Electricity amount control system and electric energy control method

Abstract

A power amount control system (1) includes: a power amount control device (10) that generates an inter-room load distribution pattern representing a load distribution ratio to rooms (100) in which a plurality of servers (3) and air conditioners (4) in a data center (1000) are arranged, and sets a load distribution ratio at which the sum of intra-room total power consumption amounts in the rooms (100) becomes minimal; and an intra-room power amount control device (20) that sets a load arrangement pattern at which the summed amount of a server power consumption amount and an air-conditioner power consumption amount becomes minimal in loads distributed according to the load distribution ratio.

Description

[Technical field]

[0001] The present invention relates to a power control system for reducing the amount of power consumed in a data center (hereinafter, sometimes referred to as "DC"). and, Power Control Method By law Regarding. [Background technology]

[0002] Air conditioning in data centers (DCs) accounts for a large proportion of the power consumption, and as the number and size of DCs increases, there is a demand to reduce the amount of power consumed by air conditioning. In addition, the amount of data processed in DCs is increasing year by year, making it necessary to improve the power consumption efficiency of the DC as a whole (the amount of power consumed by the entire DC for a certain amount of data processing).

[0003] The technology described in Non-Patent Document 1 has been disclosed as a technology for optimizing the power consumption of an entire DC by taking into consideration the power consumption of air conditioning and the power consumption of a server (IT device). In the datacenter air conditioning-linked IT load placement optimization method described in Non-Patent Document 1, future load trends of IT equipment are predicted by collecting operation information and monitoring information of IT equipment in a datacenter, and power increases of air conditioning facilities corresponding to power increases of IT equipment are calculated. Then, an optimization problem is solved to minimize the objective function, which is the power consumption of the datacenter, so that the load concentration rate of IT equipment is increased over time, that is, the number of operating IT equipment is reduced. In this way, the placement of IT loads (virtual machines) on IT equipment that minimizes the power consumption of the datacenter is calculated. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Jun Okitsu and 4 others, "Optimization of IT Load Placement in tandem with Air Conditioning for Eco-Friendly Data Centers," FIT (Forum on Information Technology) 2010, 9th Forum on Information Science and Technology, RC-009 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the technology described in Non-Patent Document 1, a general-purpose rule-based standard that is not dependent on equipment conditions that differ for each DC is adopted in the air conditioning power model used to calculate the power of the air conditioning equipment. Therefore, it was difficult to perform optimization to reduce the total power consumption of the DC, taking into account individual equipment conditions such as the placement position of the air conditioning equipment, air flow, server placement configuration in the DC, and thermal cooling efficiency. In addition, no consideration was given to how to allocate the total load to each room in which the servers are placed in one DC in order to reduce the power consumption of the entire DC.

[0006] The present invention has been made in view of these points, and an object of the present invention is to appropriately allocate the load to each room installed in a DC and reduce the power consumption of the entire DC. [Means for solving the problem]

[0007] Power control according to the present invention systema power amount control system including: an in-room power amount control device provided for each room in a data center in which a plurality of rooms are installed, the in-room power amount control device being communicatively connected to the servers and air conditioners in the rooms and controlling the power consumption in the rooms; and a power amount control device communicatively connected to the in-room power amount control device for each room and controlling the power consumption of the entire data center, wherein each room in the data center is provided with a plurality of server areas in which a group of servers is arranged to arrange virtual resources as loads, and a plurality of air conditioning areas in which an effect of air conditioning control by the air conditioners is measured, the power amount control device having a distribution pattern generation unit that acquires information on the total load amount processed in the data center, generates a plurality of inter-room load distribution patterns indicating a load distribution pattern in which a load distribution ratio to each room is changed, and transmits the distribution pattern to each of the in-room power amount control devices in charge of the rooms; and a distribution pattern generation unit that generates the distribution pattern of the load in which the total amount of server power consumption and air conditioning power consumption in the room is smallest according to the load of each room calculated from the load distribution ratio. a load allocation ratio determination unit that receives from each of the in-room power amount control devices a total in-room power consumption indicating the power consumption of the entire room in a turn for each of the inter-room load allocation patterns, sums up the total in-room power consumption received from each of the in-room power amount control devices for each of the inter-room load allocation patterns, and determines the inter-room load allocation pattern that minimizes the total power consumption indicating the sum as a load allocation ratio for each room, wherein the in-room power amount control device acquires information on external factors including an intake port temperature in the server area, an outside temperature, and a server power consumption for each of the server areas that is a predicted amount when virtual resources as the loads are allocated to the servers, divides the values ​​of each of the external factors into a predetermined range width, defines a combination of the ranges divided for each of the external factors as a Situation classification, and determines which Situation classification the acquired information on the external factors belongs to; and a situation recognition unit that calculates a load allocated to its own room using the total load and the load allocation ratio, and calculates the virtual resources as the loads calculated,a server power consumption estimation unit for estimating, for each calculated placement pattern, the server power consumption of a group of servers belonging to each of the server areas; and a placement pattern determination unit for acquiring, from the situation recognition unit, a determination result of a situation classification using the server power consumption of each of the server areas, and for each situation classification, referring to control value-power correspondence information that associates the air conditioning power consumption with the air conditioning control value of the air conditioner, acquiring the air conditioning control value of the air conditioner and the air conditioning power consumption in the situation classification for each placement pattern, summing up the server power consumption for each of the server areas in each placement pattern, calculating the total server power consumption and the air conditioning power consumption, which is the sum of the server power consumption and the air conditioning power consumption, determining the placement pattern with the smallest calculated total amount as the placement pattern in which the virtual resource as the load is placed, and transmitting the smallest calculated total amount in the determined placement pattern to the power amount control device as the total power consumption in the room. Effect of the Invention

[0008] According to the present invention, it is possible to appropriately distribute the load to each room installed in a DC, thereby reducing the power consumption of the entire DC. [Brief description of the drawings]

[0009] [Figure 1] 1 is a diagram showing an overall configuration of a power amount control system according to an embodiment of the present invention; [Diagram 2] FIG. 2 is a diagram showing the configuration within one room of a DC according to the present embodiment. [Diagram 3] 1 is a functional block diagram showing a configuration example of a power amount control device according to an embodiment of the present invention; [Figure 4] 4 is a diagram for explaining a room load allocation pattern generated by the power amount control device according to the embodiment. FIG. [Diagram 5] 1 is a functional block diagram showing a configuration example of an in-room power amount control device according to an embodiment of the present invention. [Figure 6]FIG. 2 is a diagram for explaining a situation classification according to the present embodiment. [Figure 7] 11 is a diagram for explaining a process for determining an arrangement pattern that has the smallest total power consumption in a room. FIG. [Figure 8] 4 is a sequence diagram showing the flow of processing executed by the power control system according to the present embodiment. FIG. [Figure 9] 1 is a hardware configuration diagram showing an example of a computer that realizes the functions of the power amount control device and the in-room power amount control device according to the present embodiment. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Next, an embodiment of the present invention (hereinafter, referred to as "the present embodiment") will be described. FIG. 1 is a diagram showing the overall configuration of an electric power amount control system 1 according to this embodiment.

[0011] As shown in FIG. 1, the power consumption control system 1 is configured to include a DC (data center) 1000 having a plurality of rooms 100 each having a plurality of servers 3 and one or more air conditioners 4, an in-room power consumption control device 20 that is communicatively connected to the plurality of servers 3 and one or more air conditioners 4 in the rooms 100 and is provided corresponding to each room 100, and a power consumption control device 10 that is communicatively connected to each in-room power consumption control device 20 and controls the total power consumption of the entire DC 1000. The power amount control device 10 and each room power amount control device 20 may be provided inside the DC 1000 or may be provided at a location separate from the DC 1000.

[0012] FIG. 2 is a diagram showing the configuration within one room 100 of the DC 1000 according to this embodiment. DC1000 has a plurality of rooms 100, and in each room 100, as shown in Fig. 2, a plurality of servers 3 are installed, and one or more air conditioners 4 that control the air conditioning in the room 100 are installed. Note that the power consumption for the air conditioning of the entire room caused by the operation of one or more air conditioners 4 is measured for each room 100 by a power consumption measuring device (not shown).

[0013] Each in-room power consumption control device 20 corresponding to each room 100 may obtain status information of the air conditioners 4 (air conditioners "1," "2," and "3" in FIG. 2) installed in the room 100 and transmit air conditioning control information via an air conditioning management device (not shown), or may be directly connected to communicate with each air conditioner 4 without going through the air conditioning management device. In addition, the room power control device 20 may obtain status information and transmit control information of the server 3 installed in the room 100 via a server management device (not shown), or may be directly connected to the server 3 for communication.

[0014] In each room 100 of the DC 1000 in this embodiment, the servers 3 in the entire room are divided into areas in which multiple servers 3 are arranged, as shown in Fig. 2, and are controlled as "server areas." This server area 30 is an area that accommodates a group of servers on which virtual resources are arranged. Fig. 2 shows an example in which server areas "1" to "6" are provided.

[0015] In the following explanation, it is assumed that the DC1000 will be used to build and operate a virtualization platform. Known open source virtualization platforms include OpenStack (registered trademark), which is software for building cloud environments, and Kubernetes (registered trademark), which is software for operating and managing containerized workloads and services. OpenStack is primarily used to manage and operate physical machines and virtual machines (VMs). Kubernetes is primarily used to manage and operate containers. In this specification, an application (consisting of one or more containers, one or more VMs, etc.) virtualized in a virtualization platform is called a virtual resource. In Kubernetes, the smallest execution unit of an application is a Pod consisting of one or more containers.

[0016] In this embodiment, an "air-conditioning area" is provided in correspondence with the server area 30 of the server group, as shown in Fig. 2. The air-conditioning area 40 is a collective area in which the effect of air-conditioning control is measured, and faces either the intake side or the exhaust side of the server 3. The air blown from the air conditioner 4 is blown out from the air-conditioned area 40 on the intake side (air-conditioned areas "3," "4," "7," and "8" in FIG. 2) via piping installed, for example, under the floor of the room 100. Then, air whose temperature has increased due to the heat of each server 3 is taken in from the intake of the piping installed in the air-conditioned area 40 on the exhaust side (air-conditioned areas "1," "2," "5," and "6" in FIG. 2), and an airflow that returns to the air conditioner 4 is generated.

[0017] A plurality of sensors (temperature sensors, etc.) are installed in each of the air-conditioned areas 40. A temperature sensor is also installed at the air intake of the server 3 in each server area 30. Furthermore, a sensor (temperature sensor, etc.) is also installed outside the DC 1000. Information obtained from these sensors (sensor information) can be acquired by the in-room power amount control device 20 and the power amount control device 10 via a communication line or the like.

[0018] In the power amount control system 1 according to this embodiment, the power amount control device 10 acquires the total load amount to be processed in a given time from an external device (e.g., a system management device, etc.). Then, the power amount control device 10 causes the in-room power amount control device 20 of each room 100 to calculate the total power consumption (the sum of the server power consumption and the air conditioning power consumption, which will be described later as "total power consumption in the room") in each room 100 when the load allocation ratio to each room 100 is changed, and distributes the load between the rooms 100 according to the load allocation ratio that minimizes the total amount for each room 100.

[0019] The in-room power amount control device 20 of each room 100 classifies the server power consumption determined by the intake port temperature, outside temperature, and load placement pattern of each server area 30 in the room 100 into a situation classification, and retains the air-conditioning control value and the air-conditioning power amount that satisfy the temperature reward condition in each situation as a learning history. Then, based on the learning history, the in-room power amount control device 20 calculates the total in-room power consumption when executing the load placement pattern and air-conditioning control value that minimizes the sum of the server power consumption and the air-conditioning power consumption for each intake port temperature and outside temperature of the server area 30 for the load amount allocated to each room 100, and transmits the calculated total in-room power consumption to the power amount control device 10. Then, the room power control device 20 of each room 100 executes arrangement of virtual resources and air conditioning control according to an arrangement pattern that minimizes the total power consumption in the room for the load amount based on the load distribution ratio determined by the power control device 10.

[0020] Next, the power amount control device 10 and the room power amount control device 20 which constitute the power amount control system 1 according to this embodiment will be specifically described.

[0021] <Electricity amount control device> FIG. 3 is a functional block diagram showing an example of the configuration of the power amount control device 10 according to the present embodiment. The power consumption control device 10 distributes the load to each room 100 based on the total load of the DC 1000 in such a manner that the sum of the server power consumption and the air conditioning power consumption for each room 100 (total power consumption in the room) is minimized. The power amount control device 10 is configured by a computer including a control unit 11, an input / output unit 12, and a storage unit 13.

[0022] The input / output unit 12 inputs and outputs information between each in-room power amount control device 20, etc. This input / output unit 12 is composed of a communication interface that transmits and receives information via a communication line, and an input / output interface that inputs and outputs information between an input device such as a keyboard and an output device such as a monitor (not shown).

[0023] The storage unit 13 is configured with a hard disk, a flash memory, a RAM (Random Access Memory), or the like. In this storage unit 13, programs for executing the various functions of the control unit 11 and information necessary for the processing of the control unit 11 are temporarily stored.

[0024] The control unit 11 controls all of the processes executed by the power amount control device 10, and includes an inter-room load distribution pattern generation unit 111 and a load distribution ratio determination unit 112, as shown in FIG.

[0025] The inter-room load allocation pattern generation unit 111 (allocation pattern generation unit) acquires the total load (DC total load) to be processed in the DC 1000. Then, the inter-room load allocation pattern generation unit 111 generates an inter-room load allocation pattern indicating N types of load allocation in which the load allocation ratio to each room 100 is changed for the DC total load. The inter-room load allocation pattern generation unit 111 transmits, for each generated inter-room load allocation pattern, the total DC load amount and the load allocation ratio of each room 100 to the total DC load amount indicated in the inter-room load allocation pattern to each in-room power amount control device 20.

[0026] FIG. 4 is a diagram for explaining the inter-room load allocation patterns by the inter-room load allocation pattern generating unit 111 of the power amount control device 10 according to this embodiment. As shown in FIG. 4, the inter-room load allocation pattern generation unit 111 generates the load allocation ratio to be allocated to each room 100 (room "A", room "B", and room "C" in FIG. 4) as an inter-room load allocation pattern such as "50:50:0", "100:0:0", or "0:50:50".

[0027] This load distribution ratio pattern is preset based on a predetermined distribution ratio logic. For example, a ratio of 100 divided into m equal parts (for example, 10 divided into 10 equal parts) is used as the minimum standard distribution unit, and N types of inter-room load distribution patterns are generated, such as "100:0:0" ... "90:10:0" ... "40:30:30" ... "10:10:80" ...

[0028] The inter-room load allocation pattern generation unit 111 transmits information on the load allocation ratio of each room 100 to the total DC load indicated in the generated inter-room load allocation pattern, and information on the total load (total DC load), to each in-room power amount control device 20. As a result, each in-room power amount control device 20 calculates the total in-room power consumption when controlling the air conditioners 4 with the load arrangement pattern and air conditioning control values ​​that minimize the server power consumption and air conditioning power consumption for the given load (details will be described later).

[0029] Returning to Fig. 3, the load distribution ratio determination unit 112 acquires the total in-room power consumption (total of air conditioning power consumption and server power consumption) in each room 100 from each in-room power energy control device 20. Then, the load distribution ratio determination unit 112 sums up the total in-room power consumption of each room 100 for each inter-room load distribution pattern, and determines the inter-room load distribution pattern that minimizes the sum (total power consumption). Then, the load distribution ratio determination unit 112 transmits the load distribution ratio of each room 100 indicated by the determined inter-room load distribution pattern to each in-room power energy control device 20.

[0030] 4 shows that the load distribution ratio of "0:50:50" of "distribution pattern 3" (symbol a in FIG. 4) is selected, which minimizes the sum (total power consumption) of the total room power consumption (P(A)) of room "A," the total room power consumption (P(B)) of room "B," and the total room power consumption (P(C)) of room "C." The load distribution ratio determination unit 112 transmits the load distribution ratio thus determined to each of the in-room power amount control devices 20.

[0031] <Room power consumption control device> Next, the in-room power amount control device 20 according to this embodiment will be described. FIG. 5 is a functional block diagram showing a configuration example of the in-room power amount control device 20 according to this embodiment. The in-room power energy control device 20 classifies each situation that causes a change in power consumption in the room 100 as a Situation, and calculates the total in-room power consumption when a load arrangement and air-conditioning control value are executed that minimizes the sum of the server power consumption and the air-conditioning power consumption for a given load amount in each situation. Then, each in-room power energy control device 20 executes arrangement of virtual resources and air-conditioning control in a load arrangement pattern that minimizes the total in-room power consumption for the load amount based on the load allocation ratio of each room 100 indicated by the inter-room load allocation pattern determined by the power energy control device 10. The in-room power amount control device 20 is configured by a computer including a control unit 21, an input / output unit 22, and a storage unit .

[0032] The input / output unit 22 inputs and outputs information between the power amount control device 10 and each device (each server 3 and each air conditioner 4) in the DC 1000. The input / output unit 22 is composed of a communication interface that transmits and receives information via a communication line, and an input / output interface that inputs and outputs information between an input device such as a keyboard (not shown) and an output device such as a monitor (not shown).

[0033] The storage unit 23 is configured with a hard disk, a flash memory, a RAM (Random Access Memory), or the like. The storage unit 23 temporarily stores programs for executing the functions of the control unit 21 and information necessary for the processing of the control unit 21. The storage unit 23 also stores a server power amount learning model 231, operation history information 232, and control value-power amount correspondence information 233 (details will be described later).

[0034] The control unit 21 is responsible for all of the processing executed by the room power consumption control device 20, and as shown in FIG. 5, is configured to include a situation recognition unit 211, a placement pattern calculation unit 212, a server power consumption estimation unit 213, a placement pattern determination unit 214, a load placement control unit 215, and an air conditioning control unit 216.

[0035] The situation recognition unit 211 regards the "server area intake temperature," "outside air temperature (outdoor air temperature)," and "server power consumption for each server area" determined by the load placement pattern in the room 100 in the DC 1000 before control as situation components, and acquires information on each of these situation components. These situation components are external factors that affect the increase or decrease in the air conditioning power consumption.

[0036] The situation recognition unit 211 obtains temperature information from multiple temperature sensors installed around the air intakes of the servers 3 in each server area 30, calculates the average value, and calculates the average air intake temperature for each server area 30. The situation recognition unit 211 then averages the calculated average temperatures for each server area 30 for the entire room 100, and sets the obtained temperature as the "server area air intake temperature."

[0037] The situation recognition unit 211 regards information obtained from a temperature sensor installed outside the DC 1000 as the "outside air temperature (outside air temperature)." The "server power consumption amount for each server area" is information calculated by the server power consumption amount estimation unit 213 (details will be described later).

[0038] The situation recognition unit 211 determines to which situation category each piece of acquired information (external factor information) that is an external factor belongs. Each external factor is divided into a plurality of ranges between the minimum and maximum values ​​according to the characteristics of the external factor. A combination of the ranges into which each external factor is divided is defined as one Situation. The following will be explained with reference to FIG. 6.

[0039] As shown in FIG. 6, each external factor is defined as a "factor" and a range to be divided is defined (hereinafter referred to as "division definition"). For example, the external factor of "factor1" shown in the Situation classification information 52 in FIG. 6 is "air intake temperature of server area," and the division definition is "0-48 degrees divided into 6." The external factor of "factor2" is "outside air temperature (air temperature)," and the division definition is "0-48 degrees divided into 6." The external factor of "factor3" is "server power consumption in server area "1"," and the division definition is "0-200W divided into 20." Similarly, the external factor of "factor8" is "server power consumption in server area "6"," and the division definition is "0-200W divided into 20."

[0040] Here, it is assumed that the information on the external factor acquired by the situation recognition unit 211 is the external factor information 51 shown in Fig. 6. In this case, since the value of "factor1" (air intake temperature of the server area) is "25", the situation recognition unit 211 determines that the "range" is included in the "24-32 range" (24 degrees or more and less than 32 degrees) and sets the "factor range identifier" to "factor1-4". This "factor range identifier" is information that identifies the range to which the temperature belongs, for example, by dividing 0-48 degrees into 6, with 0 degrees or more and less than 8 degrees being "factor1-1", 8 degrees or more and less than 16 degrees being "factor1-2", and 16 degrees or more and less than 24 degrees being "factor1-3". The same applies to the other "factors".

[0041] The situation recognition unit 211 combines the information of the "factor range identifier" of the external factors to create a "Situation classification" and determines that it is "factor1-4_factor2-4_factor3-4_factor4-4_factor5-5_factor6-5_factor7-4_factor8-4". In this way, the situation recognition unit 211 determines the "Situation classification" based on the acquired information of the external factors.

[0042] 5, the allocation pattern calculation unit 212 obtains the load amount allocated to the room it is responsible for based on information on the load allocation ratio of each room 100 to the total DC load amount indicated in the inter-room load allocation pattern and information on the total load amount (total DC load amount) obtained from the power amount control device 10. Then, the allocation pattern calculation unit 212 obtains information on the amount of virtual resources to be newly allocated (e.g., the number of CPU cores), and calculates an allocation pattern in which new virtual resources (VMs, containers, etc.) are allocated to each server 3 based on the most recent resource usage status (e.g., CPU usage rate). After allocating the virtual resources to each server 3, the allocation pattern calculation unit 212 makes the resource occupation amount in each server 3 equal to or less than the server capacity (upper limit value) x a predetermined threshold value.

[0043] The server power consumption estimation unit 213 estimates the power consumption of each server 3 by utilizing a learning model (server power consumption learning model 231) for each placement pattern calculated by the placement pattern calculation unit 212. Then, based on the server placement configuration for each server area 30, the server power consumption estimation unit 213 calculates the total server power consumption for each server area 30 in each placement pattern.

[0044] Specifically, the server power consumption estimation unit 213 predicts the power consumption of each server 3 for each deployment pattern using a learning model (server power consumption learning model 231) that uses the air intake temperature of the server area and information on resource usage (e.g., CPU usage, memory usage, etc.) as input data and the server power consumption as output data. The server power consumption learning model 231 is generated in advance as learning data, using the intake port temperature, the resource usage of the server 3, and information on the server power consumption, which is result information at that time.

[0045] Furthermore, the server power consumption estimation unit 213 calculates the server power consumption for each server area 30 by adding up the server power consumption of each server 3 in the server area 30 based on the server layout of each server area 30 .

[0046] The placement pattern determination unit 214 totals the server power consumption for each server area 30 in each placement pattern of the virtual resources calculated by the placement pattern calculation unit 212, calculates the sum (total power consumption in the room) of the total server power consumption, which is the total value, and the air conditioning power consumption obtained from the control value power consumption correspondence information 233 stored in the memory unit 23, and determines the placement pattern with the smallest total amount (total power consumption in the room) as the placement pattern for placing the virtual resources in the Situation classification of that room 100.

[0047] This control value-power correspondence information 233 is information stored for each Situation classification, and stores the optimal control value (air conditioning control value) of each air conditioner 4 for each situation indicated by the Situation classification determined by the situation recognition unit 211, and the amount of air conditioning power consumption when each air conditioner 4 executes that air conditioning control value. Here, the air conditioning control value is a parameter for controlling the air conditioner 4, and includes at least the temperature (target temperature), and may also include air volume, air direction, etc. In this embodiment, the parameters of the air conditioning control value will be described as the target temperature and air volume.

[0048] The information on the optimal air conditioning control values ​​(target temperature, air volume, etc.) for each situation classification and the amount of air conditioning power consumed when they are executed, which is stored in this control value-power-amount correspondence information 233, can be obtained by a method using past performance data (stored in the memory unit 23 as operation history information 232) or a rule-based calculation method, but in this embodiment, an example of calculation by constructing a learning model (air conditioning control learning model) is described below.

[0049] <Generation process of control value-power amount correspondence information> The arrangement pattern determination unit 214 generates a learning model (air-conditioning control learning model) by executing the learning phase and operation phase described below. First, in the learning phase, the arrangement pattern determination unit 214 randomly generates air conditioning control values ​​(target temperature, air volume, etc.) for each Situation classification up to a predetermined number of times (N times). Then, the air conditioning control unit 216 operates each air conditioner 4 according to the randomly generated air conditioning control values. The placement pattern determination unit 214 stores, for each Situation classification, external factor information, air conditioning control value, reward (area reward) calculated when the control is executed (details will be described later), and information on air conditioning power consumption as operation history information 232. Then, when a predetermined number of times (N times) is reached in the learning phase, the placement pattern determination unit 214 refers to the operation history information 232, and for each Situation classification, incorporates external factor information, air-conditioning control values, and rewards (area rewards), generates air-conditioning control learning data, and trains the learning model (air-conditioning control learning model).

[0050] Furthermore, after a predetermined number of times (N times) in the learning phase, the placement pattern determination unit 214 outputs air conditioning control values ​​(target temperature, air volume, etc.) by inputting external factor information into an air conditioning control learning model for each Situation classification, and stores information on the reward (area reward) and air conditioning power consumption when control of the air conditioner 4 is executed with that air conditioning control value as operation history information 232. Then, when a condition based on a predetermined reward described below is satisfied in the air conditioning control learning model for the corresponding Situation classification, the placement pattern determination unit 214 ends the learning phase and transitions to the operation phase.

[0051] Here, the reward is an index for evaluating the result of executing control of the air conditioner 4 according to the calculated air-conditioning control value, and the reward (temperature reward) is calculated as an indication of how far the target temperature has been reached. In addition, the reward is set in the form of an area reward as an index for evaluating each air-conditioned area 40, and an overall reward as an index for evaluating the room 100 as a whole. The area reward is calculated, for example, based on the difference between the target temperature and the temperature after control of the air conditioner 4 for a specified time. Specifically, assume that the temperature (area average temperature) of the relevant air-conditioned area 40 at the start of control is 38 degrees, and the target temperature is 31 degrees. In this case, if the temperature after control is 32 degrees, which is 1 degree higher than the target temperature, the reward will be "90%". Note that here, when the temperature is 1 degree lower than the target temperature, the reward is calculated as "-10%".

[0052] The overall reward is an index for judging the pass / fail of the air conditioning control for the entire room 100. This overall reward is used to judge the pass / fail of the air conditioning control for the entire room 100 based on whether or not the reward (overall reward) calculated according to a predetermined logic using the area rewards is equal to or greater than a predetermined threshold (pass threshold). The predetermined logic is, for example, that the average value of the area rewards is equal to or greater than a predetermined threshold (pass threshold), and is set arbitrarily.

[0053] If the calculated overall reward in the air-conditioning control learning model for the corresponding Situation classification exceeds a predetermined pass threshold and is determined to be pass, the placement pattern determination unit 214 ends the learning phase and transitions to the operation phase. If the placement pattern determination unit 214 determines that the overall reward is acceptable, it obtains, for each Situation classification, the air conditioning control values ​​(target temperature, air volume, etc.) and the power consumption (air conditioning power consumption) of each air conditioner 4 during that air conditioning control, and generates control value-power consumption correspondence information 233.

[0054] After the air-conditioning control learning model for each Situation classification transitions from the learning phase to the operation phase, that is, after it converges to an air-conditioning control value that satisfies a predetermined reward (overall reward) in a certain Situation classification, the allocation pattern determination unit 214 may search for an air-conditioning control value with a lower control cost (lower power consumption cost) that satisfies the reward (overall reward) from the converged value by gradually executing a trial search for air-conditioning control values ​​in the lower cost direction (degenerate operation).When degenerate operation is performed, the allocation pattern determination unit 214 updates the control value-power amount correspondence information 233 with the air-conditioning control value that has the lower power consumption cost at the end of the degenerate operation and the air-conditioning power consumption amount when the air-conditioning control value is executed. This concludes the explanation of the process for generating the control value-power amount correspondence information 233 using the learning model (air-conditioning control learning model).

[0055] The arrangement pattern determination unit 214 determines an arrangement pattern with the smallest total room power consumption in the situation classification of each room 100 as described above. For example, as shown in FIG. 7, the arrangement pattern determination unit 214 determines an arrangement pattern with the smallest total room power consumption (load arrangement pattern of each server area 30 shown by reference symbol c in FIG. 7) corresponding to the outside air temperature and the air intake temperature of the server area obtained from the temperature sensor (reference symbol b in FIG. 7) in the load amount allocated to the room 100 of the arrangement pattern determination unit 214. The arrangement pattern determination unit 214 also stores information on the total room power consumption in the load arrangement pattern of each situation classification as the operation history information 232. Then, the arrangement pattern determination unit 214 transmits information on the total room power consumption in the load arrangement pattern of the room 100 of the arrangement pattern determination unit 214 (information on the smallest total room power consumption) to the power amount control device 10.

[0056] The load placement control unit 215 places virtual resources (VMs, containers, etc.) on each server 3 based on the load placement pattern for each server 3. When the load placement control unit 215 acquires information on the load allocation ratio of each room 100 indicated by the inter-room load allocation pattern determined by the power amount control device 10, the load placement control unit 215 determines the load amount allocated to the room it is responsible for based on information on the total load amount (total DC load amount).Then, the load placement control unit 215 places virtual resources (VMs, containers, etc.) on the servers 3 in each server area 30 based on the placement pattern that has the smallest total power consumption in the rooms determined by the placement pattern determination unit 214 for that load amount.

[0057] The air conditioning control unit 216 controls each air conditioner 4 based on information on air conditioning control values ​​(target temperature, air volume, etc.) in a certain situation calculated by the arrangement pattern determination unit 214. In addition, the air conditioning control unit 216 stores, for each Situation classification, external factor information, air conditioning control values ​​(target temperature, air volume, etc.), rewards (area rewards) obtained when executing that control, and information on air conditioning power consumption as operation history information 232. In addition, when the load placement control unit 215 acquires information on the load allocation ratio of each room 100 indicated by the inter-room load allocation pattern determined by the power amount control device 10, the air conditioning control unit 216 controls each air conditioner 4 based on information on the air conditioning control values ​​(target temperature, air volume, etc.) in the placement pattern determined by the placement pattern determination unit 214 to have the smallest total power consumption in the room for the confirmed load amount.

[0058] <Processing flow> Next, the flow of processing executed by the power amount control system 1 according to this embodiment will be described. FIG. 8 is a sequence diagram showing the flow of processing executed by the power amount control system 1. As shown in FIG. In this case, it is assumed that the server power amount learning model 231 and the control value-power amount correspondence information 233 are stored in advance in the storage unit 23 of each in-room power amount control device 20.

[0059] First, the inter-room load allocation pattern generating unit 111 of the power amount control device 10 acquires the total load amount (DC total load amount) to be processed in the DC 1000 from an external device (such as a system management device) (step S1).

[0060] Next, the inter-room load allocation pattern generating unit 111 generates an inter-room load allocation pattern indicating N different load allocations of the total DC load amount obtained by changing the load allocation ratio to each room 100 (step S2). The inter-room load allocation pattern generation unit 111 transmits, for each of the generated inter-room load allocation patterns, the total DC load amount and the load allocation ratio of each room 100 to the total DC load amount indicated in the inter-room load allocation pattern to each of the in-room power amount control devices 20 (step S3).

[0061] Next, the placement pattern calculation unit 212 of each in-room power energy control device 20 obtains information on the load to be allocated to the room it is responsible for based on information on the load allocation ratio of each room 100 indicated in the inter-room load allocation pattern and information on the total load (total DC load) obtained from the power energy control device 10. Then, the arrangement pattern calculation unit 212 calculates an arrangement pattern for allocating the virtual resources to each server 3 in the own room 100 (step S4).

[0062] Next, the server power consumption estimation unit 213 of each in-room power consumption control device 20 estimates the power consumption of each server 3 for each arrangement pattern calculated by the arrangement pattern calculation unit 212 using the server power consumption learning model 231. Then, the server power consumption estimation unit 213 calculates the server power consumption for each server area 30 by adding up the server power consumptions of the servers 3 in each server area 30 based on the server layout for each server area 30 (step S5).

[0063] Next, the situation recognition unit 211 of each in-room power energy control device 20 acquires information (external factor information) on the outside air temperature, the air intake temperature of the server area, and the server power consumption of each server area 30 for each arrangement pattern, and determines the situation classification (step S6).

[0064] Next, the placement pattern determination unit 214 of each in-room power consumption control device 20 sums up the server power consumption for each server area 30 in each placement pattern, and calculates the sum of the total server power consumption, which is the sum value, and the air conditioning power consumption obtained from the control value power consumption correspondence information 233 stored in the memory unit 23 (total power consumption in the room) (step S7).

[0065] Then, the placement pattern determination unit 214 determines, from among the placement patterns, the placement pattern with the smallest total amount (total power consumption in the room) as the placement pattern in which to place the virtual resources in the Situation classification of the room 100. Next, the arrangement pattern determination unit 214 transmits information on the minimum total amount of power consumption in the room in the load arrangement pattern among the load amounts allocated to the own room 100 to the power amount control device 10 (step S8).

[0066] Next, the load distribution ratio determination unit 112 of the power amount control device 10 sums up the total power consumption of each room 100 for each inter-room load distribution pattern, and determines the inter-room load distribution pattern that minimizes the sum (total power consumption) (step S9).

[0067] Next, the load distribution ratio determination unit 112 transmits the load distribution ratios of the rooms 100 indicated by the determined inter-room load distribution pattern to each in-room power amount control device 20 (step S10).

[0068] When the load placement control unit 215 of each in-room power energy control device 10 acquires information on the load allocation ratio of each room 100 indicated by the inter-room load allocation pattern determined by the power energy control device 10, the load placement control unit 215 determines the load amount allocated to the room it is responsible for based on the information on the total load amount (total DC load amount). Then, the load placement control unit 215 places virtual resources (VMs, containers, etc.) on the servers 3 in each server area 30 based on the placement pattern with the smallest in-room total power consumption amount determined by the placement pattern determination unit 214 for that load amount. Also, the air conditioning control unit 216 refers to the control value power energy correspondence information 233 and controls each air conditioner 4 based on the information on the air conditioning control value (target temperature, air volume, etc.) in that placement pattern (step S11).

[0069] In this way, the power amount control system 1 can appropriately determine the load allocation ratio to each room 100 with respect to the total load amount set in the DC 1000. Therefore, it is possible to reduce the power consumption of the entire DC 1000, which is the sum of the power consumption of each room 100.

[0070] <Hardware configuration> The power amount control device 10 and the room power amount control device 20 according to this embodiment are realized by a computer 900 having a configuration as shown in FIG. 9, for example. 9 is a hardware configuration diagram showing an example of a computer 900 that realizes the functions of the power amount control device 10 and the room power amount control device 20 according to this embodiment. The computer 900 has a CPU (Central Processing Unit) 901, a ROM (Read Only Memory) 902, a RAM 903, an HDD (Hard Disk Drive) 904, an input / output I / F (Interface) 905, a communication I / F 906, and a media I / F 907.

[0071] The CPU 901 operates based on a program stored in the ROM 902 or the HDD 904, and performs control by the control unit. The ROM 902 stores a boot program executed by the CPU 901 when the computer 900 is started up, programs related to the hardware of the computer 900, and the like.

[0072] The CPU 901 controls an input device 910 such as a mouse or a keyboard, and an output device 911 such as a display or a printer, via an input / output I / F 905. The CPU 901 acquires data from the input device 910 via the input / output I / F 905, and outputs generated data to the output device 911. Note that a GPU (Graphics Processing Unit) or the like may be used as a processor together with the CPU 901.

[0073] The HDD 904 stores programs executed by the CPU 901 and data used by the programs, etc. The communication I / F 906 receives data from other devices via a communication network (e.g., NW (Network) 920) and outputs the data to the CPU 901, and also transmits data generated by the CPU 901 to other devices via the communication network.

[0074] The media I / F 907 reads a program or data stored in the recording medium 912, and outputs it to the CPU 901 via the RAM 903. The CPU 901 loads a program related to a target process from the recording medium 912 onto the RAM 903 via the media I / F 907, and executes the loaded program. The recording medium 912 is an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disc), a magneto-optical recording medium such as an MO (Magneto Optical disk), a magnetic recording medium, a semiconductor memory, or the like.

[0075] For example, when the computer 900 functions as the power amount control device 10 and the room power amount control device 20 of the present invention, the CPU 901 of the computer 900 executes a program loaded on the RAM 903 to realize the functions of the power amount control device 10 and the room power amount control device 20. Furthermore, the HDD 904 stores data in the RAM 903. The CPU 901 reads and executes a program related to a target process from the recording medium 912. Additionally, the CPU 901 may read a program related to a target process from another device via a communication network (NW 920).

[0076] <Effects> The effects of the power amount control system 1 according to the present invention will be described below. The power consumption control system of the present invention is an power consumption control system 1 comprising: a room power consumption control device 20 that is communicatively connected to the servers 3 and air conditioners 4 in each room 100 in a data center 1000 in which a plurality of rooms 100 are installed, each room 100 having a plurality of servers 3 and one or more air conditioners 4 arranged therein, and that is provided for each room 100 to control the amount of power consumption in the room 100; and a power consumption control device 10 that is communicatively connected to the room power consumption control devices 20 for each room 100, and that controls the amount of power consumption of the entire data center 1000. Each room 100 in the data center 1000 is provided with a plurality of server areas 30 in which a group of servers are arranged on which virtual resources are placed as loads, and a plurality of air conditioning areas 40 in which the effect of air conditioning control by the air conditioners 4 is measured.

[0077] The power consumption control device 10 is equipped with an allocation pattern generation unit (inter-room load allocation pattern generation unit 111) that acquires information on the total load to be processed within the data center, generates multiple inter-room load allocation patterns indicating load allocation patterns in which the load allocation ratio to each room 100 is changed, and transmits the multiple inter-room load allocation patterns to each in-room power consumption control device 20 responsible for the room 100, and a load allocation ratio determination unit 112 that receives from each in-room power consumption control device 20 an in-room total power consumption indicating the power consumption of the entire room 100 in a load arrangement pattern in which the total amount of server power consumption and air conditioning power consumption in the room is the smallest, for each inter-room load allocation pattern, according to the load of each room 100 calculated from the load allocation ratio, and sums up the in-room total power consumption received from each in-room power consumption control device 20 for each inter-room load allocation pattern, and determines the inter-room load allocation pattern in which the total power consumption indicating the total amount is the smallest as the load allocation ratio to each room 100.

[0078] The in-room power consumption control device 20 includes a situation recognition unit 211 that acquires information on external factors including the intake port temperature in the server area 30, the outside temperature, and the server power consumption for each server area 30, which is a predicted amount when virtual resources as loads are allocated to the servers 3, divides the value of each external factor into a predetermined range width, defines a combination of the ranges divided for each external factor as a situation classification, and determines which situation classification the acquired information on the external factor belongs to; an allocation pattern calculation unit 212 that calculates the load allocated to its own room using the total load amount and the load allocation ratio, and calculates an allocation pattern in which the virtual resources as the calculated load are allocated to one of the multiple servers 3; and a server power consumption estimation unit 213 that estimates the server power consumption of a group of servers belonging to each server area 30 for each calculated allocation pattern. and an arrangement pattern determination unit 214 that acquires from the situation recognition unit 211 a determination result of a situation classification using the server power consumption of each server area 30, refers to control value-power correspondence information 233 that associates the air conditioning control value of the air conditioner 4 with the air conditioning power consumption for each situation classification, acquires the air conditioning control value and air conditioning power consumption of the air conditioner 4 for that situation classification for each arrangement pattern, totals the server power consumption for each server area 30 for each arrangement pattern, calculates the total sum of the total server power consumption and air conditioning power consumption, determines the arrangement pattern with the smallest calculated total amount as the arrangement pattern for allocating virtual resources as a load, and transmits the calculated smallest total amount for the determined arrangement pattern to the power consumption control device 10 as the total power consumption in the room.

[0079] In this way, the power amount control system 1 can appropriately determine the load allocation ratio to each room 100 with respect to the total load amount set in the DC 1000. Therefore, it is possible to reduce the power consumption of the entire DC 1000, which is the sum of the power consumption of each room 100.

[0080] The present invention is not limited to the above-described embodiment, and many modifications can be made by those having ordinary skill in the art within the technical scope of the present invention. [Explanation of symbols]

[0081] 1. Power control system 3 Server 4 Air conditioner 10. Electricity amount control device 11,21 Control section 12,22 Input / output section 13,23 Storage part 20. Room power consumption control device 30 Server Area 40 Air-conditioned area 100 Rooms 111 Inter-room load distribution pattern generation unit (distribution pattern generation unit) 112 Load distribution ratio determination section 211 Situational Awareness Division 212 Placement pattern calculation unit 213 Server Power Consumption Estimation Unit 214 Placement pattern determination unit 215 Load placement control unit 216 Air conditioning control unit 231 Server Power Consumption Learning Model 232 Operational History Information 233 Control value power amount correspondence information 1000 Data Centers (DC)

Claims

1. An electric energy control system comprising: an in-room electric energy control device provided for each room in a data center in which a plurality of rooms are installed, each room having a plurality of servers and one or more air conditioners arranged therein, the in-room electric energy control device being communicatively connected to the servers and air conditioners in the rooms and controlling the electric energy consumption in the rooms; and an electric energy control device being communicatively connected to the in-room electric energy control device for each room and controlling the electric energy consumption of the entire data center, In each room of the data center, a plurality of server areas in which a group of servers on which virtual resources are allocated as loads are arranged, and a plurality of air conditioning areas in which effects of air conditioning control by the air conditioners are measured are set, The power amount control device includes: a distribution pattern generation unit that acquires information on a total load amount to be processed in the data center, generates a plurality of inter-room load distribution patterns indicating load distribution patterns in which a load distribution ratio to each room is changed, and transmits the inter-room load distribution patterns to each of the in-room power amount control devices in charge of the rooms; receiving from each of the in-room power amount control devices, for each of the inter-room load allocation patterns, a total in-room power consumption amount indicating the power amount of the entire room in the load arrangement pattern in which the total amount of the server power consumption and the air-conditioning power consumption in the room is the smallest according to the load of each room calculated from the load allocation ratio; a load distribution ratio determination unit that sums up the total in-room power consumption received from each of the in-room power amount control devices for each of the inter-room load distribution patterns, and determines an inter-room load distribution pattern that minimizes the total power consumption indicating the sum as a load distribution ratio to each room; The room electric power amount control device includes: a situation recognition unit that acquires information about external factors including an intake temperature and an outside temperature in the server area, and a server power consumption amount for each server area which is a predicted amount when a virtual resource as the load is allocated to the server, divides the value of each of the external factors into a predetermined range width, defines a combination of the ranges divided for each of the external factors as a situation classification, and determines to which situation classification the acquired information about the external factors belongs; an allocation pattern calculation unit that calculates a load allocated to its own room using the total load and the load allocation ratio, and calculates an allocation pattern for allocating the virtual resources as the calculated load to one of the plurality of servers; a server power consumption estimation unit that estimates server power consumption of a server group that belongs to each of the server areas for each of the calculated placement patterns; obtaining, from the situation recognition unit, a determination result of a situation classification using the server power consumption of each of the server areas, and for each of the situation classifications, referring to control value-power-amount correspondence information that associates the air-conditioning power consumption with the air-conditioning control value of the air conditioner, obtaining the air-conditioning control value and the air-conditioning power consumption of the air conditioner in the situation classification for each arrangement pattern; an arrangement pattern determination unit that sums up the server power consumption for each server area in each arrangement pattern, calculates a sum of the total server power consumption, which is the sum, and the air-conditioning power consumption, determines the arrangement pattern in which the calculated sum is the smallest as the arrangement pattern in which the virtual resources as the load are arranged, and transmits the calculated smallest sum in the determined arrangement pattern to the power amount control device as the total power consumption in the room. A power amount control system comprising:

2. A power amount control method for a power amount control system including: a data center having a plurality of rooms in which a plurality of servers and one or more air conditioners are arranged; an in-room power amount control device provided for each room, the in-room power amount control device being communicatively connected to the servers and air conditioners in the rooms and controlling the amount of power consumed in the rooms; and a power amount control device communicatively connected to the in-room power amount control device for each room and controlling the amount of power consumed in the entire data center, the method comprising: In each room of the data center, a plurality of server areas in which a group of servers on which virtual resources are allocated as loads are arranged, and a plurality of air conditioning areas in which effects of air conditioning control by the air conditioners are measured are set, The power amount control device includes: acquiring information on a total load to be processed in the data center, generating a plurality of inter-room load distribution patterns indicating load distribution patterns in which a load distribution ratio to each room is changed, and transmitting the patterns to each of the intra-room power amount control devices in charge of the rooms; receiving, for each of the inter-room load allocation patterns, from each of the in-room power amount control devices, a total in-room power amount indicating the power amount of the entire room in the load arrangement pattern in which the total amount of server power amount and air-conditioning power amount in the room is the smallest according to the load of each room calculated from the load allocation ratio; a step of summing up the total in-room power consumption amounts received from each of the in-room power consumption control devices for each of the inter-room load distribution patterns, and determining, as a load distribution ratio for each room, the inter-room load distribution pattern that minimizes the total power consumption amount indicating the sum of the total amounts; The room electric power amount control device includes: a step of acquiring information on external factors including the intake temperature and outside temperature in the server area, and the server power consumption for each server area, which is a predicted amount when the virtual resource as the load is placed on the server, dividing the value of each of the external factors into a predetermined range width, combining the ranges divided for each of the external factors to define them as a situation classification, and determining to which situation classification the acquired information on the external factors belongs; calculating a load allocated to the room by using the total load and the load allocation ratio, and calculating an allocation pattern for allocating the virtual resources as the calculated load to one of the plurality of servers; estimating server power consumption of a group of servers belonging to each of the server areas for each of the calculated placement patterns; acquiring a determination result of a situation classification using the server power consumption of each of the server areas, and for each situation classification, referring to control value-power correspondence information that associates the air conditioning power consumption with the air conditioning control value of the air conditioner, acquiring the air conditioning control value and the air conditioning power consumption of the air conditioner in the situation classification for each arrangement pattern; a step of summing up the server power consumption for each server area in each of the arrangement patterns, calculating the sum of the total server power consumption and the air conditioning power consumption, determining the arrangement pattern with the smallest calculated sum as the arrangement pattern for allocating the virtual resources as the load, and transmitting the calculated smallest sum in the determined arrangement pattern to the power amount control device as the total power consumption in the room. A power amount control method comprising:

Images