WO2014147708A1

WO2014147708A1 - Power source control device, power source control program, and power source control method

Info

Publication number: WO2014147708A1
Application number: PCT/JP2013/057671
Authority: WO
Inventors: 杉山淳一; 松浦康道
Original assignee: 富士通株式会社
Priority date: 2013-03-18
Filing date: 2013-03-18
Publication date: 2014-09-25

Abstract

The present invention adjusts the electrical power that a server autonomously supplies even if the number of devices being served changes. This power source control device has a calculation unit, a determination unit, a power control unit, and a device power source control unit. The power source control device controls the number of power sources that supply power to the server. The calculation unit calculates the total power consumption of the server by adding the actual power consumption of the running devices connected to the server and the rated power consumption of pre-running devices. The determination unit compares the total power consumption and the amount of power that the power sources can supply, and determines the minimum number of power sources that can supply the at least the total power consumption. The power control unit causes power to be supplied by the power sources in accordance with the determined number of power sources. The device power source control unit causes the pre-running devices to run when the power supplied to the server has reached the total power consumption.

Description

POWER CONTROL DEVICE, POWER CONTROL PROGRAM, AND POWER CONTROL METHOD

The present invention relates to power supply control.

Blade server is a server system that can be equipped with multiple power supplies and multiple devices. In the blade server, a plurality of boards (blades) on which the server computer function is mounted can be attached to the insertion port of the housing. In the blade server, a server having desired performance can be obtained by adjusting the number of blades to be installed according to the application. The device includes blades and switches in the blade server. The plurality of power sources is, for example, a combination of a power source that supplies power and an auxiliary power source. When the number of devices installed in the blade server is increased, the total power consumption of the blade server increases. When the total power consumption of the blade server exceeds the amount of power supplied by the power supply, the server operator or user increases the number of power supplies that supply power corresponding to the total power consumption. Further, when the number of devices mounted on the blade server is reduced, the total power consumption of the blade server is reduced. If the number of power supplies supplying power can be reduced, the server operator or user may use redundant power supplies to make the power supplies redundant. The server operator or user calculates the power consumption amount of the server and the power supply amount of the power source, and sets the combination of power sources according to the operation status of the server. It is preferable that the server operator or user decides the combination of power sources on the server side without being aware of the server operation status, the power consumption amount of the server, and the power supply amount of the power source.

In a server in which the number of blades mounted on the server does not vary, a technology is known in which the server selects a combination of power sources based on conversion efficiency. The server monitors the power consumption of the actual load, and selects a highly efficient combination of power sources from information on conversion efficiency set in advance corresponding to the power consumption information (see, for example, Patent Document 1).

Also, overvoltage protection function is known. When the power supply is in redundant operation, the operation of the apparatus is continued by stopping the power supply that has become abnormal and supplying power only with a normal power supply (see, for example, Patent Document 2).

A technique for controlling the operation of a device to be monitored according to the operation status of a specific network is known (see, for example, Patent Document 3).

Patent Publication No. 2008-204231 Patent Publication No. 2003-339116 Patent Publication No. 2007-159298

In the background art described above, power supply control in the blade server is not considered. In the case of a blade server, since the mounted device fluctuates during operation, there are the following problems.

For example, a new device may be additionally installed in the server by hot plugging, and the total power consumption of the server may exceed the amount of power supplied by the power supply. In this case, the server may go down. It is also possible that the added device cannot be activated.

The power supply control device includes a calculation unit, a determination unit, a power control unit, and a device power supply control unit. The power control device controls the number of power supplies that supply power to the server. The calculation unit calculates the total power consumption of the server by adding the actual power consumption of the active device connected to the server and the rated power consumption of the device before the operation. The determination unit compares the total power consumption with the amount of power that can be supplied by the power supply, and determines the minimum number of power supplies that can supply power that is equal to or greater than the total power consumption. The power control unit causes the power supply to supply power corresponding to the determined number of power supplies. The device power control unit operates the device before the operation when the power supplied to the server reaches the total power consumption.

∙ Adjust the power that the server supplies autonomously even if the number of devices in the server fluctuates.

It is an example of the server which concerns on embodiment. It is an example of the server which concerns on embodiment. It is an example of the hardware constitutions of a server. It is an example of a functional block diagram concerning an embodiment. It is an example of the data of power consumption. It is a flowchart explaining the example of a process of the device control part 106 when a device is added. 10 is a flowchart for explaining an example of processing of an optimization unit 104. 5 is a flowchart illustrating an example of processing of a power supply device control unit 105. It is an example of the information used for determination of the combination of power supplies. It is an example of the figure which compared the threshold value calculated from the electric energy which can be supplied of a power supply, and rated and actual power consumption. It is a flowchart explaining the example of the process of the server 100 at the time of server starting. It is a flowchart explaining the example of a process of the server 100 during server operation.

Hereinafter, the present embodiment will be described in detail with reference to the drawings.
FIG. 1 is an example of a server according to the embodiment. The server 100 in FIG. 1A includes a management device 101, devices 102-1 to 102-J, and power supplies 103-a to 103-d. The management device 101 includes an optimization unit 104, a power supply control unit 105, and a device control unit 106. In FIG. 1B, the same components as those in FIG. 1A are denoted by the same reference numerals. FIG. 1B is a diagram when a new device is added to the server 100 of FIG. 1A by hot plug. FIG. 1B further includes device 102- (J + 1).

When a new device by hot plug is additionally mounted while the server is operating, the device control unit 106 adds the rated power consumption of the added device to the power consumption P _MAXA before adding the device.
P _MAX = P _MAXA + P _{J + 1} (1) formula

The power consumption amount P _MAXA on the right side is the actual power consumption amount (hereinafter referred to as “actual power consumption”) of FIG. 1A that is the server before the device is added. The server 100 in FIG. 1A has J devices. The rated power consumption of each device is indicated by (P ₁ , P ₂ ,... P _J ). The server 100 in FIG. 1B is a server in which a J + 1th device is added to the server 100 in FIG. 1A. P _{J + 1} is the rated power consumption of the added device. The power consumption P _MAX calculated by the equation (1) is obtained by adding the rated power consumption of the added device before operation to the actual power consumption of the server 100. The power consumption P _MAX is the total power consumption of the server that is expected to be consumed when the added device is operated.

The device control unit 106 sends information on the power consumption P _MAX set by the equation (1) to the optimization unit 104. In parallel, the power supply device control unit 105 sends information on the amount of power that can be supplied by the current power supply (hereinafter referred to as “suppliable power amount”) to the optimization unit 104.

The power source 103-a to power source 103-d in FIG. 1A are a set of a power source 103-a and a power source 103-b surrounded by B1, and a set of a power source 103-c and a power source 103-d surrounded by C1. It is divided into and. The power supply in the range surrounded by B1 is a power supply that supplies power to the server. The power supply in the range surrounded by C1 is an auxiliary power supply used as a backup when the power supply in the range surrounded by B1 fails. A combination of a power source that supplies power to the server and an auxiliary power source is hereinafter referred to as a “power source combination”.

On the other hand, the server 100 increases the number of power supplies that supply power to the server when the (J + 1) th device is added and the power supplies 103-a and 103-b cannot supply power. The power supply in the range surrounded by B2 is a power supply that supplies power to the server. The power supply in the range surrounded by C2 is an auxiliary power supply used as a backup when the power supply in the range surrounded by B2 fails. In the example of FIG. 1B, a combination of a power source 103-a to power source 103-c surrounded by B2 and a power source of a power source 103-d surrounded by C2. In the server 100 of FIG. 1B, a power source 103-c is newly allocated as a power source for supplying power to the server. When the power consumption amount P _MAX is larger than the _suppliable power amount, the optimization unit 104 can secure the power consumption amount P _MAX and change the power source combination to the minimum number of power sources that supply power to the server. The power supply device control unit 105 is notified.

The optimization unit 104 receives information on the power consumption amount P _MAX and information on the suppliable power amount. The server 100 in FIG. 1A does not need to increase the number of power supplies that supply power to the server if the power supply 103-a and the power supply 103-b can supply power even when the J + 1th device is added. When the power consumption amount P _MAX is smaller than the _suppliable power amount, the optimization unit 104 notifies the power supply device control unit 105 that the combination of power sources is not changed.

The power supply control unit 105 sets the number of power supplies that supply power to the server in accordance with the notified information. The power supply control unit 105 transmits information on the suppliable power amount corresponding to the number of power supplies that supply power to the newly set server to the device control unit 106.

If the suppliable power amount corresponding to the number of power sources that supply power to the newly set server can secure the power consumption amount P _MAX , the device control unit 106 powers on the added device.

Fig. 2 shows an example of the hardware configuration of the server. FIG. 2 includes management devices 101-1 to 101-2, devices 102-1 to 102-J, and power supplies 103-1 to 103-I. Each of the devices 102-1 to 102-J includes a CPU, a memory, and a storage device. The devices 102-1 to 102-J are added to make the server have a desired performance, and the number is not limited. The power supplies 103-1 to 103-I are devices that supply power to the server. The number of power supplies installed in the server is not limited. Each of the management devices 101-1 to 101-2 includes a CPU (Central Processing Unit), a memory, and a storage device. The management devices 101-1 to 101-2 are redundant and are divided into an active system and a standby system. The management devices 101-1 to 101-2 implement the optimization unit 104, the power supply device control unit 105, and the device control unit 106 in FIG. 1 is processed by the CPU of the active management device, and information to be used is stored in a memory or a storage device. The server may include a dedicated monitor and input means for setting server setting information. Alternatively, the setting information may be set by accessing the server from an external operator terminal.

FIG. 3 is an example of a functional block diagram according to the embodiment. 1 and 2 are denoted by the same reference numerals. Each of the devices 102-1 to 102-J has information (P ₁ , P ₂ ,... P _J ) on the rated power consumption. FIG. 3 includes a memory 201 and an operator terminal 202.

The device control unit 106 includes a power consumption recording unit 203, a calculation unit 204, and a device power supply control unit 205. The power consumption recording unit 203 records the actual power consumption actually used by the power supply and the device in the memory 201 as power consumption data. The calculation unit 204 calculates the power consumption amount P _MAX . When a device is added as shown in FIG. 1B, the calculation unit 204 calculates the power consumption P _MAX by adding the rated power consumption of the added device to the actual power consumption. Similarly, even when there is a device that is not operating, the calculation unit 204 calculates the power consumption P _MAX by adding the rated power consumption of the device that is not operating to the actual power consumption. When the server is operating, the calculation unit 204 calculates, from the power consumption data of the memory 201, the maximum value of the power consumption during a predetermined period as the power consumption P _MAX . When the server is activated, the calculation unit 204 sums the rated power consumption of each device and calculates the power consumption P _MAX . The device power control unit 205 compares the amount of power that can be supplied with power and the power consumption _PMAX, and when a device that is not operating can be operated, the device that is not operating is operated.

The optimization unit 104 includes an acquisition unit 206, a determination unit 207, and a determination unit 208. The acquisition unit 206 acquires information about the server power consumption P _MAX from the device control unit 106. The acquisition unit 206 acquires information on the suppliable power amount corresponding to the number of power supplies that supply power of the current combination of power supplies from the power supply device control unit 105. The determination unit 207 compares the information on the power consumption amount P _MAX of the server and the information on the suppliable power amount, and determines the minimum number of power sources that can supply power equal to or higher than the power consumption amount P _MAX of the server. The determination unit 208 sets the number of power sources determined by the determination unit 207 as the number of power sources that supply power to the server of the combination of power sources, and determines the number of auxiliary power sources according to the priority order set by the user. The determination unit 208 notifies the power supply device control unit 105 of combination information of power sources, which is information on the number of power sources that supply power to the server and the number of auxiliary power sources.

The power supply device control unit 105 includes a power control unit 209, a setting unit 210, a notification unit 211, and a calculation unit 212. The power control unit 209 receives power source combination information from the determination unit 208 and supplies power to the power source according to the number of power sources that supply power to the server. The setting unit 210 sets the auxiliary power source determined by the power source combination information as a backup. The notification unit 211 calculates a suppliable power amount corresponding to the number of power supplies that are currently supplying power, and notifies the optimization unit 104 and the device control unit 106 of information on the suppliable power amount. The calculation unit 212 calculates the amount of power consumed by the power source. Information on the power consumption amount of the power supply is used by the power consumption recording unit 203.

The memory 201 stores server setting information, power consumption data, and various data. The server setting information includes the following information.
The value (time) of the period used when the calculation unit 204 calculates the maximum value of power consumption
Priority order of the number of auxiliary power sources used when the determination unit 208 determines the number of auxiliary power sources The determination unit 207 uses when determining the minimum number of power sources that can supply the power consumption amount P _MAX or more of the server Threshold

The information held in the memory 201 may be stored in a storage device and used by the optimization unit 104 and the device control unit 106 as appropriate.

The operator terminal 202 has input means such as a keyboard. An operator can use the operator terminal 202 to access the memory 201 and change server setting information. For example, the operator uses the operator terminal 202 to change the priority order of the number of auxiliary power supplies. The combination of power sources is N power sources and M auxiliary power sources, and is represented by N + M. The N power supplies are power supplies that supply power while the server is operating. The M auxiliary power supplies are auxiliary power supplies that operate as backup power supplies when trouble occurs in any of the N power supplies. The combination of N + M power sources is, for example, a combination of N + N, N + 1, and N + 0. The combination of N + N power supplies can continue to operate the server even if N power supplies fail. The combination of N + 1 power supplies can continuously operate the server even if one of the power supplies fails. The combination of N + N power supplies is a combination with higher redundancy than that of N + 1 power supplies. The N + 0 power supply is non-redundant, and if one power supply fails, the server may go down. The priority order of the number of auxiliary power sources set by the operator is a setting of whether or not to give priority to a combination with high redundancy by setting the number of M in N + M. Note that, in the combination of N + M power sources, power can be saved by stopping unused power sources.

FIG. 4 is an example of power consumption data. The power consumption data in FIG. 4 is stored in the memory 201. The example of FIG. 4, the actual power consumption P _R, is obtained by time saved from April 1 to May 13. In the example of FIG. 4, the actual power consumption P _R of April 11 is illustrated as the maximum value P _RMAX of actual power. Calculation unit 204 calculates the maximum value P _RMAX of actual power as power consumption P _MAX.
P _MAX = P _RMAX (2) formula

Calculation unit 204 notifies the information of the power consumption P _MAX to the device control unit 106. The process of acquiring information about the maximum value P _RMAX of the actual power consumption from the power consumption data may be executed every predetermined period. Further, the user / operator may be able to change the setting during a predetermined period.

FIG. 5 is a flowchart illustrating an example of processing of the device control unit 106 when a device is added. The device control unit 106 detects that the device J + 1 has been added (step S101). The device control unit 106 acquires information on the rated power consumption of the detected device J + 1 (step S102). The device control unit 106 adds the detected rated power consumption P _{J + 1} of the device J + 1 to the actual power consumption of the server before the device is added, and sets the added value as a new power consumption P _MAX (step) S103). The device control unit 106 sends information on the power consumption P _MAX to the optimization unit 104 (step S104). After the process of step S104, the processes of FIGS. 6 and 7 described later are executed, and the combination of the power sources is changed. The device control unit 106 acquires information on the suppliable power amount corresponding to the changed combination of power sources from the power supply device control unit 105 (step S105). The device control unit 106, the power consumption P _MAX of the entire server or smaller is determined than the supply power amount (Step S106). The device control unit 106 powers on the device J + 1 (YES in step S106, step S107). The device control unit 106 suppresses power-on of the device J + 1 (NO in step S106, step S108). When the process of step S107 or step S108 is executed, the device control unit 106 moves the process to a flowchart for operation. The processes in steps S101 to S104 are processes executed by the calculation unit 204. Further, the processing of steps S105 to S108 is processing executed by the device power supply control unit 205.

FIG. 6 is a flowchart for explaining an example of processing of the optimization unit 104. Process of Figure 6, the optimization unit 104, the notification information of the power consumption P _MAX (e.g., step S104 in FIG. 5) is a process after receiving the. The optimization unit 104 inquires of the power supply control unit 105 about information relating to the amount of power that can be supplied (step S201). The optimization unit 104 acquires information on the amount of power that can be supplied from the power supply device control unit 105 (step S202). The optimization unit 104 determines an optimal combination of power sources (step S203). The determination of the combination of power sources in step S203 includes the priority order of the combination of power sources. The optimization unit 104 notifies the power supply device control unit 105 of the determined combination information of the power supplies, and transmits an instruction to set (step S204). The processes in steps S201 to S202 are processes executed by the acquisition unit 206. The process of step S203 is a process executed by the determination unit 207 and the determination unit 208. The process in step S204 is a process executed by the determination unit 208.

FIG. 7 is a flowchart for explaining an example of processing of the power supply device control unit 105. The process in FIG. 7A is a process after the power supply control unit 105 receives an inquiry about information about the amount of power that can be supplied from the optimization unit 104 to the power supply control unit 105 (step S201 in FIG. 6). . The power supply device control unit 105 notifies the optimization unit 104 of information related to the suppliable power amount corresponding to the current combination of power sources (step S301). When the process of step S301 ends, the process proceeds to step S203 of FIG. The process in step S301 is a process executed by the notification unit 211. The process in FIG. 7B is a process after receiving an instruction (step S204 in FIG. 6) for setting the combination information of the power source determined by the power supply device control unit 105 from the optimization unit 104. The power supply device control unit 105 changes the power supply combination setting in accordance with the received power supply combination information (step S401). The power supply device control unit 105 notifies the device control unit 106 of information related to the combination of power sources (step S402). The information regarding the combination of power supplies notified in step S402 corresponds to the information acquired in step S104 of FIG. The process of step S401 is a process executed by the power control unit 209 and the setting unit 210. The process in step S402 is a process executed by the setting unit 210.

When a device is added to the server, the combination of power supplies is automatically changed, so that the server operator or user is aware of the server operating status, the amount of power consumed by the server, and the amount of power that can be supplied. The server can be operated without having to When a server device is added by hot plugging and the combination of power sources reaches the suppliable power amount, for example, in Patent Document 1, the power source of the added device may not be activated. However, the server 100 according to the embodiment can automatically supply the power of the added device by securing the amount of power that can be supplied by automatically changing the combination of the power sources. By automatically changing the combination of power sources, the server can stop unused power sources and save power.

Even if there is a device that is not operating among the devices installed in the server, it is preferable that the server autonomously changes the number of power supplies that supply power to the server and starts up the device power. In this case, the processing is the same as when a device is added. For example, if the device X is not running, (1) the P _{J + 1} the rated power consumption of the added device type, may be replaced with the rated power P _X of the device X. The device X is one or a plurality of devices 201-1 to 201-J.

In addition, while the server is operating, the calculation unit 204 calculates the maximum power consumption amount during a predetermined period from the power consumption data of the memory 201 as the power consumption amount P _MAX . When the power consumption amount P _MAX reaches the suppliable power amount, the server autonomously changes the number of power supplies that supply power to the server.

<Processing at server startup>
When the server is started, the server may not have power consumption data. In that case, the calculation unit 204 sums the rated power consumption of each device to calculate the power consumption P _MAX.
P _MAX = P ₁ + P ₂ + ... + P _J (3)

The power consumption amount P _{MAX in the} expression (3) is the sum of the information (P ₁ , P ₂ ... P _J ) of the rated power consumption of each device 201-1 to 201-J. In this way, the calculation unit 204, without the power consumption data recorded actual power consumption, it is possible to calculate the power consumption P _MAX. Then, the server determines the number of power sources that can supply power that is greater than or equal to the power consumption.

FIG. 8 is an example of information used for determining a combination of power sources. FIG. 8 shows an example of a combination of power supplies and a threshold value calculated from the suppliable power amount P _{0 (N + M)} and the suppliable power amount corresponding to the combination of power sources. The combination of power sources in FIG. 8 is an example in the case where six power sources are used. The suppliable power amount P _{0 (N + M)} indicates the suppliable power amount when the combination of the power sources is N + M. The suppliable power amount P _{0 (N + M)} is calculated at 2000 W (watts) for each power source that supplies power. The threshold value calculated from the suppliable power amount is a value used when the determination unit 207 determines the minimum number of power sources that can supply power equal to or greater than the power consumption amount P _{MAX of the} server. As the threshold value, a value obtained by multiplying the suppliable power amount P _{0 (N + M)} by α is used. In this example, α is set to 0.8. α is a value used to determine the threshold. α is 1 or less, and may be set by a user or an operator. The values shown in FIG. 8 are an example and do not limit the numerical values. Similarly, the number of power supplies is not limited. The threshold value may be set directly by an operator or user of the server. The suppliable power amount P _{0 (N + M) in} FIG. 8 is used, for example, in step S106 in FIG. Further, the threshold value calculated from the suppliable power amount in FIG. 8 is used in, for example, step S203 in FIG.
P _{0 (N + M)} × α ≧ P _MAX (4) formula

Expression (4) is a conditional expression used when the determination unit 207 determines the minimum number of power sources that can supply power equal to or greater than the power consumption amount P _{MAX of the} server.

When the power consumption P _MAX of the server is 5000 W, the combination of power sources that can supply the power more than the power consumption P _MAX of the server is the threshold value P _{0 (4 + M)} × α to P _{0 (6 + M)} × α It is. Since the determination unit 207 determines the minimum number of power supplies that can supply power equal to or greater than the power consumption amount P _MAX , the minimum number of power supplies that supply power is four.

FIG. 9 is an example of a diagram comparing the amount of power that can be supplied by the power supply and the rated and actual power consumption. The dotted line in FIG. 9 represents the threshold value P _{0 (N + M)} × α calculated from the suppliable power amount used when the optimization unit 104 in FIG. 3 determines the optimal combination of power sources. The dotted line indicates the threshold value P _{0 (N + M)} × α of P _{0 (5 + 1)} , P _{0 (4 + 1)} , P _{0 (3 + 3)} or P _{0 (3 + 1)} , P _{0 (2 + 2)} or P _{0 (2 + 1)} from the top. To express. The threshold value P _{0 (N + M)} × α in α calculated from the suppliable power amount may be 1. Therefore, the threshold value P _{0 (N + M)} × α calculated from the suppliable power amount may be the suppliable power amount P _{0 (N + M)} . Hereinafter, the threshold used when the determination unit 207 determines the minimum number of power sources that can supply power equal to or higher than the power consumption amount P _{MAX of the} server assumes that α is 1, and the suppliable power amount P _{0 (N + M )} . When the server is activated, the calculation unit 204 calculates the power consumption amount P _MAX from the rated power consumption, and the determination unit 207 determines the number of power supplies that can supply power that is equal to or higher than the power consumption amount. Thereafter, when the server enters an operating state, the calculation unit 204 calculates the maximum value P _RMAX of the actual power consumption as the power consumption P _MAX as shown in the equation (2), and the determination unit 207 Determine the number of power supplies that can supply the power. The flow from server activation to server operating state will be described using the example of FIG. When the server is started, the calculation unit 204 calculates the power consumption amount P _MAX from the rated power consumption. The example of the rated power consumption in FIG. 9 is 4500 W, which exceeds the suppliable power amount 4000 W of P _{0 (2 + 2)} or P _{0 (2 + 1)} . On the other hand, the example of the rated power consumption in FIG. 9 does not exceed 6000 W, which is the suppliable power amount of P _{0 (3 + 3)} or P _{0 (3 + 1)} . Therefore, the power supply of the server is set to P _{0 (3 + 3)} or P _{0 (3 + 1)} . When the server enters the operating state, the calculation unit 204 calculates the maximum value P _RMAX of the actual power consumption as the power consumption P _MAX . The example of the actual power consumption in FIG. 9 is 3000 W, and does not exceed the supplyable power amount 4000 W of P _{0 (2 + 2)} or P 0 _{(2 + 1)} . Therefore, the combination of the server power sources is set to P _{0 (2 + 2)} or P 0 _{(2 + 1)} . As in this example, the actual power consumption of the server is often smaller than the rated power consumption, and the number of power supplies that supply power to the server can be reduced during operation rather than when the server is started.

The determination unit 208 sets the number of power sources determined by the determination unit 207 as the number of power sources that supply power to the server of the combination of power sources, and determines the number of auxiliary power sources according to the priority order set by the user. Next, an example of a case where the power consumption P _MAX of the entire server is calculated to 5000 W, a combination of the power of the N + N is set priority as the priority. In the example in which the power consumption P _MAX is 5000 W, the supplyable power amount 4000 W of P _{0 (2 + 2)} or P _{0 (2 + 1)} is exceeded. On the other hand, the example in which the power consumption amount P _MAX is 5000 W does not exceed 6000 W that is the suppliable power amount of P _{0 (3 + 3)} or P _{0 (3 + 1)} . Therefore, the power supply of the server is set to P _{0 (3 + 3)} or P _{0 (3 + 1)} . Here, the power supply combinations are set so that the N + N power supply combinations are prioritized, so the 3 + 3 power supply combinations are selected. Further, in the example where the power consumption P _MAX is 5000 W, when the priority order is set so that the combination of N + 1 power supplies is given priority, the 3 + 1 redundant configuration is selected. The power control unit 209 stops two power supplies that are not used for power saving.

FIG. 10 is a flowchart for explaining an example of processing of the server 100 at server startup. When the server is activated, the device control unit 106 acquires information on the power consumption of each device (step S501). The device control unit 106 calculates the power consumption P _MAX for the entire server (step S502). The server 100 notifies the optimization unit 104 of information about the power consumption P _{MAX for} the entire server, and changes the combination of power sources. (Step S503). The device control unit 106 acquires information related to the suppliable power amount from the power supply device control unit 105 (step S504). The device control unit 106, the power consumption P _MAX of the entire server is less than or judges than the supply power amount of a combination of power (step S505). The device control unit 106 turns on the power of each of the devices 201-1 to 201-J (YES in step S505, step S506). The server enters the operating state and moves the process to FIG. On the other hand, the device control unit 106 performs setting so as to suppress the power-on of the device X (NO in step S505, step S507). The device X is any one of the devices 201-1 to 201-J, and may be a plurality of devices. From power consumption P _MAX of the entire server, obtains a value obtained by subtracting the rated power P _X set devices X to suppress power-on, the power consumption P _MAX of the entire server (step S508) . When the process of step S508 is executed, the process is repeated from step S503. The user interface 301 may determine the device X by setting the priority order of the devices 201-1 to 201-J as server setting information. Although the power-on of the device X is set to be suppressed in step S507, this process is a temporary response. Thereafter, when the server is in an operating state, the server executes a process corresponding to the operating state, determines whether the device X can be started, and starts up if it can be started. The supplyable power amount used in S505 may use a threshold value calculated from the supplyable power amount. The processes in steps S501 to S503 are processes executed by the calculation unit 204. Further, the processing in steps S504 to S508 is processing executed by the device power supply control unit 205.

FIG. 11 is a flowchart illustrating an example of processing of the server 100 during server operation. This flowchart is processing after step S506 in FIG. The device control unit 106 monitors the actual power consumption P _R (step S601).

The device control unit 106 stores the information of the actual power consumption P _R as the data of the power consumption (step S602). The device control unit 106 determines whether there is power consumption data for the set period (step S603). If there is no power consumption data for the set period (NO in step S603), the process is repeated from step S601.

The device control unit 106 acquires the maximum value P _RMAX of actual power consumption from the power consumption data (step S604). The device control unit 106 compares the power consumption amount P _MAX and the actual power consumption maximum value P _RMAX to determine whether or not they are the same value (step S605). When the power consumption amount P _MAX and the actual power consumption maximum value _PRMAX are the same value (YES in step S605), the processing is repeated from step S601. The device control unit 106 sets the actual power consumption maximum value P _RMAX to the power consumption amount P _MAX (NO in step S605, step S606).

The device control unit 106 determines whether there is a device X whose power-on is inhibited (step S607). The server 100 notifies the optimization unit 104 of information about the power consumption P _{MAX for} the entire server, and changes the combination of power sources. (NO in step S607, step S613). When the process of step S613 is executed, the process is repeated from step S601.

The device control unit 106 newly sets the power consumption P _MAX value obtained by adding the rated power P _X to the power consumption P _MAX (YES in step S607, the step S608). The server 100 notifies the optimization unit 104 of information about the power consumption P _{MAX for} the entire server, and changes the combination of power sources. (Step S609).

The device control unit 106 acquires information on the amount of power that can be supplied from the power supply device control unit 105 (step S610).

The device control unit 106, the power consumption P _MAX of the entire server is less than or judges than the supply power amount of a combination of power (step S611). If the power consumption amount P _MAX of the entire server is larger than the suppliable power amount of the combination of power sources (NO in step S611), the processing is repeated from step S601.

The device control unit 106 powers on the device X (YES in step S611, step S612). When the process of step S612 is executed, the process repeats the process from step S607. As the suppliable power amount used in S611, a threshold value calculated from the suppliable power amount may be used. The process of step S613 is a process when there is no component X that is not operating by the determination process of step S607. At the time of server startup, step S503 sets a combination of power sources that is the minimum number of power sources that can supply power that exceeds the rated power consumption.

When the server is in an operating state, step S613 sets a combination of power sources with the minimum number of power sources that can supply power equal to or greater than the maximum value P _RMAX of actual power consumption. Servers often have lower actual power consumption than rated power consumption. Therefore, when the supply of power is excessive, step S613 can reduce the number of power supplies that supply power, thereby saving power.

On the other hand, regularly step S613 sets a combination of the minimum number of power source capable of supplying a maximum value P _RMAX more power actual power. For example, when the number of users who use the server increases and the use of the server increases, the maximum value P _RMAX of actual power consumption also increases.

Then, it may happen that power supply is insufficient. In step S613, by setting a combination of the minimum number of power sources that can supply power that is equal to or greater than the increased actual power consumption maximum value P _RMAX , power can be supplied in accordance with the usage status of the server.

The processing of steps S601 to S603 is processing executed by the power consumption recording unit 203. The processes in steps S604 to S609 and step S613 are processes executed by the calculation unit 204. Further, the processing of steps S610 to S612 is processing executed by the device power supply control unit 205.

As described above, when the combination of power sources is determined based on the rated power consumption at the time of starting the server, and the rated power consumption is larger than the suppliable power, the device X is temporarily suppressed. Thereafter, when the server is in operation, the combination of power sources is determined from the maximum value of the actual power consumption of the server. Servers often have lower actual power consumption than rated power consumption. Therefore, when the suppliable power amount is larger than the value obtained by adding the actual power consumption and the rated power consumption of the device X, the power of the device X can be turned on.

As described above, in the method according to the embodiment, the server operator or the user can operate the server without being aware of the operation status of the server, the power consumption amount of the server, and the power supply amount of power. .

In addition, it is possible to start the device even when there is a device that is not turned on by selecting the optimal combination of the power sources among the power sources installed in the server. When there is a device that is not in operation, the server can automatically change the combination of the power sources to ensure the amount of power that can be supplied and start the power source of the device that is not in operation. By automatically changing the combination of power sources, the server can stop unused power sources and save power.

Note that the embodiment is not limited to the above, and can be variously modified. Some examples are described below.

The server may store various information about the device that was operating last time and information about the power consumption P ^′ _MAX of the server. When starting the server, compare the various information of the device that was operating last time with the various information of the current device, and if the device has not been newly added or deleted, the server will use the power consumption P _MAX . Set the power consumption P ^′ _MAX . The process is executed by the calculation unit 204.
P _MAX = P ^' _MAX (5) formula

The calculation unit 204 can calculate the power consumption amount P _MAX from the actual power consumption amount even when starting the server by setting as in the equation (5), and the determination unit 207 determines the power consumption amount P _The minimum number of power supplies capable of supplying more than _MAX power can be determined. In addition, the number of processes for changing the combination of power supplies can be reduced. Server may be stored and various information and the amount of power consumption P ^_'MAX of the server information of the device when was the last time running, when to turn off the restart or power the server, to periodically save It may be.

The memory 201 may have a setting for enabling or disabling power consumption data. When the power consumption data is validated, the server executes the processes in FIGS. 5, 6, 7, 10, and 11. On the other hand, when the power consumption data is invalidated, the processes of FIGS. 6, 7 and 10 are executed.

The event stored in the memory 201 may be notified to the user or the operator by an SNMP (Simple Network Management Protocol) trap, mail, or the like.
The memory 201 may have a setting for determining a notification method for an event that has occurred.
The operator can access the memory 201 from the operator terminal 202 and confirm the event. The user or operator can confirm that the combination of the power sources has been changed, and can confirm the activation status of the device.

As described in equation (2), the calculation unit 204 calculates the maximum power consumption amount during a predetermined period from the power consumption data of the memory 201 as the power consumption amount P _MAX . The maximum value of power consumption may be a relatively large value in the power consumption data. Further, the power consumption P _MAX may be a value obtained by adding the margin value to the maximum value, with further margin from the maximum value.

100 server 101 management device 101-1 to 101-2 management device 102-1 to 102-J device 103-a to 103-d, 103-1 to 103-I power supply 104 optimization unit 105 power supply control unit 106 device control Unit 201 memory 202 operator terminal 203 power consumption recording unit 204 calculation unit 205 device power supply control unit 206 acquisition unit 207 determination unit 208 determination unit 209 power control unit 210 setting unit 211 notification unit 212 calculation unit

Claims

A power control device that controls the number of power supplies that supply power to a server,
A calculation unit for calculating the total power consumption of the server by adding the actual power consumption of the active device connected to the server and the rated power consumption of the device before the operation;
A determination unit that compares the total power consumption and the amount of power that can be supplied by a power source, and determines the minimum number of power sources that can supply power equal to or greater than the total power consumption;
A power control unit for supplying power to the power supply corresponding to the determined number of power supplies;
And a device power control unit configured to operate the device before the operation when the power supplied to the server reaches the total power consumption.
2. The power source that is not turned on by the power control unit is used as an auxiliary to the power source that is supplying power, or the power source is kept down for power saving. Power supply control device.
3. The power control apparatus according to claim 1, wherein priority is given to a setting having a large number of power supplies used as an auxiliary to a power supply that supplies power.
4. The actual power consumption of the operating device is a relatively large value of the power consumption actually consumed by the server in a predetermined period. The power supply control device described.
2. The calculation unit, when starting up the server, sets a value obtained by adding all rated power consumptions of devices mounted on the server as a total power consumption of the server. 5. The power supply control device according to any one of 4 to 4.
The calculation unit stores power consumption during operation when the server is shut down, and sets the power consumption during operation as the total power consumption of the server when the server is started. The power supply control device according to claim 5.
A power supply control program for controlling the number of power supplies that supply power to a server,
Calculate the total power consumption of the server by adding the actual power consumption of the active device connected to the server and the rated power consumption of the device before the operation,
Comparing the total power consumption and the amount of power that can be supplied by a power source, and determining the minimum number of power sources that can supply power equal to or greater than the total power consumption amount,
Power is supplied to the power supply corresponding to the determined number of the power supplies,
When the power supplied to the server reaches the total power consumption, the server controls the server to perform a process for operating the device before the operation.
The power supply control program according to claim 7, wherein a power supply that is not turned on is used as an auxiliary to the power supply that supplies power, or the power supply is kept turned off for power saving.
9. The power supply control program according to claim 7, wherein priority is given to a setting with a large number of power supplies used as an auxiliary to a power supply that supplies power.
10. The actual power consumption amount of the active device is a relatively large value among the power consumption amounts actually consumed by the server in a predetermined period. The power supply control program described.
11. The total power consumption of the server is a value obtained by adding all the rated power consumptions of devices mounted on the server when the server is started. The power supply control program described in 1.
The power consumption during operation is stored when the server is shut down, and the power consumption during operation is set as the total power consumption of the server when the server is started. Item 12. The power supply control program according to Item 11.
Calculate the total power consumption of the server by adding the actual power consumption of the active device connected to the server and the rated power consumption of the device before operation.
Comparing the total power consumption and the amount of power that can be supplied by a power source, and determining the minimum number of power sources that can supply power equal to or greater than the total power consumption amount,
Power is supplied to the power supply corresponding to the determined number of the power supplies,
A power control method characterized in that when the power supplied to the server reaches the total power consumption, the server performs processing for operating the device before the operation.
14. The power supply control method according to claim 13, wherein a power supply that is not turned on is used as an auxiliary to a power supply that supplies power, or the power supply is kept down for power saving.
The power control method according to claim 13 or 14, wherein priority is given to a setting with a large number of power supplies used as an auxiliary to a power supply that supplies power.
16. The actual power consumption of the operating device is a relatively large value among the power consumption actually consumed by the server in a predetermined period. The power supply control method described.
The total power consumption of the server is a value obtained by adding all the rated power consumptions of devices mounted on the server when the server is started. The power supply control method described in 1.
The power consumption during operation is stored when the server is shut down, and the power consumption during operation is set as the total power consumption of the server when the server is started. Item 18. The power supply control method according to Item 17.