identification, a virtual machine operating on the electronic computer to be stopped is arranged in an electronic computer that does not stop,
Start the computer that was stopped after the time period ended,
The program according to claim 3, which causes a computer to execute a process of rearranging an operating virtual machine to an operating electronic computer. The identification is performed with reference to a table in which the calculated number, the load on the electronic computer, and the installation location of the electronic computer to be stopped are associated with each other. Program. The electronic computer is installed in a place where it can be divided into unit sections having a reference length defined in advance in one direction,
The program according to any one of claims 1 to 4, wherein the specifying is performed based on the calculated number and the formula (1).

k is a positive integer.
y (k) ≦ j
y (k) is a number indicating the unit section in which the computer to be stopped is installed. round (x) is a function that rounds x to obtain an integer.
n is the number of electronic computers to be stopped. j is the total number of unit sections in which the electronic computers are installed.
hsv is the number of unit sections provided in a section where one computer is installed. A section composed of the same number of unit sections can be divided into unit sections having a reference length defined in advance in one direction, and a serial number starting from 1 is assigned to the unit section. When placing a computer and storage device that can be placed in
Formula (5) is satisfied based on the number of the storage devices to be arranged, the total number of the unit partitions in which the electronic computer and the storage device are arranged, and the number of unit partitions included in the partition in which the electronic computer or the storage device is arranged. P (I) is calculated,
A program for causing a computer to execute a process of determining a partition starting from a unit partition to which P (I) is assigned as a partition in which a storage device is arranged.

I is a positive integer.
P (I) is a number indicating a unit partition in which the storage device is arranged.
P (I) ≦ j
round (x) is a function that rounds x to obtain an integer.
m is the number of storage devices j is the total number of unit partitions in which the electronic computer and the storage device are arranged.
hst is the number of unit sections constituting a section in which one electronic computer or storage device is arranged. Obtain a restriction on the amount of power consumed by a plurality of computers and a time zone for performing the restriction,
Calculate the number of computers to stop based on the limit and the relationship between the CPU usage rate and power consumption of the computer ,
Based on the calculated number, identify the installation location of the computer to be stopped,
An operation method for causing a computer to execute a process of stopping the electronic computer at the installation location specified before the time period. An acquisition unit that acquires a restriction on the amount of power consumed by a plurality of electronic computers and a time zone for performing the restriction;
A calculation unit that calculates the number of electronic computers to be stopped based on the relationship between the restriction and the CPU usage rate and power consumption of the electronic computer ;
A specifying unit for specifying the installation location of the electronic computer to be stopped based on the calculated number;
An operating device comprising: a stop unit that stops the electronic computer at the installation location specified before the time period.

Images