Apparatus, method, optimal placement determination apparatus, optimal placement determination method, and program

Abstract

An optimum arrangement determination device according to one embodiment of the present disclosure is for determining the optimum arrangement of ICT loads that can be arranged at a plurality of bases constituting a subject network, the optimum arrangement determination device having: a pre-processing unit configured to offset, for each of the bases where an electricity supply generated by renewable energy is usable, an electricity demand demanded by the corresponding ICT load arranged at the base against the electricity supply; a main processing unit configured to, by using the electricity supply and the electricity demand having been offset by the pre-processing unit, determine as an optimum arrangement, an arrangement that can minimize, throughout the entirety of the plurality of bases, a cost required for purchasing the amount of electricity that is, of an electricity demand demanded by a ICT load when the ICT load is arranged in a base, an amount exceeding the electricity supply usable in the base; a post-processing unit configured to, by using the electricity supply and the electricity demand having been offset by the pre-processing unit, offset the electricity demand of the ICT load after the optimum arrangement determined by the main processing unit against the electricity supply usable in the base where the ICT load is arranged; and an output unit configured to, when all the electricity demands and the electricity supply can be offset by the post-processing unit and the electricity supply after the offset is greater than 0, output the electricity supply after the offset as amount of selling electricity.

Description

Optimal placement determination device, optimal placement determination method, and program

[0001] The present disclosure relates to an optimal placement determination device, an optimal placement determination method, and a program.

[0002] In recent years, there has been a global movement to introduce renewable energy with the aim of reducing environmental impact, etc. Accordingly, the use of renewable energy in data centers has also been promoted. However, renewable energy has large fluctuations in power generation output due to weather changes, etc., which can result in surplus or shortage of renewable energy power supply. In response to this, Non-Patent Document 1 discloses a method for effectively utilizing renewable energy power supply throughout a data center network by rearranging ICT (Information and Communication Technology) loads within a data center network where renewable energy power generation facilities are installed.

[0003] Ryota Nakamura, Kunaki Harada, "Proposal of an ICT Load Allocation Method Considering Renewable Energy," IEICE General Conference, B-14-4, March 2022.

[0004] However, the method described in Non-Patent Document 1 does not take into consideration the purchase of power shortages or the sale of surplus power after rearranging ICT loads. As a result, for example, even if the power shortage in the entire data center network is reduced by rearranging the ICT loads, the cost required to purchase the power shortage may increase.

[0005] The present disclosure has been made in consideration of the above points, and aims to provide a technology for determining an optimal load allocation of renewable energy power, taking into consideration the buying and selling of surplus or shortage power.

[0006] An optimal placement determination device according to one aspect of the present disclosure is an optimal placement determination device that determines the optimal placement of ICT loads that can be placed at multiple bases that make up a target network, and includes a pre-processing unit that is configured, for each base that can use a power supply generated by renewable energy, to offset the power demand of the ICT load placed at the base with the power supply amount; and, using the power supply amount and power demand amount after offset by the pre-processing unit, calculates the cost required to purchase the amount of power that exceeds the power supply amount available at the base out of the power demand required by the ICT load when the ICT load is placed at the base. a post-processing unit configured to offset the power demand of the ICT load after the optimal arrangement determined by the main processing unit with the power supply amount available at the base where the ICT load is located, using the power supply amount and power demand amount after the offset by the pre-processing unit; and an output unit configured to output the power supply amount after the offset as the amount of power sold if all the power demand amounts can be offset with the power supply amount by the post-processing unit and the power supply amount after the offset is greater than 0.

[0007] A technology is provided for determining the optimal load allocation of renewable energy power, taking into consideration the buying and selling of surplus or shortage power.

[0008] FIG. 1 is a diagram illustrating an example of the hardware configuration of an optimal placement determination device according to the present embodiment; FIG. 2 is a diagram illustrating an example of the functional configuration of an optimal placement determination device according to the present embodiment; FIG. 3 is a flowchart illustrating an example of an optimal placement determination process according to the present embodiment; FIG. 4 is a diagram illustrating an example of pre-processing (part 1); FIG. 5 is a diagram illustrating an example of pre-processing (part 2); FIG. 6 is a diagram illustrating an example of main processing; FIG. 7 is a diagram illustrating an example of the influence of power purchase costs on main processing; and FIG. 8 is a diagram illustrating an example of post-processing.

[0009] An embodiment of the present invention will now be described, which is an optimal allocation determination device 10 that can determine the optimal allocation of ICT loads across an entire data center network in which renewable energy power generation facilities are installed, taking into consideration the sale of surplus power and the purchase of power shortages.

[0010] Here, a data center network is a network consisting of a data center (DC) where ICT loads can be placed, nodes that transfer communications, and links that interconnect them. Each node has multiple user terminals that use the services provided by the data center network. The data center network is assumed to be a wide-area network, such as a nationwide core network.

[0011] In the following, as an example, the ICT load is a virtual machine (VM), and at least one data center in the data center network is assumed to have solar panels installed as renewable energy power generation equipment. However, the ICT load is not limited to virtual machines, and any ICT load that can be installed on a data center and whose placement can be changed or moved can be used. Furthermore, the renewable energy power generation equipment is not limited to solar panels, and may be other renewable energy power generation equipment (e.g., wind power generation equipment, hydroelectric power generation equipment, geothermal power generation equipment, biomass power generation equipment, etc.).

[0012] Furthermore, a data center is just one example, and the present invention is not necessarily limited to a data center, and any facility can be used as long as it is a base where ICT loads can be placed.

[0013] <Preparation of Symbols> Several symbols to be used in this embodiment are prepared.

[0014] Assuming that there are multiple data centers in a data center network, let A be the set of data centers j that have power demand. Let V(j) be the set of virtual machines in data center j (more precisely, virtual machines deployed on physical servers in data center j). Let B be the set of data centers j to which solar cells supply power.

[0015] The power demand of the virtual machine, the power supply amount of the solar cell, and the power purchase cost per unit of power of the data center j are respectively expressed as follows:

[0016] a ｊｋ : Power demand of virtual machine k∈V(j) in data center j∈A b ｊ : Amount of power supplied by solar cells at data center j∈B c ｊ : Power purchase cost per unit of power of data center j∈A Note that the data center where the virtual machine is placed can be dynamically changed at predetermined time intervals ΔT (for example, ΔT = 30 minutes or 1 hour). Therefore, if a certain time is t, more accurately, A = A(t) and V(j) = V(t;j). Similarly, the amount of power demand, the amount of power supply, and the cost of purchasing power can also change depending on the time t, so more accurately, a ｊｋ = a ｊｋ (t), b ｊ = b ｊ (t), c ｊ = c ｊ (t). Although renewable energy power generation facilities are not installed (or removed) frequently, if new installations or removals of renewable energy power generation facilities are taken into consideration, the data center to which power is supplied by solar cells may also change depending on the time, so B can also be expressed as B = B(t). However, in the following, a certain specific time will be considered as fixed, and time t will not be specified unless misunderstandings arise.

[0017] <Example of Hardware Configuration of Optimal Allocation Determination Apparatus 10> An example of the hardware configuration of the optimal allocation determination apparatus 10 according to this embodiment is shown in Fig. 1. As shown in Fig. 1, the optimal allocation determination apparatus 10 according to this embodiment includes an input device 101, a display device 102, an external I / F 103, a communication I / F 104, a RAM (Random Access Memory) 105, a ROM (Read Only Memory) 106, an auxiliary storage device 107, and a processor 108. Each of these pieces of hardware is connected to each other via a bus 109 so as to be able to communicate with each other.

[0018] The input device 101 is, for example, a keyboard, a mouse, a touch panel, a physical button, etc. The display device 102 is, for example, a display, a display panel, etc. Note that the optimal layout determination device 10 does not necessarily have to include at least one of the input device 101 and the display device 102, for example.

[0019] The external I / F 103 is an interface with an external device such as a recording medium 103a. The optimal allocation determination apparatus 10 can read from and write to the recording medium 103a via the external I / F 103. Examples of the recording medium 103a include a flexible disk, a CD (Compact Disc), a DVD (Digital Versatile Disk), an SD memory card (Secure Digital memory card), and a USB (Universal Serial Bus) memory card.

[0020] The communication I / F 104 is an interface through which the optimal placement determination apparatus 10 communicates with other devices and equipment. The RAM 105 is a volatile semiconductor memory (storage device) that temporarily stores programs and data. The ROM 106 is a non-volatile semiconductor memory (storage device) that can store programs and data even when the power is turned off. The auxiliary storage device 107 is a storage device (storage device) such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a flash memory. The processor 108 is an arithmetic device such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit).

[0021] The optimal placement determination device 10 according to this embodiment has the hardware configuration shown in Fig. 1 and is therefore capable of implementing the optimal placement determination process described below. Note that the hardware configuration shown in Fig. 1 is merely an example, and the hardware configuration of the optimal placement determination device 10 is not limited to this. For example, the optimal placement determination device 10 may have multiple auxiliary storage devices 107 or multiple processors 108, may not have some of the hardware shown in the figure, or may have various hardware components other than the hardware shown in the figure.

[0022] <Example of Functional Configuration of Optimal Allocation Determination Apparatus 10> An example of the functional configuration of the optimal allocation determination apparatus 10 according to this embodiment is shown in Fig. 2. As shown in Fig. 2, the optimal allocation determination apparatus 10 according to this embodiment has an input unit 201, a pre-processing unit 202, a main processing unit 203, a post-processing unit 204, and an output unit 205. Each of these units is realized, for example, by processing in which one or more programs installed in the optimal allocation determination apparatus 10 are executed by the processor 108.

[0023] The input unit 201 receives a set A of data centers with power demand, a set V(j) of virtual machines in each data center j∈A, and a power demand a of each virtual machine k∈V(j) in each data center j∈A. ｊｋ , the amount of power supplied by the solar cells of each data center j b ｊ , and the power purchase cost per unit of power of each data center j cｊ The input unit 201 inputs information such as the above. Generally, this information is collected and managed by an EMS (Energy Management System) or NMS (Network Management System) that manages the data center network. Therefore, the input unit 201 may acquire and input this information from the EMS, NMS, or the like.

[0024] The preprocessing unit 202 calculates the power supply amount b of the solar cell in each data center j∈A∩B. ｊ to the power demand a of the virtual machine k ∈ V(j) ｊｋ Subtract the power supply amount b ｊ and electricity demand a ｊｋ The amount of power supply b ｊ and electricity demand a ｊｋ The cancellation of these two is called power cancellation.

[0025] The main processing unit 203 calculates the power demand amount a after the power offset. ｊｋ and power supply amount b ｊ Using this, the power supply amount b between the data centers is calculated. ｊ and electricity demand a ｊｋ The main processing unit 203 then determines the placement of the virtual machines so as to offset the above as much as possible. At this time, the main processing unit 203 reduces the problem to a multiple knapsack problem (see, for example, Reference 1) and calculates the optimal placement of the virtual machines (in other words, the optimal destinations for the virtual machines) as the solution. Hereinafter, the "optimal placement of virtual machines" obtained as the solution to the multiple knapsack problem will be referred to as the "optimal VM placement."

[0026] The post-processing unit 204 calculates the power supply amount b ｊ and electricity demand a ｊｋ If there is a power surplus or a power shortage, the amount of power surplus or power shortage is calculated. As a result of the power offset, if there is a power supply remaining, the amount of power surplus is obtained, and if there is a power demand remaining, the amount of power shortage is obtained.

[0027] The output unit 205 outputs the optimal VM placement to a predetermined output destination. If a surplus power amount is obtained, the output unit 205 outputs the surplus power amount as the amount of power to be sold, and if a shortage power amount is obtained, the output unit 205 outputs the surplus power amount as the amount of power to be purchased, to a predetermined output destination. Here, an example of an output destination of the optimal VM placement is a virtual machine control device that controls the placement of virtual machines. Furthermore, an example of an output destination of the amount of power to be sold and the amount of power to be purchased is a power control device that controls the sale and purchase of power.

[0028] <Optimal Allocation Determination Process> The optimal allocation determination process according to this embodiment will be described below with reference to Fig. 3. Here, the following steps S101 to S105 are repeatedly executed, for example, at predetermined time intervals ΔT (for example, ΔT = 30 minutes or 1 hour). Steps S101 to S105 at a certain time will be described below.

[0029] The input unit 201 receives a set A of data centers with power demand, a set V(j) of virtual machines in each data center j∈A, and a power demand a of each virtual machine k∈V(j) in each data center j∈A. ｊｋ , the amount of power supplied by the solar cells of each data center j b ｊ , and the power purchase cost per unit of power of each data center j c ｊ The above information is input (step S101).

[0030] Next, the preprocessing unit 202 calculates the power supply amount b of the solar cell in each data center j∈A∩B. ｊ to the power demand a of the virtual machine k ∈ V(j) ｊｋ Subtract the power supply amount b ｊ and electricity demand a ｊｋ In other words, the preprocessing unit 202 performs power cancellation for each data center j∈A∩B according to the following steps 1-1 and 1-2.

[0031] Step 1-1: b ｊ ≧a ｊｋ' (k′∈V(j)) ｊｋ' Then, determine whether b exists. ｊ ≧aｊｋ' (k′∈V(j)) ｊｋ' If b exists, execute step 1-2. ｊ ≧a ｊｋ' Satisfy a ｊｋ' does not exist (i.e., b ｊ <a ｊｋ (∀k∈V(j)) or V(j)=φ), power cancellation is terminated.

[0032] Step 1-2: b ｊ ←b ｊ -a ｊｋ' Also, V(j)←V(j)\{k'} and return to step 1-1.

[0033] In addition, in the above step 1-1, b ｊ ≧a ｊｋ' (k′∈V(j)) ｊｋ' Therefore, in step 1-2 above, these multiple a ｊｋ' Which of the a ｊｋ' A b ｊ There are various ways to determine whether to subtract from a. ｊｋ' Among them, the one with the largest value is a ｊｋ' A b ｊ One possible solution is to subtract the power demand from the virtual machine's power demand. This is a solution that prioritizes power offsetting for virtual machines with large power demands, since it is difficult to change their placement. Another solution is to evaluate the difficulty of changing the placement of virtual machines from a location or functionality perspective, and prioritize power offsetting for the power demand of virtual machines with high evaluation values. This is also a solution that prioritizes power offsetting for the power demand of virtual machines whose placement is difficult to change.

[0034] An example of the power offset in step S102 is shown in FIG. 4. In the example shown in FIG. 4, ｊ１ , a ｊ２ and a ｊ３ and power supply amount b ｊ This is an example of power cancellation in a data center j where there are two power sources. In this example, as a result of power cancellation, the power supply amount is finally reduced to b ｊ ←b ｊ -a ｊ１ -a ｊ２The power demand a ｊ３ (> b ｊ ) remains.

[0035] Another example of the power cancellation in step S102 is shown in Fig. 5. In the example shown in Fig. 5, ｊ１ and a ｊ２ and power supply amount b ｊ This is an example of power cancellation in a data center j where there are two power sources. In this example, as a result of power cancellation, the power supply amount is finally reduced to b ｊ ←b ｊ -a ｊ１ -a ｊ２ As a result, there is no remaining demand for electricity.

[0036] Next, the main processing unit 203 calculates the power demand amount a after the power offset in step S102. ｊｋ and power supply amount b ｊ Using this, the power supply amount b between the data centers is calculated. ｊ and electricity demand a ｊｋ In this case, the main processing unit 203 reduces the problem to a multiple knapsack problem and calculates the optimal VM placement as a solution to the problem (step S103).

[0037] Variable x in the multiple knapsack problem ｉｊｋ is defined as follows:

[0038] x ｉｊｋ : If virtual machine k∈V(j) is moved from data center j∈A to data center i∈A∪B, then x ｉｊｋ = 1, otherwise x ｉｊｋ = 0 In this case, the multiple knapsack problem for optimizing the placement of virtual machines can be formulated as follows:

[0039] However, the second constraint (inequality constraint) of the multiple knapsack problem shown in the above equation 1 is to be satisfied on a best effort basis (that is, for example, it is permitted that it is not satisfied for some i∈B).

[0040] The multiple knapsack problem shown in the above equation 1 represents that for virtual machines that have not been moved so that power can be supplied by solar cells, their power demand is met by purchasing power, but optimization is performed to minimize the cost of purchasing power.

[0041] An example of the optimal VM allocation in step S103 is shown in FIG. 6. In the example shown in FIG. 6, the power supply amount b ｊ Let j be a data center, and let a be the power demand a ｊ'１ and a ｊ'２ are present in a data center j', and the power demand a ｊ'１ The corresponding virtual machine and power demand a ｊ'２ The figure shows a case where a virtual machine corresponding to b is moved (placed) to data center j. ｊ -a ｊ'１ -a ｊ'２ ≧0 indicates that the second constraint of the multiple knapsack problem shown in Equation 1 above is satisfied.

[0042] In the multiple knapsack problem shown in the above equation (1), optimization is performed to minimize the cost of purchasing power to cover the power demand of the virtual machines that have not been moved. Therefore, the optimal placement of virtual machines is determined taking into consideration the power purchase cost. For example, as shown in FIG. 7, when the power supply amount b ｊ Let j be a data center, and let a be the power demand a ｊ'１ and a ｊ'２ is a data center j', and the power demand a ｊ''１ and a ｊ''２ Assume that there exists a data center j′′ with a certain value. ｊ' >c ｊ'' (i.e., when the power purchase cost of data center j′ is greater than that of data center j″), in order to minimize the total power purchase cost required to cover the power demands of the unmigrated virtual machines, the power demand a ｊ'１ The corresponding virtual machine and power demand a ｊ'２ The virtual machine that corresponds to the virtual machine is preferentially moved.

[0043] Next, the post-processing unit 204 calculates the power supply amount b ｊ and electricity demand a ｊｋ If there is a power surplus or a power shortage, the post-processing unit 204 calculates the amount of power surplus or power shortage (step S104). That is, for each data center j∈A∩B, the post-processing unit 204 performs power cancellation and calculates the amount of power surplus or power shortage by the following steps 2-1 to 2-4.

[0044] Step 2-1: b ｊ ≧a ｊｋ' (k′∈V(j)) ｊｋ' Then, determine whether b exists. ｊ ≧a ｊｋ' (k′∈V(j)) ｊｋ' If b exists, execute step 2-2. ｊ ≧a ｊｋ' Satisfy a ｊｋ' does not exist (i.e., b ｊ <a ｊｋ (If ∀k∈V(j)) or V(j)=φ), execute step 2-3.

[0045] Step 2-2: b ｊ ←b ｊ -a ｊｋ' Also, V(j)←V(j)\{k'} and return to step 2-1.

[0046] Step 2-3: b ｊ Then, it is determined whether b ｊ If b = 0, neither surplus nor shortage of power occurs, and the process ends. ｊ If not 0, perform steps 2-4.

[0047] Step 2-4: a ｊｋ (> b ｊ ) does not exist (i.e., V(j) = φ), b ｊ is the amount of surplus power. ｊｋ (> b ｊ ) exists (i.e., V(j) ≠ φ), then Σ ｋ∈Ｖ（ｊ） a ｊｋ -bｊ is the power shortage.

[0048] An example of the calculation of the surplus power amount and the shortage power amount in step S104 is shown in Fig. 8. In the example shown in Fig. 8, when V(j) = φ, the power supply amount b ｊ is used as the surplus power amount, and if V(j) ≠ φ, a ｊ１ -b ｊ In the example shown in Fig. 8, it is assumed that when V(j) ≠ φ, V(j) = {1}.

[0049] Finally, the output unit 205 outputs the optimal VM placement determined in step S103 to a predetermined output destination (e.g., a virtual machine control device, etc.), and also outputs surplus power obtained in a certain data center in step S104 as the amount of power to be sold, or outputs a power shortage obtained in step S104 as the amount of power to be purchased to a predetermined output destination (e.g., a power control device, etc.) (step S105). As a result, the virtual machine control device, etc. changes the placement of the virtual machines to the optimal VM placement. Furthermore, when surplus or shortage of power occurs in a certain data center, the surplus power is sold and the shortage of power is purchased by the power control device, etc.

[0050] <Summary> As described above, the optimal allocation determination device 10 according to this embodiment determines the optimal allocation of virtual machines through three stages of processing: pre-processing, main processing, and post-processing. This allows the power demand of as many virtual machines as possible to be met by the power supply from the solar cells, taking into account the balance between power supply and demand, thereby minimizing the cost of purchasing power. Furthermore, if there is surplus power supply from the solar cells after the power demand of the virtual machines has not been met, this surplus can be used as the amount of power sold.

[0051] Therefore, even if the amount of renewable energy power supply fluctuates due to weather changes, etc., the amount of power supply can be used efficiently and effectively. In other words, through the above three-stage process, the amount of power demand for ICT loads is covered as much as possible by the amount of renewable energy power supply so that the total cost of purchasing power for the shortfall is minimized, and the amount of power supply can be used efficiently. Furthermore, even if there is a surplus in the amount of renewable energy power supply, the surplus can be sold, so the surplus can be used effectively.

[0052] The present invention is not limited to the above-described specifically disclosed embodiments, and various modifications, changes, and combinations with known technologies are possible without departing from the scope of the claims.

[0053] [References] Reference 1: Handbook of Applied Mathematical Programming, Mikio Kubo et al., Asakura Publishing, 2002, pp. 240-241.

[0054] 10 Optimal placement determination device 101 Input device 102 Display device 103 External I / F 103a Recording medium 104 Communication I / F 105 RAM 106 ROM 107 Auxiliary storage device 108 Processor 109 Bus 201 Input unit 202 Pre-processing unit 203 Main processing unit 204 Post-processing unit 205 Output unit

Claims

1. An apparatus for determining optimal placement of ICT load, comprising: A device that, for each base station where a power supply is available, uses the power demand required by the ICT load located at the base and determines as the optimal layout an arrangement that minimizes, across all bases, the cost of purchasing the amount of power that exceeds the power supply available at the base, out of the power demand required by the ICT load when the ICT load is located at the base.

2. For each base where a power supply amount is available, a power demand amount required by an ICT load disposed at the base is offset with the power supply amount, The system is configured to offset all of the power demand of the ICT load after the optimal placement with the power supply available at the base where the ICT load is placed, using the power supply and power demand after the offset, 2. The device according to claim 1, wherein when the amount of power supply after the offset is greater than a predetermined value, the amount of power supply after the offset is output as an amount of surplus power to be sold, and when the amount of power demand after the offset is greater than a predetermined value, the amount of power supply after the offset is output as an amount of shortage power.

3. The device according to claim 2, configured to, when at least a portion of the power demand cannot be offset against the power supply, output an amount of power obtained by subtracting the offset power supply amount from the power demand that could not be offset as an amount of purchased power.

4. The apparatus according to claim 1 , further comprising: a device for determining the optimal placement of a plurality of objects; a device for determining the optimal placement of the plurality of objects;

5. The multiple knapsack problem is formulated as a problem of minimizing the sum of costs required to purchase an amount of electricity that exceeds the amount of electricity supply available at the base where the ICT load is located, out of the amount of electricity demanded by the ICT load whose location has not been changed, The device described in claim 4, wherein the multiple knapsack problem includes a first constraint indicating that each ICT load is located at one of the locations, and a second constraint indicating that the sum of the power demands of the ICT loads located at each location is less than or equal to the power supply available to the location.

6. The apparatus according to claim 5 , wherein the second constraint is a best-effort constraint to be satisfied.

7. A device for determining the optimal placement of ICT loads, A method for determining, for each base station where a power supply is available, the optimal arrangement is one that minimizes, across all bases, the cost of purchasing the amount of power that exceeds the power supply available at the base out of the amount of power demanded by the ICT load located at the base when the ICT load is located at the base, using the amount of power demanded by the ICT load located at the base.

8. A device that determines the optimal placement of ICT loads. A program that uses the amount of electricity demanded by the ICT load located at each base where electricity supply is available to determine as the optimal arrangement an arrangement that minimizes, across all bases, the cost of purchasing the amount of electricity demanded by the ICT load when the ICT load is located at the base that exceeds the amount of electricity supply available at the base.

9. An optimal placement determination device that determines an optimal placement of ICT loads that can be placed at multiple bases that constitute a target network, comprising: a pre-processing unit configured to offset, for each base that can use a power supply generated by renewable energy, a power demand required by an ICT load disposed at the base with the power supply; a main processing unit configured to determine, as the optimal arrangement, an arrangement that minimizes, across the multiple locations, the cost of purchasing an amount of electricity that exceeds the amount of electricity supply available at the location out of the amount of electricity demanded by the ICT load when the ICT load is placed at the location, using the amount of electricity supply and the amount of electricity demand after the offset by the pre-processing unit; a post-processing unit configured to offset the power demand of the ICT load after optimal placement determined by the main processing unit with the power supply amount available at the base where the ICT load is placed, using the power supply amount and power demand amount after offset by the pre-processing unit; an output unit configured to output the power supply amount after the offset as a sales power amount when all of the power demand amount can be offset with the power supply amount by the post-processing unit and the power supply amount after the offset is greater than a predetermined value; An optimal placement determination device having the above-mentioned.

10. The output unit:

10. The optimal allocation determination device according to claim 9, configured to, when the post-processing unit is unable to offset at least a portion of the power demand with the power supply, output an amount of power obtained by subtracting the amount of power supply after offsetting by the post-processing unit from the amount of power demand that could not be offset as an amount of purchased power.

11. The main processing unit:

11. The optimal placement determination device according to claim 9, configured to formulate a problem for determining the optimal placement as a multiple knapsack problem, and to determine the optimal placement by solving the multiple knapsack problem.

12. The main processing unit: The multiple knapsack problem is formulated as a problem of minimizing the sum of costs required to purchase an amount of electricity that exceeds the amount of electricity supply available at the base where the ICT load is located, out of the amount of electricity demanded by the ICT load whose location has not been changed, The optimal placement determination device of claim 11, wherein the multiple knapsack problem includes a first constraint condition indicating that each ICT load is placed at one of the bases, and a second constraint condition indicating that the total amount of electricity demanded by the ICT loads placed at each base is less than or equal to the amount of electricity supply available to the base.

13. The optimal placement determination device as described in claim 12, wherein the second constraint is a constraint that should be satisfied on a best effort basis.

14. An optimal placement determination device that determines an optimal placement of ICT loads that can be placed at multiple bases that constitute a target network, comprising: a pre-processing step of offsetting, for each base that can use the amount of power generated by renewable energy, the amount of power demanded by the ICT loads located at the base with the amount of power generated by renewable energy; a main processing procedure for determining, as the optimal arrangement, an arrangement that minimizes, across the multiple locations, the cost of purchasing an amount of electricity that exceeds the amount of electricity supply available at the location out of the amount of electricity demanded by the ICT load when the ICT load is placed at the location, using the amount of electricity supply and the amount of electricity demand after the offsetting in the preprocessing procedure; a post-processing step of offsetting the power demand of the ICT load after optimal placement determined by the main processing step with the power supply amount available at the site where the ICT load is placed, using the power supply amount and power demand amount after the offset by the pre-processing step; an output step of outputting the power supply amount after the offset as a sales power amount when all of the power demand amounts can be offset with the power supply amount by the post-processing step and the power supply amount after the offset is greater than a predetermined value; A method for determining optimal placement.

15. An optimal placement determination device for determining an optimal placement of ICT loads that can be placed at multiple bases that constitute a target network, comprising: a pre-processing step of offsetting, for each base that can use the amount of power generated by renewable energy, the amount of power demanded by the ICT loads located at the base with the amount of power generated by renewable energy; a main processing procedure for determining, as the optimal arrangement, an arrangement that minimizes, across the multiple locations, the cost of purchasing an amount of electricity that exceeds the amount of electricity supply available at the location out of the amount of electricity demanded by the ICT load when the ICT load is placed at the location, using the amount of electricity supply and the amount of electricity demand after the offsetting in the preprocessing procedure; a post-processing step of offsetting the power demand of the ICT load after optimal placement determined by the main processing step with the power supply amount available at the site where the ICT load is placed, using the power supply amount and power demand amount after the offset by the pre-processing step; an output step of outputting the power supply amount after the offset as a sales power amount when all of the power demand amounts can be offset with the power supply amount by the post-processing step and the power supply amount after the offset is greater than a predetermined value; A program that executes the following.