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

Abstract

An optimal arrangement determination device according to one aspect of the present disclosure comprises: a first input unit that inputs power demand amounts of ICT loads arranged at each base, power supply amounts of renewable energy power generation facilities installed at each base, and the capacities of storage batteries installed at each base; a second input unit that inputs a first parameter indicating the degree of suppression of a power consumption increase due to the movement of the ICT loads at each base, and a second parameter indicating the degree of suppression of the use of the storage batteries at each base; and an output unit that outputs at least one arrangement of the ICT loads determined, on the basis of the first parameter and the second parameter, so as to satisfy the power demand amounts with the power supply amounts and the capacities of the storage batteries.

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] A technology is known that optimizes the placement of ICT loads such as virtual machines (VMs) within a data center (DC) network that is equipped with renewable energy power generation facilities, taking into account weather conditions (e.g., Non-Patent Documents 1 and 2).

[0003] Ryota Nakamura, K. Harada, "Proposal of an ICT Load Placement Method Considering Renewable Energy," IEICE General Conference, B-14-4, March 2022. Ryutaro Matsumura, K. Harada, "A Method for Determining the Location of Multiple VMs Considering Solar Power Generation and Battery Storage in Data Centers," IEICE General Conference, B-14-6, March 2023.

[0004] However, while storage batteries may be installed at bases such as data centers, the conventional technologies described in Non-Patent Documents 1 and 2, etc., do not take into consideration the optimization of storage battery capacity when the placement of ICT loads is changed. Also, depending on the base, there may be a desire to reduce the consumption of storage battery capacity or to minimize the movement of ICT loads, but it has not been possible to flexibly respond to these requests.

[0005] The present disclosure has been made in consideration of the above points, and aims to provide technology that can flexibly optimize the placement of ICT loads, taking into account the capacity of storage batteries installed at base stations.

[0006] An optimal placement determination device according to one aspect of the present disclosure has a first input unit for inputting the power demand of ICT loads to be placed at each base, the power supply amount of renewable energy power generation equipment installed at each base, and the capacity of storage batteries installed at each base, a second input unit for inputting a first parameter representing the degree of suppression of increase in power consumption due to ICT load movement at each base, and a second parameter representing the degree of suppression of storage battery usage at each base, and an output unit for outputting at least one placement of the ICT loads determined based on the first parameter and the second parameter so as to satisfy the power demand amount with the power supply amount and the capacity of the storage batteries.

[0007] The technology provides the ability to flexibly optimize the placement of ICT loads, taking into account the capacity of storage batteries installed at base locations.

[0008] It is a diagram showing an example of a hardware configuration of an optimal placement determination device according to the present embodiment. It is a diagram showing an example of a functional configuration of an optimal placement determination device according to the present embodiment. It is a flowchart showing an example of an optimal placement determination process according to the present embodiment.

[0009] An embodiment of the present invention will be described below.

[0010] <Theoretical Configuration> The following describes the theoretical configuration of a technology that can flexibly optimize the placement of virtual machines (VMs) within a data center network in which solar power generation equipment and storage batteries are installed in at least some data centers (DCs), taking into account the storage battery capacity. Note that a data center network is a communication network that includes, for example, data centers where VMs can be placed, communication equipment that transfers communications to users, and links that connect the data centers and the communication equipment. A data center network spans a wide area, such as the entire country, and the amount of power generated by solar power generation equipment installed in a data center can vary depending on factors such as the weather in the area where the data center is located.

[0011] Note that a solar power generation facility is an example of a renewable energy power generation facility, and a data center may be equipped with renewable energy power generation facilities other than a solar power generation facility (for example, a wind power generation facility, a geothermal power generation facility, etc.). Furthermore, a VM is an example of an ICT load, and the placement of ICT loads other than VMs may be targeted for optimization. Furthermore, a data center is an example of a base, and ICT loads may be placed at bases other than a data center, or solar power generation facilities or storage batteries may be installed.

[0012] First, it is assumed that a number for identifying the data center is assigned to each data center. Similarly, it is assumed that a number for identifying the VM is assigned to each VM. In this case, a set of data centers with power demand is denoted as A, a set of VMs in data center j∈A is denoted as V(j), and a set of data centers to which power is supplied by a photovoltaic power generation facility or a storage battery is denoted as B. Furthermore, the power demand of a VM, the power supply amount of a photovoltaic power generation facility, and the storage battery capacity are respectively represented by the following symbols.

[0013] a ｊｋ : Power demand of VM k∈V(j) in data center j∈A b ｊ : Power supply amount of solar power generation equipment of data center j∈B s ｉ : Battery capacity of data center i∈B Here, the power demand of VM a ｊｋ , the power supply amount of the solar power generation facility b ｊ , and the battery capacity s ｉ can be obtained from, for example, an EMS (Energy Management System) or NMS (Network Management System) that manages the data center network. In the following, the power supply amount b ｉ and the battery capacity s of data center i∈B ｉ Sum of b ｉ +s ｉ It is assumed that the power demand of the VMs placed in the data center i can be met by the power supply amount represented by

[0014] In addition, the following parameters are introduced as parameters that can be set by users such as administrators and operators of the data center network.

[0015] l ｉ : Coefficient m representing the degree of suppression of power consumption increase due to VM migration in data center i∈B ｉ : Coefficient representing the degree of suppression of battery usage at data center i∈B. Here, parameter l ｉ can be set to any value equal to or greater than 0. However, if it is desired to suppress the increase in power consumption of data center i, a value greater than 1 should be set, and if it is not desired to suppress the increase in power consumption of data center i, a value between 0 and 1 should be set. Similarly, parameter m ｉ Any value greater than or equal to 0 can be set to , but if you want to suppress the use of storage batteries in data center i, you can set a value greater than 1, and if you do not want to suppress the use of storage batteries in data center i, you can set a value between 0 and 1.

[0016] Next, the variable b related to the power supply when the VM is migrated is ｉｊ and s ｉｊ (i∈B, j∈A) are defined as follows, respectively.

[0017] b ｉｊ : A variable s representing the amount of power supplied by the photovoltaic power generation equipment to the VM that has moved from the data center j ∈ A in the data center i ∈ B ｉｊ : A variable representing the battery capacity (i.e., the amount of battery power supply) supplied (distributed) to a VM migrated from a data center j∈A in a data center i∈B. Furthermore, a variable x ｉｊｋ is defined as follows:

[0018] x ｉｊｋ : If VM k∈V(j) is moved from data center j∈A to data center i∈B, then x ｉｊｋ = 1, otherwise x ｉｊｋ In this case, the allocation problem for optimizing the placement of VMs when VM migration is permitted and the use of storage batteries is also taken into consideration is expressed as the following formula (1). ｉｊ , s ｉｊ and x ｉｊｋ is the variable whose value changes.

[0019] min (Σ ｉ∈Ｂ Σｊ∈Ａ，ｊ≠ｉ Σ ｋ∈Ｖ（ｊ） l ｉ a ｊｋ x ｉｊｋ +Σ ｉ∈Ｂ Σ ｊ∈Ａ m ｉ s ｉｊ ) Subject to Σ ｉ∈Ｂ x ｉｊｋ =1(j∈A, k∈V(j)) Σ ｋ∈Ｖ（ｊ） a ｊｋ x ｉｊｋ =s ｉｊ +b ｉｊ (j∈A, i∈B) Σ ｊ∈Ａ b ｉｊ ≦b ｉ (i∈B) Σ ｊ∈Ａ s ｉｊ ≦s ｉ (i∈B) b ｉｊ ≧0(i∈B, j∈A) s ｉｊ ≧0(i∈B, j∈A) x ｉｊｋ ∈{0, 1}, i∈B, j∈A, k∈V(j) ...Equation (1) Here, the allocation problem shown in the above equation (1) means a problem of optimizing the placement of VMs so as to minimize the power consumption covered by moving VMs and the consumption of storage battery capacity (i.e., the consumption of power stored in the storage battery). This is to make the most of the power supply amount from the solar power generation equipment of each data center. ｉ is a coefficient that represents the degree of suppression of the increase in power consumption of data center i, and when its value is greater than 1, it is possible to reduce the increase in power consumption due to VM migration to data center i. ｉ is a coefficient that represents the degree of suppression of battery usage in data center i, and when its value is greater than 1, the amount of power supplied by the battery in data center i (i.e., the amount of power supplied by discharging the battery (battery power supply amount)) can be reduced.

[0020] The formulation of the above equation (1) is a general expression, ｉ and m ｉ By adjusting the following, the following control can be realized. For example, m ｉ= 0 (i∈B), the problem becomes an allocation problem that optimizes the placement of VMs without moving VMs as much as possible when the use of storage batteries is possible. This is to make the most of the power supply of each data center, and is set, for example, when the weather is good both on the day and the next day and there is no need to suppress the power consumption of storage batteries too much. Also, for example, if l ｉ If we set i∈B = 0 (i∈B), VMs can be moved, but the allocation problem is to optimize the placement of VMs so as to minimize the use of storage batteries. This is the case when we want to reduce the power consumption of storage batteries in each data center. For example, we set it when the weather is fine on the day, but the weather will be bad the next day and solar power generation cannot be expected. Also, for example, if we set l only for a certain data center i, ｉ is made greater than 1, while m ｉ If parameter l is set to a value smaller than 1, the allocation problem is to utilize the power supply from the storage battery while minimizing the increase in power consumption due to VM migration. This means that even if the weather is bad on the day, it will be set to a data center i that will be blessed with good weather the next day. In this way, parameter l can be set according to the situation of each data center. ｉ and m ｉ You may change the setting value.

[0021] By finding the optimal solution to the allocation problem shown in the above formula (1), the optimal VM placement and the optimal battery power supply amount can be obtained. That is, given a set A of data centers with power demand, a set V(j) of VMs in data center j∈A, a set B of data centers to which power is supplied by photovoltaic power generation facilities or storage batteries, and the optimal solution x ｉｊｋ The optimal VM placement is obtained from (i∈B, j∈A, k∈V(j)). ｉｊ The optimal battery power supply amount can be obtained from the above equation. If multiple solutions are found, one feasible solution can be randomly selected from the multiple solutions and set as the optimal solution.

[0022] Hereinafter, an optimal placement determination device 10 will be described that can flexibly optimize the placement of VMs by using the allocation problem described in this theoretical configuration, taking into account the capacity of storage batteries installed in a data center.

[0023] <Example of Hardware Configuration of Optimal Placement Determination Apparatus 10> An example of the hardware configuration of the optimal placement determination apparatus 10 according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of the hardware configuration of the optimal placement determination apparatus 10 according to this embodiment.

[0024] 1, an 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.

[0025] 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.

[0026] The external I / F 103 is an interface with an external device such as a recording medium 103a. Examples of the recording medium 103a include 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.

[0027] The communication I / F 104 is an interface for connecting the optimal placement determination apparatus 10 to a communication network. 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 non-volatile storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a flash memory. The processor 108 is one of various arithmetic devices such as a CPU (Central Processing Unit) or a GPU (Graphic Processing Unit).

[0028] 1 is an example, and the hardware configuration of the optimal allocation determination device 10 is not limited to this. For example, the optimal allocation 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 other than the hardware shown in the figure.

[0029] <Example of Functional Configuration of Optimal Placement Determination Apparatus 10> An example of the functional configuration of the optimal placement determination apparatus 10 according to this embodiment will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of the functional configuration of the optimal placement determination apparatus 10 according to this embodiment.

[0030] 2, the optimal allocation determination device 10 according to this embodiment includes an input unit 201, a formulation unit 202, an optimal solution calculation unit 203, and an output unit 204. Each of these units is realized, for example, by a process in which one or more programs installed in the optimal allocation determination device 10 are executed by a processor 108 or the like.

[0031] The input unit 201 inputs a set A of data centers with power demand, a set V(j) of VMs in each data center j∈A, and a set B of data centers to which power is supplied by a solar power generation facility or a storage battery. The input unit 201 also inputs the power demand a of VM k∈V(j) in each data center j∈A. ｊｋand the power supply amount b of the solar power generation equipment of each data center j∈B ｊ and the battery capacity s of each data center i∈B ｉ Furthermore, the input unit 201 inputs the parameter l ｉ and parameter m ｉ Enter .

[0032] The formulation unit 202 uses the various information and parameters input by the input unit 201 to formulate the assignment problem shown in the above equation (1).

[0033] The optimal solution calculation unit 203 calculates an optimal solution to the assignment problem shown in the above equation (1) formulated by the formulation unit 202. Note that the assignment problem shown in the above equation (1) can be solved using, for example, a solver that executes an existing solution method.

[0034] The output unit 204 outputs the optimal VM placement and the optimal battery power supply amount obtained from the optimal solution calculated by the optimal solution calculation unit 203 to a predetermined output destination. Note that the output destination is not limited to a specific output destination, and examples thereof include the display device 102 such as a display, the auxiliary storage device 107 such as an HDD, SSD, or flash memory, other devices connected via a communication network, other programs, etc. In particular, the output unit 204 may output the optimal solution to a control device, control program, etc. that actually controls the placement of VMs in accordance with the optimal VM placement.

[0035] <Optimal Placement Determination Process> An example of the optimal placement determination process according to this embodiment will be described with reference to Fig. 3. Fig. 3 is a flowchart showing an example of the optimal placement determination process according to this embodiment. Here, the processes of steps S101 to S105 in Fig. 3 are repeatedly executed at predetermined time intervals (also called control periods, etc.) that are set in advance by a user such as a data center network administrator or operator. Note that the time interval can be set to any value, and it is conceivable to set a value such as 30 minutes or 1 hour, for example.

[0036] First, the input unit 201 receives various information (a set A of data centers with power demand, a set V(j) of VMs in each data center j∈A, a set B of data centers to which power is supplied by photovoltaic power generation facilities or storage batteries, and the power demand a of VM k∈V(j) in each data center j∈A). ｊｋ , the power supply amount b of the solar power generation equipment of each data center j∈B ｊ , the battery capacity s of each data center i∈B ｉ ) (Step S101). The input unit 201 receives, for example, the power demand a of VM k∈V(j) in each data center j∈A from a system such as an EMS or NMS that manages a data center network. ｊｋ and the power supply amount b of the solar power generation equipment of each data center j∈B ｊ and the battery capacity s of each data center i∈B ｉ It is possible to obtain the following.

[0037] Next, the input unit 201 inputs various parameters (parameter l representing the degree of suppression of the increase in power consumption due to VM migration in data center i∈B, ｉ , parameter m ｉ ) is input (step S102).

[0038] Next, the formulation unit 202 formulates the allocation problem shown in equation (1) above using the various information input in step S101 above and the various parameters input in step S102 above (step S103).

[0039] Next, the optimal solution calculation unit 203 calculates an optimal solution to the allocation problem formulated in step S103 (step S104).

[0040] Then, the output unit 204 outputs the optimal VM placement and the optimal battery power supply amount obtained from the optimal solution calculated in the above step S104 to a predetermined output destination (step S105). Note that, if multiple solutions are calculated in the above step S104, the output unit 204 may, for example, randomly select one feasible solution from the multiple solutions and set it as the optimal solution.

[0041] <Summary> As described above, the optimal allocation determination device 10 according to this embodiment uses the parameter l ｉ and m ｉ By setting , and solving the allocation problem shown in the above formula (1), it is possible to determine the optimal VM placement taking into consideration minimizing VM migration, minimizing the consumption of storage battery capacity, or both, depending on the situation of each data center. Therefore, for example, the parameter l can be set according to a desire to avoid VM migration as much as possible in consideration of the increase in power consumption and deterioration of reliability due to VM migration, a desire to minimize the use of storage battery capacity, which is a scarce asset, or a desire to achieve both. ｉ and m ｉ By appropriately setting the value of , it is possible to determine the optimal VM placement for the request.

[0042] That is, the amount of power supplied by the solar power generation equipment installed in the data center fluctuates depending on the weather. However, by optimally migrating VMs between data centers in response to this fluctuation, for example, many VMs can be migrated to a data center with a large amount of power supplied by the solar power generation equipment, and VMs can be migrated to another data center from a data center with a small amount of power supplied by the solar power generation equipment. On the other hand, if a storage battery is installed in the data center, a shortage of power supplied by the solar power generation equipment can be compensated for by consuming the storage battery capacity without migrating VMs. Therefore, by considering both the amount of power supplied by the solar power generation equipment and the storage battery capacity, it is possible to determine an optimal VM placement that considers whether to limit VM migration, limit the use of the storage battery, or both.

[0043] 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.

[0044] REFERENCE SIGNS LIST 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 Formulation unit 203 Optimal solution calculation unit 204 Output unit

Claims

1. An optimal placement determination device having: a first input unit for inputting the power demand of ICT loads placed at each base, the power supply amount of renewable energy power generation equipment installed at each base, and the capacity of storage batteries installed at each base; a second input unit for inputting a first parameter representing the degree of suppression of the increase in power consumption due to the movement of ICT loads at each base, and a second parameter representing the degree of suppression of storage battery usage at each base; and an output unit for outputting at least one placement of the ICT loads determined so as to satisfy the power demand amount with the power supply amount and the capacity of the storage batteries based on the first parameter and the second parameter.

2. An optimal placement determination device as described in claim 1, comprising: a formulation unit that uses the power demand, the power supply, the capacity of the storage battery, the first parameter, and the second parameter to formulate an optimization problem that satisfies the power demand with the power supply and the capacity of the storage battery, while not moving the ICT load as much as possible, and minimizing utilization of the storage battery; and an optimal solution calculation unit that calculates an optimal solution to the optimization problem, wherein the output unit outputs at least one placement of the ICT load obtained from the optimal solution.

3. The first parameter and the second parameter are set for each of the bases, and the formulation unit calculates the first parameter set for the i-th base by ｉ , the second parameter set for the i-th site is m ｉ , the power demand of the kth ICT load located at the jth site is a ｊｋ , a variable x is a variable that takes a value of 1 when the kth ICT load located at the jth site is moved to the ith site, and takes a value of 0 when it is not moved. ｉｊｋ The capacity of the storage battery distributed to the ICT load moved from the jth site to the ith site is denoted by s ｉｊ Then, for i and j (where j ≠ i), ｉ a ｊｋ x ｉｊｋ and m for i and j ｉ s ｉｊ 3. The optimal placement determination apparatus according to claim 2, further comprising: a step of formulating an optimization problem for minimizing a sum of 4. The first parameter l ｉ is set to a value of 0 or more, and when a value greater than 1 is set, it indicates that the increase in power consumption due to the movement of the ICT load at the i-th location is suppressed, and when a value between 0 and 1 is set, it indicates that the increase in power consumption due to the movement of the ICT load at the i-th location is not suppressed; ｉ is set to a value of 0 or greater, and when a value greater than 1 is set, it indicates that the storage battery utilization at the i-th location is suppressed, and when a value between 0 and 1 is set, it indicates that the storage battery utilization at the i-th location is not suppressed.

5. An optimal placement determination method executed by a computer, comprising: a first input step of inputting the power demand of ICT loads placed at each base, the power supply amount of renewable energy power generation equipment installed at each base, and the capacity of storage batteries installed at each base; a second input step of inputting a first parameter representing the degree of suppression of the increase in power consumption due to the movement of ICT loads at each base, and a second parameter representing the degree of suppression of storage battery use at each base; and an output step of outputting at least one placement of the ICT loads determined so as to satisfy the power demand amount with the power supply amount and the capacity of the storage batteries based on the first parameter and the second parameter.

6. A program that causes a computer to execute the following steps: a first input procedure for inputting the power demand of ICT loads placed at each base, the power supply amount of renewable energy power generation equipment installed at each base, and the capacity of storage batteries installed at each base; a second input procedure for inputting a first parameter representing the degree of suppression of the increase in power consumption due to the movement of ICT loads at each base, and a second parameter representing the degree of suppression of storage battery use at each base; and an output procedure for outputting at least one placement of the ICT loads determined so as to satisfy the power demand amount with the power supply amount and the capacity of the storage batteries based on the first parameter and the second parameter.

Images