Data centers and information processing systems

Abstract

To provide a data center and information processing system that can improve cooling performance. [Solution] The data center comprises a server having a heat-generating element, a cooling system for cooling the heat-generating element, and a control device for controlling the cooling system. The cooling system comprises a cooling section for cooling the heat-generating element with a refrigerant, a cooling heat source unit for cooling the refrigerant, a circulation channel for circulating the refrigerant between the cooling section and the cooling heat source unit, a bypass channel that bypasses the circulation channel so that the refrigerant that has passed through the cooling heat source unit in the circulation channel is introduced to the upstream portion of the circulation channel relative to the cooling section, and a tank in the bypass channel capable of storing the refrigerant cooled by the cooling heat source unit. The control device controls the cooling system so that the refrigerant in the tank can be released into the circulation channel if it determines that the predicted value of power consumption, which is the sum of the power consumed by the server and the power consumed by the cooling heat source unit, exceeds the predicted value of available power, including the contracted power.

Description

Technical Field

[0001] The present disclosure relates to a data center and an information processing system.

Background Art

[0002] Patent Document 1 discloses an information processing system that performs migration between data centers located at a plurality of geographically separated positions. Here, the migration means causing a job executed by a server in one data center to be executed by a server in another data sensor. This information processing system predicts the transition of power generated by natural energy in each data center and migrates jobs to a data center with a large amount of available power. Thereby, jobs can be operated in a data center with a large amount of available power.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, a data center is provided with a mechanism for cooling a server that generates heat. In the information processing system described in Patent Document 1, for example, when the prediction of the available power amount is off, there is a problem that the power for server cooling becomes insufficient and the server is not sufficiently cooled.

[0005] The present disclosure has been made to solve the above problems, and an object thereof is to provide a data center and an information processing system capable of improving cooling performance.

Means for Solving the Problems

[0006] To solve the above problems, the data center according to this disclosure comprises a server having a heat-generating element, a cooling system for cooling the heat-generating element, and a control device for controlling the cooling system, wherein the cooling system comprises a cooling unit for cooling the heat-generating element with a refrigerant, a cooling heat source unit for cooling the refrigerant, a circulation channel for circulating the refrigerant between the cooling unit and the cooling heat source unit, a bypass channel that bypasses the circulation channel so that the refrigerant that has passed through the cooling heat source unit in the circulation channel is introduced into the upstream portion of the circulation channel relative to the cooling unit, and a tank provided in the bypass channel capable of storing the refrigerant cooled by the cooling heat source unit, wherein the control device controls the cooling system so as to release the refrigerant in the tank into the circulation channel when it is determined that the predicted value of power consumption, which is the sum of the power consumed by the server and the power consumed by the cooling heat source unit, exceeds the predicted value of usable power, including contracted power.

[0007] The information processing system relating to this disclosure comprises a plurality of data centers located at geographically separated locations, and a migration control unit that migrates jobs between the plurality of data centers according to the available power in each of the data centers. [Effects of the Invention]

[0008] According to the data center and information processing system disclosed herein, cooling performance can be improved. [Brief explanation of the drawing]

[0009] [Figure 1] This is a schematic diagram of an information processing system according to an embodiment of the present disclosure. [Figure 2] This is a schematic diagram of the data center according to the embodiment of this disclosure. [Figure 3] This is a functional block diagram of a control device according to an embodiment of the present disclosure. [Figure 4] This flowchart shows an example of the procedure for the operating mode of the cooling system according to the embodiment of this disclosure. [Figure 5] This is a functional block diagram of a control device relating to the first modified example of the present disclosure. [Figure 6] This is a functional block diagram of the migration control unit relating to the first modified example of the present disclosure. [Figure 7] This flowchart shows an example of the procedure for the operating mode of the cooling system according to the first modified example of this disclosure. [Figure 8] This flowchart shows an example of the migration procedure for an information processing system relating to the first modified example of this disclosure. [Figure 9] This is a schematic diagram of another example of a data center relating to the first modified example of this disclosure. [Figure 10] This is a functional block diagram of a control device relating to a second modified example of the present disclosure. [Figure 11] This flowchart shows an example of the procedure for the operating mode of the cooling system according to the second modified example of the present disclosure. [Figure 12] This flowchart shows an example of the procedure for the operating mode of the cooling system according to the second modified example of the present disclosure. [Figure 13] This is a hardware configuration diagram according to an embodiment of the present disclosure. [Modes for carrying out the invention]

[0010] (Configuration of the information processing system) Hereinafter, the information processing system 1 and data center 10 according to the embodiments of this disclosure will be described with reference to Figures 1 to 4. As shown in Figure 1, the information processing system 1 comprises a data center 10, a migration control unit 2, and a network 3.

[0011] The data center 10 is a facility equipped with multiple servers 11. In this embodiment, the data center 10 has a renewable energy generator 51 (see Figure 2). Renewable energy refers to natural energy that is replenished steadily or repeatedly by natural forces, such as solar power, wind power, wave power / tidal power, flowing water / tides, geothermal energy, and biomass. Examples of renewable energy power generation include solar power generation, wind power generation, hydroelectric power generation, wave power generation, geothermal power generation, and biomass power generation. Hereinafter, the electricity generated by the renewable energy generator 51 will be referred to as renewable energy power. The data center 10 is capable of using renewable energy power in addition to electricity purchased from an external power generation facility 4 (purchased power). The other components of the data center 10 will be described in detail later.

[0012] Furthermore, the data center 10 in this embodiment is a so-called distributed data center, and multiple data centers are arranged. The information processing system 1 distributes jobs to the multiple data centers 10 and executes jobs individually at each data center 10. In addition, each data center 10 is located in a geographically separated location so that if an abnormality occurs in any of the data centers 10 due to a disaster or the like, the other data centers 10 will not be affected by the abnormality. For example, if the data centers 10 utilize electricity generated by solar power, each data center 10 is located in a region with different hours of sunlight.

[0013] Furthermore, a plurality of data centers 10 are communicably connected to each other by a network 3. Each data center 10 is connected to a migration control unit 2 by the network 3. The migration control unit 2 migrates jobs among the plurality of data centers 10 according to the available power in each data center 10. The available power includes the contracted power. The contracted power is the upper limit of the amount of power that can be used within a predetermined period. Generally, if power exceeding the contracted power is used, the power unit price increases and the cost becomes high. For this reason, it is desirable that the purchased power be within the contracted power. In addition, in the present embodiment, since the data center 10 uses renewable energy power in addition to the purchased power, the available power in the present embodiment means the total value of the contracted power and the renewable energy power unless otherwise specified.

[0014] The migration control unit 2 migrates jobs among the plurality of data centers 10 so that the data center 10 with more available power executes jobs with high power consumption or more jobs than other data centers 10. In the present embodiment, the migration control unit 2 predicts the transition of the power generated by renewable energy (renewable energy power) in each data center 10 and migrates jobs to the data center 10 with more renewable energy power. Specifically, the migration control unit 2 performs migration based on weather information such as weather forecasts in each data center 10, for example. More specifically, the migration control unit 2 predicts the power consumption and the renewable energy power of each data center 10, determines the jobs to be assigned to each data center 10, and optimally operates the plurality of data centers 10. In addition, the power consumption in the present embodiment is the total value of the power consumed by the server 11 (computing processing power) and the power consumed by a cooling heat source machine 32 (see FIG. 2) described later (cooling processing power).

[0015] Each data center 10 executes a job assigned by the migration control unit 2. By migrating jobs among a plurality of data centers 10 in this way, it is possible to efficiently use the available power among the data centers 10 without waste and to operate the plurality of data centers 10 efficiently. In this embodiment, the migration control unit 2 is provided separately from each data center 10. However, the control device 60 (see FIG. 2) of any one of the plurality of data centers 10 may have the migration control unit 2.

[0016] (Configuration of Data Center) Next, the configuration of the data center 10 will be described. The configurations of the plurality of data centers 10 are substantially the same. Therefore, one of the plurality of data centers 10 will be described, and the description of the other data centers 10 will be omitted. As shown in FIG. 2, the data center 10 includes a server 11, a cooling system 20, a power supply system 50, a control device 60, and a network 13.

[0017] (Server) The server 11 has a heating element 12. Hereinafter, the case where the heating element 12 is a plurality of chips such as CPUs and GPUs mounted on the server 11 will be taken as an example to describe the embodiment. These chips generate heat during operation. The heating element 12 is a cooling target to be cooled by the cooling system 20.

[0018] (Cooling System) The cooling system 20 cools the heating element 12. The cooling system 20 includes a main cooling system 30, a storage system 40, and a refrigerant mixing unit 21.

[0019] (Main Cooling System) The main cooling system 30 includes a cooling unit 31, a cooling heat source machine 32, a circulation flow path 33, a main pump 34, and a main valve 35.

[0020] (Cooling Unit) The cooling unit 31 cools the heat-generating element 12 with a refrigerant R. The cooling unit 31 is driven by power supplied from a power supply system 50, which will be described later. An example of the cooling unit 31 is a cold plate attached to the heat-generating element 12. However, the cooling unit 31 is not limited to a cold plate. For example, the cooling unit 31 may immerse the heat-generating element 12 in a refrigerant R stored inside. The cooling unit 31 only needs to cool the heat-generating element 12 by heat exchange with the refrigerant R. Examples of the refrigerant R include cooling water, LLC (Long Life Coolant), antifreeze, etc. Furthermore, the heat-generating element 12 has a specified appropriate temperature for the refrigerant R. In this embodiment, when the heat-generating element 12 is a chip such as a GPU or CPU, the temperature of the refrigerant R suitable for cooling the heat-generating element 12 is, for example, in the range of 10 to 20 degrees Celsius. The refrigerant R exchanges heat with the heat-generating element 12 in the cooling unit 31. As a result, the heat-generating element 12 is cooled and the refrigerant R is heated.

[0021] (Cooling heat source machine) The cooling heat source unit 32 cools the refrigerant R heated in the cooling unit 31. The cooling heat source unit 32 is driven by power supplied from the power supply system 50, which will be described later. In this embodiment, the cooling heat source unit 32 is, for example, a chiller. This chiller cools the refrigerant R by exchanging heat with outside air blown by a fan. The cooling heat source unit 32 has a cooling capacity of, for example, 1.5 times or more the normal load on the heat generating element 12.

[0022] (Circulation channel) The circulation channel 33 circulates the refrigerant R between the cooling unit 31 and the cooling heat source unit 32. The circulation channel 33 comprises a low-temperature circulation channel 33a and a high-temperature circulation channel 33b. The low-temperature circulation channel 33a guides the refrigerant R cooled by the cooling heat source unit 32 to the cooling unit 31. The high-temperature circulation channel 33b guides the refrigerant R heated by the cooling unit 31 to the cooling heat source unit 32.

[0023] (Main pump) The main pump 34 is located in the low-temperature circulation channel 33a. The main pump 34 pumps the refrigerant R from the cooling heat source unit 32 to the cooling unit 31.

[0024] (Main valve) The main valve 35 is located downstream of the main pump 34 in the low-temperature circulation channel 33a. The main valve 35 opens and closes the circulation channel 33.

[0025] (Storage system) The storage system 40 stores the refrigerant R cooled by the cooling heat source unit 32. The storage system 40 has better thermal insulation than the cooling system 20. The storage system 40 includes a bypass channel 41, a tank 42, a pump 43, a supply valve 44, and a discharge valve 45.

[0026] (Bypass channel) The bypass channel 41 bypasses the circulation channel 33. The bypass channel 41 introduces the refrigerant R that has passed through the cooling heat source unit 32 in the circulation channel 33 to the upstream portion of the circulation channel 33 relative to the cooling section 31. In this embodiment, the upstream end of the bypass channel 41 is connected between the main pump 34 and the main valve 35 in the low-temperature circulation channel 33a. The downstream end of the bypass channel 41 is connected to the downstream portion of the high-temperature circulation channel 33b, near the cooling heat source unit 32.

[0027] (tank) Tank 42 is located in the bypass channel 41. Tank 42 is capable of storing refrigerant R cooled by the cooling heat source unit 32. For example, if refrigerant R is LLC, tank 42 can store refrigerant R at a temperature of -20°C to -10°C. It is desirable that the temperature of the refrigerant R stored in tank 42 be about 5°C higher than the freezing point (freezing temperature) of refrigerant R.

[0028] (pump) Pump 43 is located in the bypass passage 41 downstream of tank 42. Pump 43 pumps the refrigerant R stored in tank 42 towards circulation passage 33.

[0029] (Supply valve) The supply valve 44 is located in the bypass passage 41 upstream of the tank 42. The supply valve 44 opens and closes the bypass passage 41.

[0030] (Discharge valve) The discharge valve 45 is located in the bypass passage 41 downstream of the pump 43. The discharge valve 45 opens and closes the bypass passage 41.

[0031] (refrigerant mixing section) The refrigerant mixing unit 21 mixes the refrigerant R flowing through the circulation channel 33 with the refrigerant R released from the tank 42. The refrigerant mixing unit 21 is located at the connection point between the circulation channel 33 and the downstream end of the bypass channel 41. Examples of the refrigerant mixing unit 21 include a mixer and a mixing tank.

[0032] (Power supply system) The power supply system 50 supplies power to the cooling system 20. The power supply system 50 includes a renewable energy generator 51, a power supply line 52, and a power meter 53.

[0033] (Renewable energy generator) The renewable energy generator 51 generates electricity using renewable energy. The renewable energy generator 51 is capable of supplying renewable energy power to the cooling unit 31 and the cooling heat source unit 32. As mentioned above, renewable energy refers to natural energy that is replenished steadily or repeatedly by natural forces, such as solar power, wind power, wave power / tidal power, flowing water / tides, geothermal energy, and biomass. Renewable energy power refers to electricity generated by the renewable energy generator 51 using such renewable energy. Examples of the renewable energy generator 51 include solar power generators and wind power generators. The renewable energy generator 51 may also consist of two or more generators that generate electricity using different renewable energy sources. For example, the renewable energy generator 51 may include a solar power generator and a wind power generator.

[0034] (Power supply line) The power supply line 52 connects the externally installed power generation equipment 4 and renewable energy generator 51 to the motors of the cooling system 20 (for example, the cooling unit 31 and the cooling heat source unit 32). The power supply line 52 supplies the electricity generated by the power generation equipment 4 and renewable energy generator 51 to the cooling unit 31 and the cooling heat source unit 32.

[0035] (wattmeter) The power meter 53 measures the amount of electricity supplied to the cooling unit 31 and the cooling heat source unit 32. The power meter 53 can measure the amount of purchased electricity purchased from the power generation equipment 4 and the amount of renewable energy electricity generated by the renewable energy generator 51.

[0036] (Control device) The control device 60 controls various equipment and devices that make up the data center 10. The control device 60 can control the cooling system 20, the power supply system 50, and the server 11. As shown in Figure 3, the control device 60 has functional units consisting of an acquisition unit 61, a prediction unit 62, and a determination unit 63.

[0037] (Acquisition part 61) The acquisition unit 61 acquires information related to the control of the cooling system 20. For example, the acquisition unit 61 acquires information from within the data center 10, such as the cooling system 20 and the power supply system 50, as well as information from other data centers 10 and information input from the outside. Specifically, the acquisition unit 61 acquires the amount of purchased power and renewable energy power measured by the power meter 53. The acquisition unit 61 also acquires weather information related to renewable energy power.

[0038] (Prediction section) The prediction unit 62 includes a renewable energy power prediction unit 62a, a usable power prediction unit 62b, and a power consumption prediction unit 62c.

[0039] (Renewable Energy Power Forecasting Department) The renewable energy power prediction unit 62a predicts renewable energy power based on the information acquired by the acquisition unit 61. In other words, the renewable energy power prediction unit 62a calculates a predicted value for renewable energy power.

[0040] (Available power prediction unit 62b) The available power prediction unit 62b predicts the available power based on the information acquired by the acquisition unit 61 and the predicted values ​​calculated by the renewable energy power prediction unit 62a. In other words, the available power prediction unit 62b calculates a predicted value of the available power. The available power includes the contracted power. In this embodiment, the available power is the sum of the contracted power and the renewable energy power.

[0041] (Power consumption prediction unit) The power consumption prediction unit 62c predicts power consumption based on the information acquired by the acquisition unit 61. That is, the power consumption prediction unit 62c calculates a predicted value of power consumption. In this embodiment, power consumption is the sum of the power consumed by the server 11 and the power consumed by the cooling heat source unit 32. Hereinafter, the power consumed by the server 11 may be referred to as calculation processing power, and the power consumed by the cooling heat source unit 32 may be referred to as cooling processing power.

[0042] (Judgment Department) The determination unit 63 includes a power consumption determination unit 63a.

[0043] (Power consumption determination section) The power consumption determination unit 63a determines whether the predicted power consumption exceeds the predicted usable power.

[0044] (Cooling System Control Unit) The cooling system control unit 64 controls the cooling system 20 according to the conditions. For example, if the cooling system control unit 64 determines that the predicted value of power consumption exceeds the predicted value of available power, it controls the cooling system 20 so that the refrigerant R in the tank 42 can be released into the circulation channel 33.

[0045] (Operation of the cooling system) Next, we will explain the operation of the cooling system 20.

[0046] First, the basic operation of the main cooling system 30 will be explained. The cooling system control unit 64 drives the main pump 34 to circulate the refrigerant R in the circulation channel 33. As a result, the refrigerant R cooled by the cooling heat source unit 32 is sent to the cooling unit 31 through the low-temperature circulation channel 33a. At this time, the cooling system control unit 64 adjusts the flow rate of the refrigerant R flowing through the circulation channel 33 by adjusting the pumping flow rate of the main pump 34. The refrigerant R sent to the cooling unit 31 is at a temperature suitable for cooling the heat source unit 12 (for example, 10 to 20 degrees Celsius). Heat exchange takes place between the refrigerant R and the heat source unit 12 in the cooling unit 31. As a result, the heat source unit 12 is cooled and maintained at a temperature suitable for operation. On the other hand, the refrigerant R is heated in the cooling unit 31. The refrigerant R is heated to, for example, about 30 degrees Celsius. The refrigerant R heated in the cooling unit 31 is sent to the cooling heat source unit 32 through the high-temperature circulation channel 33b. The refrigerant R is cooled in the cooling heat source unit 32 and then sent back to the cooling unit 31 to cool the heat-generating element 12. In this way, the main cooling system 30 circulates the refrigerant R in the circulation channel 33 to cool the heat-generating element 12 and maintain a constant temperature for the heat-generating element 12.

[0047] Here, for example, if the load on the heat-generating element 12 increases, the refrigerant R becomes hot, which may adversely affect the operation of the main cooling system 30 and the server 11. For this reason, the main cooling system 30 is designed with low thermal insulation to maximize heat dissipation, making it difficult to maintain the refrigerant R in the main cooling system 30 at a temperature suitable for cooling the heat-generating element 12. Also, for example, if the load on the server 11 increases and the computing power and cooling power increase, or if the weather forecast is wrong and renewable energy power is insufficient, the refrigerant R may not be sufficiently heated with the available power (the sum of contracted power and renewable energy power), resulting in insufficient cooling capacity.

[0048] Therefore, in this embodiment, a storage system 40 with higher thermal insulation than the main cooling system 30 is provided. The storage system 40 has a tank 42 capable of storing the refrigerant R cooled by the cooling heat source 32. The cooling system control unit 64 stores the refrigerant R cooled by the cooling heat source 32 in the tank 42 under conditions where sufficient cooling capacity can be ensured with the available power. Conditions under which sufficient cooling capacity can be ensured with the available power include, for example, when the load on the server 11 is smaller than expected or when the renewable energy power is greater than expected. In such cases, the cooling system control unit 64 opens the supply valve 44 of the storage system 40 while operating the main cooling system 30, and stores the refrigerant R cooled by the cooling heat source 32 in the tank 42. At this time, the cooling system control unit 64 adjusts the flow rate of the refrigerant R flowing through the circulation channel 33 and the flow rate of the refrigerant R flowing through the bypass channel 41 by adjusting the pumping flow rate of the main pump 34. For example, the flow rate of refrigerant R flowing through the circulation channel 33 and the flow rate of refrigerant R flowing through the bypass channel 41 are both adjusted to 10 ml / l. Note that the flow rates of the circulation channel 33 and the bypass channel 41 can be adjusted as appropriate.

[0049] Subsequently, if the cooling capacity is actually insufficient or is predicted to be insufficient with the operation of the main cooling system 30 alone, the cooling system control unit 64 opens the discharge valve 45 and drives the pump 43 of the storage system 40 to release the refrigerant R in the tank 42 to the main cooling system 30. As a result, the refrigerant R with cold energy stored in the tank 42 is supplied to the main cooling system 30, improving the cooling capacity.

[0050] The following describes an example of an operating mode for the cooling system 20 that releases the refrigerant R from the tank 42, referring to the flow chart in Figure 4. First, the usable power prediction unit 62b predicts the usable power (step S1). The prediction in step S1 is performed based on the information acquired by the acquisition unit 61. In step S1, the predicted value of usable power is calculated. When predicting usable power, the renewable energy power prediction unit 62a calculates the predicted value of renewable energy power, and the usable power prediction unit 62b calculates the usable power by adding the predicted value of renewable energy power and the contracted power. Next, the power consumption prediction unit 62c predicts the power consumption (step S2). The prediction in step S2 is performed based on the information acquired by the acquisition unit 61. In step S2, the predicted value of power consumption is calculated. Note that the order of steps S1 and S2 can be interchanged. After that, the power consumption determination unit 63a determines whether the predicted value of power consumption exceeds the predicted value of usable power (step S3). If the predicted power consumption does not exceed the predicted available power (step S3; NO), this flow terminates. If the predicted power consumption exceeds the predicted available power (step S3; YES), the cooling system control unit 64 releases the refrigerant R in the tank 42 into the circulation channel 33 (step S4). In step S4, the cooling system control unit 64 controls the pumping flow rate of the main pump 34 and pump 43. This adjusts the temperature of the refrigerant R after it has been mixed in the refrigerant mixing unit 21. By adjusting the temperature of the refrigerant R in the refrigerant mixing unit 21, the temperature of the refrigerant R sent to the cooling unit 31 is adjusted to a temperature suitable for cooling the heat-generating element 12. After step S4 is executed, this flow terminates. This flow may be executed continuously or at regular intervals.

[0051] Furthermore, if the refrigerant R in the tank 42 is allowed to flow as is, it will damage the heat-generating element 12 that is being cooled. Therefore, it is desirable to thoroughly mix the refrigerant R released from the tank 42 with the refrigerant R flowing through the circulation channel 33 in the refrigerant mixing section 21 so that the temperature reaches an appropriate level (for example, 10 to 20 degrees Celsius). However, it is also possible to supply refrigerant R to the cooling section 31 within a temperature range that minimizes damage to the heat-generating element 12, even if it is temporarily outside the appropriate temperature range (for example, 0 to 10 degrees Celsius).

[0052] The timing and flow rate of storing refrigerant R in tank 42 and releasing refrigerant R from tank 42 are controlled according to the situation. For example, refrigerant R may be stored in tank 42 while the release of refrigerant R from tank 42 is stopped, or refrigerant R may be released from tank 42 while storing refrigerant R in tank 42, or refrigerant R may be released from tank 42 while refrigerant R is stored in tank 42. Also, for example, all of the refrigerant R cooled by the cooling heat source unit 32 may be temporarily stored in tank 42.

[0053] (Effects and Benefits) The data center 10 and information processing system 1 of this embodiment can achieve the following effects and advantages.

[0054] In this embodiment, the data center 10 is equipped with a tank 42 capable of storing the refrigerant R cooled by the cooling heat source unit 32. When it is determined that the predicted power consumption, which is the sum of the power consumed by the server 11 (computation processing power) and the power consumed by the cooling heat source unit 32 (cooling processing power), exceeds the predicted value of the available power, including the contracted power, the control device 60 controls the cooling system 20 so that the refrigerant R in the tank 42 can be released into the circulation channel 33.

[0055] According to the above configuration, for example, if sufficient cooling capacity is ensured with power consumption within the available power, the cooling system 20 can store refrigerant R in the tank 42. Subsequently, when the available power becomes less than expected, the cooling system 20 can lower the temperature of the refrigerant R supplied to the cooling unit 31 by releasing the low-temperature refrigerant R stored in the tank 42 into the circulation channel 33. Therefore, the cooling system 20 can maintain its cooling capacity while suppressing the power consumed by the cooling heat source unit 32. Thus, according to this embodiment, sufficient cooling capacity can be ensured regardless of the conditions, and cooling performance can be improved compared to conventional systems. In addition, since purchased power can be kept within the contracted power, there is no need to pay high electricity rates. Furthermore, the cooling system 20 can adjust the total power consumption to the minimum by adjusting the temperature of the refrigerant R in the tank 42 and the temperature of the refrigerant R discharged from the cooling heat source unit 32. Thus, according to this embodiment, electricity costs can be reduced significantly, leading to energy conservation and decarbonization.

[0056] In this embodiment, a renewable energy generator 51 is further provided. The renewable energy generator 51 generates electricity using renewable energy and is capable of supplying renewable energy power to the cooling unit 31 and the cooling heat source unit 32. The usable power is the sum of the contracted power and the renewable energy power.

[0057] This allows usable power to be increased by the amount of renewable energy power. In addition, a portion of the power consumption can be covered by renewable energy. Therefore, the amount of purchased electricity can be reduced. Furthermore, even if, for example, the weather forecast is wrong and the expected amount of renewable energy power cannot be obtained, resulting in less usable power than expected, the cooling system 20 can maintain its cooling performance while keeping power consumption within the usable power limit by releasing the refrigerant R in the tank 42.

[0058] In this embodiment, the cooling system 20 also has a refrigerant mixing unit 21 provided in the circulation channel 33. The refrigerant mixing unit 21 mixes the refrigerant R flowing through the circulation channel 33 with the refrigerant R released from inside the tank 42.

[0059] This ensures that the low-temperature refrigerant R in the tank 42 is not supplied to the heat-generating element 12 at a temperature lower than the appropriate temperature range. Therefore, damage to the heat-generating element 12, which is being cooled, can be reduced.

[0060] The information processing system 1 comprises multiple data centers 10 located at geographically separated locations, and a migration control unit 2. The migration control unit 2 migrates jobs between the multiple data centers 10 according to the available power at each data center 10.

[0061] Incidentally, if power consumption in one data center 10 is expected to exceed the available power, one method is to migrate jobs from that data center 10 to another data center 10 to prevent power consumption from exceeding the available power. However, this method cannot be executed if the other data center 10 also has insufficient available power. Furthermore, migrating jobs causes delays in the processing time of the entire information processing system 1, so there is a desire to minimize the number of job migrations as much as possible.

[0062] In this embodiment, each data center 10 is equipped with the tank 42 and control device 60 described above. Therefore, for example, if power consumption in one data center 10 is expected to exceed the available power, the low-temperature refrigerant R stored in the tank 42 can be released to reduce the cooling power required for cooling the server 11, thereby keeping power consumption within the available power. Thus, the information processing system 1 can execute jobs initially assigned to one data center 10 while ensuring the necessary cooling capacity for that data center 10, without migrating jobs from one data center 10 with insufficient available power to another data center 10 with insufficient available power. Furthermore, since it is possible to suppress the purchase of power exceeding the contracted power in each data center 10, the information processing system 1 as a whole can reduce the purchase of expensive power and lower its electricity costs.

[0063] (First variation) Next, the first modified example will be described with reference to Figures 5 to 8. For configurations similar to those in the embodiments described above, the same names and reference numerals will be used, and explanations will be omitted as appropriate.

[0064] In this modified example, as shown in Figure 5, the determination unit 63 of the control device 60 further includes a renewable energy power determination unit 63b.

[0065] (Renewable energy power determination unit) The renewable energy power determination unit 63b determines whether the predicted value of renewable energy power exceeds the surplus determination value. The surplus determination value is the value obtained by subtracting the purchased power within the contracted power range from the predicted value of power consumption.

[0066] (Cooling System Control Unit) In this modified example, the cooling system control unit 64 controls the cooling system 20 to increase the power consumed by the cooling heat source unit 32 so that refrigerant R can be stored in the tank 42, for example, when it is determined that the predicted value of renewable energy power exceeds the surplus determination value.

[0067] In this modified example, as shown in Figure 6, the migration control unit 2 has the following functional units: an acquisition unit 70, a determination unit 71, and a migration execution unit 72.

[0068] (Acquisition Department) The acquisition unit 70 acquires information necessary to perform migration between multiple data centers 10. The acquisition unit 70 includes a power forecast value acquisition unit 70a and a storage amount acquisition unit 70b.

[0069] (Power forecast value acquisition unit) The power prediction unit 70a acquires predicted power consumption and predicted available power from each data center 10.

[0070] (Storage volume acquisition unit) The storage volume acquisition unit 70b acquires the storage volume in the tank 42 from each data center 10.

[0071] (Judgment Department) The determination unit 63 includes a power shortage data center determination unit 71a and a cooling / heating surplus data center determination unit 71b.

[0072] (Power shortage data center detection unit) The power shortage data center determination unit 71a determines whether or not there is a power shortage data center among the multiple data centers 10 that constitute the information processing system 1. A power shortage data center is a data center 10 in which the available power is insufficient relative to the power consumption.

[0073] (Data center surplus cooling / heating determination unit) The cooling / heating surplus data center determination unit 71b determines whether or not there is a cooling / heating surplus data center among the multiple data centers 10 that constitute the information processing system 1. A cooling / heating surplus data center is a data center 10 in which a predetermined amount or more of refrigerant R is stored in the tank 42. Here, the state in which a predetermined amount or more of refrigerant R is stored in the tank 42 means a state in which the amount of refrigerant R stored in the tank 42 is such that the amount of purchased power can be reliably kept within the range of the contracted power by reducing the purchased power by releasing the refrigerant from the tank 42. The value of this "predetermined amount," which is the threshold for determining whether or not a data center is cooling / heating surplus, can be set appropriately for each data center 10.

[0074] (Migration Execution Team) The migration execution unit 72 enables the migration of jobs between multiple data centers 10 based on the information acquired by the acquisition unit 70. For example, based on the determination results of the power shortage data center determination unit 71a and the cooling / heating surplus data center determination unit 71b, the migration execution unit 72 migrates jobs from a data center 10 where the available power is insufficient relative to the power consumption (power shortage data center) to a data center 10 where a predetermined amount or more of refrigerant R is stored in the tank 42 (cooling / heating surplus data center).

[0075] (Operation of the cooling system) Next, the operation of the cooling system 20 according to this modified example will be described. The operation, which is the same as that of the embodiment described above, will not be explained.

[0076] The following describes an example of an operating mode for a cooling system 20 that stores refrigerant R in a tank 42 according to this modified example, with reference to the flow chart in Figure 7. First, the renewable energy power prediction unit 62a predicts the renewable energy power (step S11). The prediction in step S11 is performed based on the information acquired by the acquisition unit 61. In step S11, the predicted value of renewable energy power is calculated. Next, the power consumption prediction unit 62c predicts the power consumption (step S12). The prediction in step S12 is performed based on the information acquired by the acquisition unit 61. In step S12, the predicted value of power consumption is calculated. Note that the order of steps S11 and S12 can be interchanged. After that, the renewable energy power determination unit 63b determines whether the predicted value of renewable energy power exceeds the surplus determination value (step S13). If the predicted value of renewable energy power does not exceed the surplus determination value (step S13; NO), this flow ends. If the predicted value of renewable energy power exceeds the surplus threshold (step S13; YES), the cooling system control unit 64 increases the power consumed by the cooling heat source unit 32 (cooling processing power) to store refrigerant R in the tank 42 (step S14). After step S14 is completed, this flow is terminated. This flow may run continuously or at regular intervals.

[0077] In this way, by storing refrigerant R in the tank 42, it becomes easier for refrigerant R to be stored in the tank 42 at each data center 10. Then, by migrating high-load jobs or many jobs to a data center 10 equipped with a tank 42 that has a sufficient amount of refrigerant R stored, jobs can be processed efficiently. Below, an example of the procedure for performing such an efficient migration will be explained with reference to the flow in Figure 8.

[0078] First, the power prediction acquisition unit 70a acquires predicted power consumption and predicted usable power from each data center 10 (step S21). Next, the storage amount acquisition unit 70b acquires the amount stored in the tank 42 from each data center 10 (step S22). Note that the order of steps S21 and S22 can be interchanged. After that, the cooling surplus data center determination unit 71b determines whether or not there is a cooling surplus data center among the multiple data centers 10 that constitute the information processing system 1, where a predetermined amount or more of refrigerant R is stored in the tank 42 (step S23). In this modified example, the "predetermined amount" which serves as the threshold for determining whether or not a data center is a cooling surplus data center is the maximum storage amount in the tank 42. In other words, in step S23, the cooling surplus data center determination unit 71b determines whether there is a data center 10 among the multiple data centers 10 that has a tank 42 that is full of refrigerant R, and designates the data center 10 that has a tank 42 that is full of refrigerant R as a cooling surplus data center.

[0079] If there are no data centers with surplus cooling / heating capacity (Step S23; NO), this flow terminates. If there are data centers with surplus cooling / heating capacity (Step S23; YES), the power shortage data center determination unit 71a determines whether there are any power shortage data centers among the multiple data centers 10 that constitute the information processing system 1, where the available power is insufficient relative to the power consumption (Step S24). If there are no power shortage data centers (Step S24; NO), this flow terminates. If there are power shortage data centers (Step S24; YES), the migration execution unit 72 migrates jobs from power shortage data centers to data centers with surplus cooling / heating capacity (Step S25). Note that the order of Step S23 and Step S24 can be interchanged.

[0080] After step S25 is executed, this flow terminates. This flow may run continuously or at regular intervals.

[0081] (Effects and Benefits) According to this modified version, the following effects can be achieved.

[0082] In this modified example, if it is determined that the surplus value obtained by subtracting the purchased power is exceeded, the control device 60 controls the cooling system 20 to increase the power used by the cooling heat source unit 32 (cooling processing power) so that refrigerant R can be stored in the tank 42.

[0083] This allows the data center 10 to actively store cooling energy in the tank 42 by increasing the power consumed by the cooling heat source unit 32 when there is a surplus of renewable energy, even if purchased electricity is reduced. Therefore, the data center 10 can utilize surplus electricity without waste.

[0084] In this modified example, the migration control unit 2 is capable of migrating jobs from a data center 10 where the available power is insufficient relative to the power consumption (power-deficient data center) to a data center 10 where a predetermined amount or more of refrigerant R is stored in the tank 42 (refrigeration surplus data center).

[0085] As a result, excess refrigerant R is stored in the tank 42, allowing jobs from power-deficient and power-deficient data centers to be transferred to data centers with surplus cooling / heating. At this time, the data center 10 to which the jobs are transferred will experience an increase in power consumption (computation processing power) by the server 11, but by releasing the excess refrigerant R stored in the tank 42, the power consumed by the cooling heat source 32 (cooling processing power) is reduced, keeping power consumption within the available power. Furthermore, in this embodiment, when it is determined that there is a surplus of renewable energy power, the power consumed by the cooling heat source 32 is increased to actively store refrigerant R in the tank 42. Therefore, by migrating jobs to data center 10 where cooling / heating is actively stored in this way, the cooling / heating stored by the surplus renewable energy can be actively utilized. Consequently, the entire information processing system 1 can utilize the surplus renewable energy power without waste.

[0086] Next, another example of the data center 10 according to the first modification will be described with reference to Figure 9. As shown in Figure 9, the bypass flow path 41 may have a supply-side bypass flow path 41a, a branch flow path 41b, and a discharge-side bypass flow path 41c. The supply-side bypass flow path 41a guides the refrigerant R cooled by the cooling heat source 32 to the tank 42. Multiple branch flow paths 41b are provided in parallel so as to branch off from the supply-side bypass flow path 41a. Each branch flow path 41b is provided with one tank 42. The multiple tanks 42 are connected to each other in parallel. The discharge-side bypass flow path 41c guides the refrigerant R stored in each tank 42 to the circulation flow path 33. In the illustrated example, there are three tanks 42, but this is not limited to this. The number of tanks 42 can be changed as appropriate.

[0087] Furthermore, each branch channel 41b is provided with a branch supply valve 46 and a branch discharge valve 47. The branch supply valve 46 is located upstream of the tank 42 in the branch channel 41b, and the branch discharge valve 47 is located downstream of the tank 42 in the branch channel 41b. Both the branch supply valve 46 and the branch discharge valve 47 are provided to open and close the branch channel 41b.

[0088] In this example as well, the refrigerant R is stored in the tanks 42 in the same steps as described in steps S11 to S14 above. However, there are differences in the following respects. Specifically, if the predicted value of renewable energy power is determined to exceed the surplus determination value (step S13; YES), the cooling system control unit 64 increases the power consumed by the cooling heat source unit 32 and controls the cooling system 20 so that refrigerant R can be stored in each tank 42 sequentially (step S14). In step S13, for example, the cooling surplus data center determination unit 71b determines whether there is a data center 10 among the multiple data centers 10 in which all tanks 42 have been filled with refrigerant R, and designates the data center 10 in which all tanks 42 have been filled with refrigerant R as a cooling surplus data center.

[0089] In this way, the cooling system 20 can store more cold energy. In addition, since the cooling system 20 can store and release refrigerant R in each tank 42, efficient management of refrigerant R becomes possible.

[0090] (Second variation) Next, a second modified example will be described with reference to Figures 10 to 12. For configurations similar to those in the embodiments described above, the same names and reference numerals will be used, and explanations will be omitted as appropriate.

[0091] In this modified example, as shown in Figure 10, the determination unit 63 of the control device 60 further comprises a time zone determination unit 63c and a probability determination unit 63d.

[0092] (Time zone determination unit) The time zone determination unit 63c determines whether or not there are time periods in which high electricity rates are not incurred. Hereinafter, this time period will be referred to as the low-cost time period. Examples of low-cost time periods include time periods in which the amount of electricity purchased per unit time does not exceed the contracted power per unit time, and nighttime electricity periods in which electricity rates are cheaper than daytime rates.

[0093] (Probability determination unit) The probability determination unit 63d determines whether the probability of power consumption exceeding usable power is lower than a predetermined threshold. The predetermined threshold can be set appropriately for each data center 10.

[0094] (Cooling System Control Unit) In this modified example, the cooling system control unit 64 controls the cooling system 20 to increase the power consumed by the cooling heat source unit 32 during low-cost periods, for example, to enable the storage of refrigerant R in the tank 42. Furthermore, the cooling system control unit 64 controls the cooling system 20 to release the refrigerant R in the tank 42 when, for example, the probability of power consumption exceeding available power is lower than a predetermined threshold.

[0095] (Operation of the cooling system) Next, the operation of the cooling system 20 according to this modified example will be described. The operation, which is the same as that of the embodiment described above, will not be explained.

[0096] The following describes an example of an operating mode for a cooling system 20 that stores refrigerant R in a tank 42 according to this modified example, with reference to the flow chart in Figure 11. First, the time zone determination unit 63c determines whether or not there is a low-cost time zone in which high electricity charges are not incurred (step S31). If there is no low-cost time zone (step S31; NO), this flow ends. If there is a low-cost time zone (step S31; YES), the cooling system control unit 64 increases the power consumed by the cooling heat source unit 32 (cooling processing power) to store refrigerant R in the tank 42 (step S32). After step S32 is executed, this flow ends. This flow may be executed continuously or at regular intervals.

[0097] Next, an example of an operating mode for the cooling system 20 that releases refrigerant R from tank 42 will be described with reference to the flow in Figure 12. First, the probability determination unit 63d determines, based on the information acquired by the acquisition unit 61, whether the probability that the power consumption exceeds the available power is lower than a predetermined threshold (step S41). If the probability that the power consumption exceeds the available power is higher than the predetermined threshold (step S41; NO), this flow ends. If the probability that the power consumption exceeds the available power is lower than the predetermined threshold (step S41; YES), the cooling system control unit 64 releases refrigerant R from tank 42 into the circulation channel 33 (step S42). In step S42, the cooling system control unit 64 controls the amount of refrigerant R released so that some of the refrigerant R remains in tank 42. For example, the cooling system control unit 64 releases refrigerant R from tank 42 until the temperature of tank 42 does not exceed 0 degrees. After step S42 is executed, this flow ends. This flow may be executed continuously or at regular intervals.

[0098] (Effects and Benefits) According to this modified version, the following effects can be achieved.

[0099] In this modified example, the control device 60 controls the cooling system 20 so that the power consumed by the cooling heat source unit 32 is increased during low-cost periods when high electricity rates are not incurred, thereby enabling the storage of refrigerant R in the tank 42.

[0100] As a result, the cooling system 20 can efficiently store low-temperature refrigerant R in the tank 42. For example, even when renewable energy power is not surplus and there is a high probability that purchased power will exceed the contracted power, low-temperature refrigerant R can be actively stored in the tank 42 during specific time periods when purchased power per unit time does not exceed the contracted power per unit time. Alternatively, for example, refrigerant R cooled using inexpensive nighttime electricity can be stored in the tank 42. This allows the refrigerant R flowing through the circulation channel 33 to be cooled using the chilled water in the tank 42, even if no surplus power is expected the following day.

[0101] If the probability that the power consumption exceeds the available power is lower than a predetermined threshold, the control device 60 controls the cooling system 20 to allow the release of the refrigerant R in the tank 42.

[0102] This allows the refrigerant R stored in the tank 42 to be actively used to cool the refrigerant R in the circulation channel 33. As a result, the power consumed by the cooling heat source unit 32 is reduced, and electricity costs can be further reduced.

[0103] (Other embodiments) Although embodiments of this disclosure have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments and may include design changes and the like that do not depart from the gist of this disclosure.

[0104] In the above embodiment, the downstream end of the bypass channel was assumed to be the downstream portion of the high-temperature circulation channel 33b and connected near the cooling heat source 32, but this is not limited to this. For example, the downstream end of the bypass channel may be provided downstream of the cooling section 31 in the low-temperature circulation channel 33a.

[0105] In the above embodiment, the case in which the data center 10 is equipped with a renewable energy generator 51 was described, but the data center is not limited to this. The data center 10 does not need to be equipped with a renewable energy generator 51. In this case, the available power becomes the contracted power, and the contracted power becomes the predicted value of the available power calculated by the available power prediction unit 62b.

[0106] <Hardware Configuration> The control device 60 and migration control unit 2 of the above embodiments and modified examples are implemented in the computer 1100 shown in Figure 13. Figure 13 is a schematic block diagram showing the configuration of the computer 1100 according to each embodiment. The computer 1100 includes a processor 1110, main memory 1120, storage 1130, and interface 1140.

[0107] The operations of each of the above-mentioned functional units of the control device 60 and the migration control unit 2 are stored in the storage 1130 in the form of a program. The processor 1110 reads the program from the storage 1130, loads it into the main memory 1120, and executes the above-mentioned processes according to the program. The processor 1110 also allocates storage space in the main memory 1120 according to the program.

[0108] The program may be for the purpose of realizing some of the functions that the computer 1100 is to perform. For example, the program may perform functions in combination with other programs already stored in the storage 1130, or in combination with other programs implemented in other devices. In addition, the computer 1100 may be equipped with a custom LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device) in addition to or instead of the above configuration. Examples of PLDs include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array). In this case, some or all of the functions realized by the processor 1110 may be realized by the integrated circuit.

[0109] Examples of storage 1130 include magnetic disks, magneto-optical disks, and semiconductor memory. Storage 1130 may be an internal medium directly connected to the bus of computer 1100, or an external medium connected to computer 1100 via interface 1140 or a communication line. Furthermore, if this program is distributed to computer 1100 via a communication line, computer 1100 that receives the program may expand it into main memory 1120 and execute the above processing. Storage 1130 may also be a tangible storage medium that is not temporary.

[0110] Furthermore, the program may be intended to implement some of the functions described above. In addition, the program may be a so-called differential file (differential program) that implements the functions described above in combination with other programs already stored in the storage 1130.

[0111] <Note> The data center 10 and information processing system 1 described in each embodiment can be understood, for example, as follows.

[0112] (1) The data center 10 according to the first embodiment comprises a server 11 having a heat-generating element 12, a cooling system 20 for cooling the heat-generating element 12, and a control device 60 for controlling the cooling system 20, wherein the cooling system 20 comprises a cooling unit 31 for cooling the heat-generating element 12 with a refrigerant R, a cooling heat source unit 32 for cooling the refrigerant R, a circulation channel 33 for circulating the refrigerant R between the cooling unit 31 and the cooling heat source unit 32, and the refrigerant R that has passed through the cooling heat source unit 32 in the circulation channel 33 for the cooling unit 31 in the circulation channel 33 The cooling system 20 includes a bypass channel 41 that bypasses the circulation channel 33 so that the refrigerant R cooled by the cooling heat source 32 is introduced into the upstream portion, and a tank 42 provided in the bypass channel 41 that can store the refrigerant R cooled by the cooling heat source 32. The control device 60 controls the cooling system 20 so that the refrigerant R in the tank 42 can be released into the circulation channel 33 if it determines that the predicted value of power consumption, which is the sum of the power consumed by the server 11 and the power consumed by the cooling heat source 32, exceeds the predicted value of available power, including the contracted power.

[0113] According to the above configuration, for example, if sufficient cooling capacity is ensured with power consumption within the available power, the cooling system 20 can store refrigerant R in the tank 42. Subsequently, when the available power becomes less than expected, the cooling system 20 can lower the temperature of the refrigerant R supplied to the cooling unit 31 by releasing the low-temperature refrigerant R stored in the tank 42 into the circulation channel 33. Therefore, the cooling system 20 can maintain its cooling capacity while suppressing the power consumed by the cooling heat source unit 32. Accordingly, according to this embodiment, sufficient cooling capacity can be ensured regardless of the conditions, and thus cooling performance can be improved compared to the conventional method.

[0114] (2) The data center 10 in the second embodiment is the data center 10 of (1), further comprising a renewable energy generator 51 that generates electricity using renewable energy and can supply renewable energy power to the cooling unit 31 and the cooling heat source unit 32, wherein the usable power may be the sum of the contracted power and the renewable energy power.

[0115] This allows usable power to be increased by the amount of renewable energy power. In addition, a portion of the power consumption can be covered by renewable energy. Furthermore, even if, for example, the weather forecast is wrong and the expected amount of renewable energy power cannot be obtained, resulting in less usable power than expected, the cooling system 20 can maintain its cooling performance while keeping power consumption within the usable power limit by releasing the refrigerant R in the tank 42.

[0116] (3) The data center 10 in the third embodiment is the data center 10 of (2), wherein the control device 60 may control the cooling system 20 to increase the power consumed by the cooling heat source 32 so that the refrigerant R can be stored in the tank 42 if it is determined that the predicted value of the renewable energy power exceeds a surplus determination value obtained by subtracting the purchased power within the range of the contracted power from the predicted value of the power consumption.

[0117] This allows the data center 10 to actively store cooling energy in the tank 42 by increasing the power consumed by the cooling heat source unit 32 when there is a surplus of renewable energy, even if purchased electricity is reduced. Therefore, the data center 10 can utilize surplus electricity without waste.

[0118] (4) The data center 10 in the fourth embodiment is the data center 10 of (3), wherein a plurality of tanks 42 are provided, and the control device 60 may control the cooling system 20 such that, when it is determined that the predicted value of the renewable energy power exceeds the surplus determination value, the power consumed by the cooling heat source 32 is increased so that the refrigerant R can be sequentially stored in each of the tanks 42.

[0119] This allows the cooling system 20 to store more cold energy. In addition, the cooling system 20 can store and release refrigerant R in each tank 42, enabling efficient management of refrigerant R.

[0120] (5) The data center 10 in the fifth embodiment is any one of the data centers 10 described in (1) to (4), wherein the control device 60 may control the cooling system 20 such that the power consumed by the cooling heat source 32 is increased during times when high electricity rates are not incurred, so that the refrigerant R can be stored in the tank 42.

[0121] This allows the cooling system 20 to efficiently store low-temperature refrigerant R in the tank 42.

[0122] (6) The data center 10 in the sixth embodiment is any one of the data centers 10 described in (1) to (5), wherein the control device 60 may control the cooling system 20 to release the refrigerant R in the tank 42 when the probability that the power consumption exceeds the available power is lower than a predetermined threshold.

[0123] This allows the refrigerant R stored in the tank 42 to be actively used to cool the refrigerant R in the circulation channel 33. As a result, the power consumed by the cooling heat source unit 32 is reduced, and electricity costs can be further reduced.

[0124] (7) The information processing system 1 of the seventh embodiment comprises one of the data centers 10 of (1) to (6) located in multiple geographically separated locations, and a migration control unit 2 that migrates jobs between the multiple data centers 10 according to the available power in each of the data centers 10.

[0125] In this embodiment, each data center 10 is equipped with the aforementioned tank 42 and control device 60. Therefore, for example, if power consumption in one data center 10 is expected to exceed the available power, the cooling power required for cooling the server 11 can be reduced by releasing the low-temperature refrigerant R stored in the tank 42, thereby keeping power consumption within the available power. Thus, the information processing system 1 can execute the jobs initially assigned to one data center 10 while ensuring the necessary cooling capacity for that data center 10, without migrating jobs from one data center 10 with insufficient available power to another data center 10 with insufficient available power.

[0126] (8) The information processing system 1 of the eighth embodiment is the information processing system 1 of (7), wherein the migration control unit 2 is capable of migrating jobs from the data center 10 in which the available power is insufficient relative to the power consumption to the data center 10 in which a predetermined amount or more of the refrigerant R is stored in the tank 42.

[0127] As a result, excess refrigerant R is stored in the tank 42, and jobs from data center 10 that have been determined to be experiencing a power shortage can be transferred to the data center 10 with excess cooling energy stored. At this time, although the power consumed by the server 11 in the data center 10 to which the jobs have been transferred will increase, the power consumed by the cooling heat source unit 32 can be reduced by releasing the excess refrigerant R stored in the tank 42, keeping the power consumption within the available power. [Explanation of symbols]

[0128] 1. Information Processing System 2. Migration Control Unit 3 Network 4. Power generation equipment 10 Data Centers 11 Servers 12 Heating element 13 Networks 20 Cooling System 21 Refrigerant mixing section 30 Main Cooling System 31 Cooling section 32 Cooling heat source machine 33 Circulation channels 33a Low-temperature circulation channel 33b High-temperature circulation channel 34 Main pump 35 Main valve 40 Storage Systems 41 Bypass channel 41a Supply-side bypass channel 41b Branch channel 41c Discharge side bypass channel 42 tanks 43 pumps 44 Supply valve 45. Exhaust valve 46 Branch supply valve 47 Branch discharge valve 50 Power supply systems 51 Renewable energy generators 52 Power supply lines 53 Power meter 60 Control device 61 Acquisition Department 62 Prediction Section 62a Renewable Energy Power Forecasting Section 62b Available Power Prediction Unit 62c Power Consumption Prediction Unit 63 Judgment section 63a Power consumption determination section 63b Renewable energy power determination unit 63c Time zone determination unit 63d Probability determination unit 64 Cooling System Control Unit 70 Acquisition Department 70a Power Prediction Value Acquisition Unit 70b Storage volume acquisition unit 71 Judgment section 71a Power shortage data center determination unit 71b Cooling and Heat Surplus Data Center Determination Unit 72 Migration Execution Department 1100 Computer 1110 processor 1120 Main Memory 1130 storage 1140 Interface R refrigerant

Claims

1. A server having a heat source, A cooling system for cooling the aforementioned heat-generating element, A control device for controlling the cooling system, Equipped with, The cooling system is, A cooling unit that cools the heat-generating element with a refrigerant, A cooling heat source unit for cooling the aforementioned refrigerant, A circulation channel for circulating the refrigerant between the cooling unit and the cooling heat source unit, A bypass channel is provided that bypasses the circulation channel so that the refrigerant that has passed through the cooling heat source in the circulation channel is introduced into the upstream portion of the circulation channel relative to the cooling section, A tank provided in the bypass channel and capable of storing the refrigerant cooled by the cooling heat source unit, Equipped with, If the control device determines that the predicted power consumption, which is the sum of the power consumed by the server and the power consumed by the cooling heat source, exceeds the predicted value of the available power, including the contracted power, it controls the cooling system to allow the refrigerant in the tank to be released into the circulation path. Data center.

2. The system further includes a renewable energy generator capable of generating electricity using renewable energy and supplying renewable energy power to the cooling unit and the cooling heat source unit, The available power is the sum of the contracted power and the renewable energy power. The data center according to claim 1.

3. If the control device determines that the predicted value of renewable energy power exceeds a surplus determination value obtained by subtracting the purchased power within the range of the contracted power from the predicted value of power consumption, it controls the cooling system to increase the power consumed by the cooling heat source to enable the storage of the refrigerant in the tank. The data center according to claim 2.

4. Multiple tanks are provided, If the control device determines that the predicted value of the renewable energy power exceeds the surplus determination value, it increases the power consumed by the cooling heat source unit and controls the cooling system so that the refrigerant can be sequentially stored in each of the tanks. The data center according to claim 3.

5. The control device controls the cooling system so that the power consumed by the cooling heat source unit is increased during times when high electricity rates are not incurred, thereby enabling the storage of the refrigerant in the tank. A data center according to any one of claims 1 to 4.

6. The control device controls the cooling system so as to allow the release of the refrigerant in the tank when the probability that the power consumption exceeds the available power is lower than a predetermined threshold. A data center according to any one of claims 1 to 4.

7. A data center according to any one of claims 1 to 4, which is installed in multiple geographically separated locations, A migration control unit that migrates jobs between multiple data centers according to the available power in each of the data centers, An information processing system equipped with the following features.

8. The migration control unit is capable of migrating jobs from a data center where the available power is insufficient relative to the power consumption to a data center where a predetermined amount or more of the refrigerant is stored in the tank. The information processing system according to claim 7.

Images