Operation plan creation device and operation plan creation method for data center

Abstract

To increase speed and optimize a cooperative operation plan creation work between a data center operator and a user.SOLUTION: In an operation plan creation method in a data center, an operation plan server 60 calculates a handling cost to handle a given consumption pattern of a DC user 98 when an operation plan of a created workload (WL) is to be created or changed, and creates an optimization equation for maximizing or minimizing an index such as a cost including the handling cost. At this time, a solution of the optimization equation is obtained such that a consumption pattern obtained by changing a consumption pattern in a past response history of the user according to a certain rule is available at a corresponding handling cost, thereby creating an operation plan draft for the consumption pattern of each user and a device held by a data center operator. The operation plan server 60 confirms, to a WL execution management server 70, whether the operation plan draft can be executed or not, and instructs the WL execution management server to execute the operation plan draft.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an operation plan creation device and an operation plan creation method that use a hierarchical operation plan optimization technique for quickly creating an operation plan that includes negotiations within a data center. [Background technology]

[0002] As a measure against climate change, the introduction of renewable energy is progressing rapidly along with the efficient use of energy. However, because the output of renewable energy fluctuates depending on weather conditions, it can be difficult to adjust the amount of power generated to match power consumption. For this reason, it is important to take measures to match supply and demand by utilizing storage batteries and the ability to adjust for demand fluctuations. While power transmission and distribution companies, microgrid operators, and electricity retailers require this adjustment ability to ensure a stable power supply and avoid imbalances, consumers and distributed energy resource (DER) operators are seeking benefits from the provision of adjustment ability and the use of renewable energy for decarbonization, and cooperative operation to exchange power between these operators is required.

[0003] Data centers (hereinafter referred to as "DC"), which are one of the major consumers of electric energy, consume a lot of power and can vary their demand by varying the execution time and execution location of workloads (hereinafter referred to as "WL" or "jobs") executed on servers (hereinafter referred to as "WL shift"). This makes them promising candidates for cooperative operation. Meanwhile, workloads executed in data centers (DCs) include not only the DC operator's own workload but also the workloads of DC users, who are separate businesses from the DC operator. In cases other than in-house use, the majority of workloads are those of DC users, and the DC operator cannot directly control them. To effectively utilize the balancing power of data centers, cooperative operation between DC operators and DC users is required.

[0004] By creating an optimal cooperative operation plan between data center operators and users from the perspectives of cost and power supply stability, efficient cooperative operation of data centers can be expected. However, when creating a cooperative operation plan, it is undesirable to disclose the equipment information and operation information owned by each operator from the perspective of privacy. Therefore, it is necessary to optimize the operation plan without disclosing the equipment information of each DC operator. Patent Literature 1 discloses a method for creating an overall optimal operation plan regarding the creation of a cooperative operation plan between a distribution system operator and a DER operator that owns distributed energy resources (DERs). The method decomposes the problem of minimizing overall energy costs into a main problem in which the distribution system operator determines the amount of power generated by each operator and a subordinate problem in which the distribution system operator determines the control of the equipment owned by each operator. The method then creates an overall optimal operation plan by repeatedly solving these problems while exchanging information such as power generation amount and energy cost. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent Publication No. 2021-136757 Summary of the Invention [Problem to be solved by the invention]

[0006] In Patent Document 1, it is possible to optimize operation plans between operators with a two-tier relationship, such as a distribution system operator and a DER operator, or a DC operator and a DC user, where the operation plan of one party becomes part of the operation plan of the other party. However, the DC operator needs to create a cooperative operation plan with operators with a three-tier relationship, such as a DC user (within the data center) who is part of the DC operator's operation plan, and a transmission and distribution operator or an electricity retailer (outside the data center) where the DC operator's operation plan becomes part of the other party's operation plan. In cooperative operation plan optimization with an operator outside the data center, the data center's operation plan is repeatedly required to be re-planned, so cooperative operation plan optimization within the data center needs to be performed quickly in response to re-planning requests from an operator outside the data center.

[0007] The present invention has been made in consideration of the above background, and its purpose is to provide an operation plan creation device and an operation plan creation method that are capable of quickly planning and re-planning in collaborative operation plan optimization between an operation plan control unit of a data center and data center users. [Means for solving the problem]

[0008] To solve the above problems, the present invention is applied to a datacenter that has multiple server devices that execute workloads, a workload execution management server installed on each server device, and an operation planning server connected to the execution management server, and in which multiple users use the server devices via a network. The workload execution management server manages the execution of workloads (WL) for users in the datacenter and has a cost calculation unit that calculates the cost of responding to power consumption (hereinafter referred to as consumption pattern) in each given time block. The operation planning server is configured to include an operation plan optimization problem creation unit that stores user consumption patterns and creates optimization problems that maximize or minimize indicators such as costs, including the corresponding cost; an optimization formula update unit that changes consumption patterns in the user response history according to certain rules so that the consumption patterns can be used at the corresponding corresponding cost; and an operation plan proposal calculation unit that solves the updated optimization formula to create an operation plan proposal for each user's consumption pattern and the equipment owned by the operator. The cost calculation unit, cooperative operation plan optimization problem creation unit, optimization formula update unit, and operation plan proposal calculation unit can be realized by software when a processor in the workload execution management server or the operation planning server executes a specific computer program to achieve these functions.

[0009] According to another feature of the present invention, the operation planning server compares the consumption patterns of users with the consumption pattern baselines of the corresponding users that have been registered in advance, calculates the lower limit of power consumption corresponding to the amount of job queues waiting to be executed in each time block for each consumption pattern, and each time it receives a consumption pattern and corresponding cost from a user, it updates the operation planning optimization formula for all users so that the consumption pattern that can be realized by accelerating the execution of the job queues calculated from that consumption pattern (hereinafter referred to as early execution patterns) becomes available at the corresponding corresponding cost.

[0010] According to yet another feature of the present invention, when creating or changing an operation plan for the created workload, the operation planning server calculates the consumption pattern and corresponding cost for each user of the server device, creates an optimization formula that maximizes or minimizes an index of the corresponding cost taking into account the consumption pattern of the user of the server device, updates the optimization formula so that, for consumption patterns in the past response history from users, consumption patterns that have been changed according to certain rules can be used at the corresponding corresponding cost, and by finding a solution to the updated optimization formula, creates an operation plan for each user's consumption pattern and the equipment held by the data center operator. [Effects of the Invention]

[0011] According to the present invention, an operation planning server creates an operation plan while negotiating with a workload (WL) execution management server about the consumption patterns of each data center (DC) user. This allows for the creation of a collaborative operation plan that adjusts the data center's power consumption to approach the target without disclosing the individual workload information of the data center users to the data center operator. This also makes it possible to minimize the total cost paid by the data center operator, including the corresponding costs paid to the data center users, for this adjustment. Furthermore, by taking into consideration the early execution of jobs whose execution was delayed in previously proposed consumption patterns during negotiation, the overall number of negotiations can be reduced, allowing for faster replanning.

[0012] Configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of an operation plan creation system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram illustrating an example of hardware and functions included in the operation plan server 60 of FIG. [Figure 3] FIG. 3 is a diagram showing an example of a user management table 100 in FIG. 2. [Figure 4] FIG. 3 is a diagram showing an example of a response result management table 200 of FIG. [Figure 5] 3 is a diagram showing an example of a consumption pattern baseline management table 300 of FIG. 2. FIG. [Figure 6] 3 is a diagram illustrating an example of hardware and functions included in the WL execution management server 70 of FIG. 2. FIG. [Figure 7] FIG. 7 is a diagram showing an example of a WL management table 500 in FIG. 6. [Figure 8] FIG. 7 is a diagram showing an example of a response history table 600 in FIG. [Figure 9] FIG. 10 is a diagram illustrating an example of a consumption pattern of a DC user. [Figure 10] FIG. 10 is a diagram illustrating an example of fluctuations in consumption patterns due to early execution and cost identification. [Figure 11] FIG. 10 is a diagram illustrating an example of a first coordinated operation plan creation process. [Figure 12] 12 is a flowchart illustrating the details of the cooperative operation plan optimization problem creation process of FIG. 11. [Figure 13] 12 is a flowchart illustrating the details of the job queue calculation process of FIG. 11. [Figure 14] 12 is a flowchart illustrating details of the optimization formula update (equalizing costs of early execution patterns) process of FIG. 11. [Figure 15] 12 is a flowchart illustrating details of the corresponding cost calculation process of FIG. 11. [Figure 16]FIG. 10 is a diagram illustrating an example of a management screen for a DC business operator. [Figure 17] FIG. 10 is a diagram showing an example of a management screen for a DC user. [Figure 18] FIG. 10 is a diagram illustrating an example of a second coordinated operation plan creation process. DETAILED DESCRIPTION OF THE INVENTION

[0014] FIG. 1 is a diagram showing an example of the overall configuration of an operation plan creation system 1 according to this embodiment. The operation plan creation system 1 of the present invention includes a data center 20 and an integrated DC management system 90 connected via a network 5. Necessary power is supplied to the data center 20 by an electricity retailer 2. An operation plan creation device according to this embodiment is installed in the data center 20. The data center 20 includes an operation plan server 60 and a WL execution management server 70 for each DC user (A, B, DC operator). In addition, one or more server devices 81, storage devices 82, and NW (network) devices 83 are provided for each DC operator and DC user. The server devices 81 are devices for executing the workloads of the DC users, and the WL execution management server 70 is devices for managing the execution of the workloads in the server devices 81. The NW devices 83 are well-known devices including network switches and the like. These facilities and devices operate while consuming power proportional to the workload (WL) executed by the server devices 81.

[0015] The DC business operator terminal 95 is an information processing device used by the DC business operator 96 that manages the data center 20. The DC user terminal 97 is an information processing device used by a DC user 98 that uses the data center 20. Here, only one DC user 98 is illustrated, but in reality there are many users, and multiple DC user terminals 97 used by each of them are connected to the network 6.

[0016] The integrated DC management system 90 includes a plan viewing server 91 as one of its components. In response to access from a user (such as a DC user), the plan viewing server 91 acquires necessary information from the operation plan server 60 and the WL execution management server 70 via the networks 5 and 6, and presents a plan viewing screen on a DC user terminal 98. The user terminal 98 can receive input instructions from the user on the plan viewing screen and transmit them to the integrated DC management system 90 and the data center 20. The integrated DC management system 90 is provided to a DC provider 96 and a DC user 98 as, for example, SaaS (Software as a Service). That is, the integrated DC management system 90 is accessed from a DC provider terminal 95 and a DC user terminal 97 via the network 6.

[0017] Electricity from the electricity retailer 2 is supplied to a second data center (DC2) 21, a DER operator 40, and general consumers 50. The second data center 21 can be realized with a configuration similar to that of the first data center (DC1). Note that while FIG. 1 illustrates only two data centers, 20 and 21, that receive electricity supply from the electricity retailer 2, in reality, there are many data centers. The supply and demand adjustment server 10 is a computer installed within the electricity retailer 2, which creates an electricity operation plan based on a demand forecast, and when demand adjustment becomes necessary, it interchanges electricity by requesting the data centers 20 and 21, the DER operator 40, and the general consumers 50 to re-plan their operation plans.

[0018] The operation planning server 60 creates an operation plan for each server in the data center 20. That is, the operation planning server 60 divides the time period to be planned into several time blocks, and determines the control parameters of the devices owned by the DC business operator 96 and the power consumption by each DC user 98 for each time block. In this embodiment, the operation plan is created in units of 30 minutes as the length of the time block. Note that the time block does not have to be in units of 30 minutes, and may be in units longer or shorter than that, for example, in units of one hour.

[0019] The WL execution management server 70 creates a workload execution plan to realize the power consumption (hereinafter referred to as consumption pattern) in each time block determined by the operation planning server 60, and calculates the cost required to realize that consumption pattern (hereinafter referred to as corresponding cost). The operation planning server 60 and the WL execution management server 70 exchange the consumption pattern and corresponding cost, and create an operation plan that maximizes or minimizes an index (for example, total cost) determined by the DC operator.

[0020] FIG. 2 is a diagram illustrating an example of the hardware and functions of the operation planning server 60. The operation planning server 60 includes a processing device 61, a memory 62, a storage device 63, and a communication device 64. The processing device 61 includes processors such as a central processing unit (CPU), a digital signal processor (DSP), a graphics processing unit (GPU), and a field-programmable gate array (FPGA), and executes various programs stored in the storage device 63 to implement the operation plan creation method according to this embodiment. The memory 62 is called a main storage device and includes a read-only memory (ROM), a random access memory (RAM), and the like. The storage device 63 is an auxiliary storage device such as a hard disk drive (HDD) or a solid-state drive (SSD). The communication device 64 is configured using a known network interface such as a network interface card (NIC), a wireless communication module, a universal serial interface (USB) module, or a serial communication module.

[0021] The storage device 63 of the operation planning server 60 stores the following programs: a cooperative operation plan optimization problem creation program 65, a job queue calculation program 66, an operation plan proposal calculation program 67, and an optimization formula update program 68. In addition to the programs shown here, other programs are also stored in the storage device 63. The storage device 63 stores a DC user management table 100, a response result management table 200, and a consumption pattern baseline management table 300, which are referenced and updated by the programs (65 to 68) executed by the processing device 61.

[0022] The cooperative operation plan optimization problem creation program 65 uses the control of each device owned by the DC business operator 96 and the consumption pattern of each DC user 98 as variables to create an optimization problem that maximizes or minimizes indicators such as costs including the corresponding costs of the consumption pattern of each DC user 98. The function achieved by executing the cooperative operation plan optimization problem creation program 65 corresponds to the "cooperative operation plan optimization problem creation unit" of the present invention.

[0023] The job queue calculation program 66 compares the consumption pattern of the DC user 98 with the consumption pattern of the user of the corresponding data center registered in the consumption pattern baseline management table 300, and calculates the lower limit of power consumption corresponding to the amount of job queue waiting to be executed in each time block for each consumption pattern. The function achieved by executing the job queue calculation program 66 corresponds to the "job queue calculation unit" of this invention.

[0024] Every time the optimization formula update program 68 receives a consumption pattern and a corresponding cost from a DC user 98, it updates the operation plan optimization formula so that a consumption pattern that can be realized by accelerating the execution of the job queue calculated by the job queue calculation program 66 (hereinafter referred to as an early execution pattern) can be used at the corresponding cost. The function achieved by executing the optimization formula update program 68 corresponds to the "optimization formula update unit" of the present invention.

[0025] The operation plan proposal calculation program 67 solves the optimization problem created by the cooperative operation plan optimization problem creation program 65 and updated by the optimization formula update program 68, and outputs the control plan for the equipment of the DC operator 96 and the consumption pattern for each DC user 98 obtained as the optimal solution. The function achieved by executing the operation plan proposal calculation program 67 corresponds to the "operation plan proposal calculation unit" of the present invention. Note that the operation plan server 60 shown in Fig. 2 may be equipped with an input device consisting of a mouse, keyboard, etc., and an output device consisting of a liquid crystal display, organic EL (Electro-Luminescence) display, etc.

[0026] Next, each piece of information stored in the operation plan server 60 will be described in detail.

[0027] (User list table) 3 is a diagram showing an example of a user list table 100, which stores the ID information of each DC user 98 and whether or not they are able to respond to operation plan optimization. In the user list table 100, one record is configured by the name of the DC user 98 (user name 101), the user's identification information (user ID 102), and information 103 indicating whether the DC user 98 identified by the user name 101 and user ID 102 is able to respond when a collaborative operation plan is created. In the example of FIG. 3, it is stored that user A, whose user ID 102 is 1, is able to respond when a collaborative operation plan is created, and it is stored that user B, whose user ID 102 is 2, is unable to respond when a collaborative operation plan is created.

[0028] (Response result management table) The response result management table 200 is a table that stores each response from the WL execution management server and the calculated job queue. FIG. 4 is a diagram showing an example of the response result management table 200. The response result management table 200 sets unique identification information for each response for each DC user 98 indicated by a user ID 201 (corresponding to the user ID 102 in the user list table 100 in FIG. 3), in which the identification information of the DC user 98 is set. For example, when two responses are obtained from a DC user 98 whose user ID 201 is "1," "1" and "2" are assigned as the response IDs 202. The response cost 203 stores the cost required for the response with the response ID 202 to the user ID 201. Here, the cost per response is stored as 500 yen and 1000 yen. The reflection flag 204 stores the result of whether or not the response information has been reflected in the optimization formula.

[0029] The columns after Reflected 204 store power consumption 210 calculated from the consumption pattern in the response and power consumption 220 calculated by the job queue calculation program 66 during job queue execution. Here, a plurality of columns 210 and 220 are provided, each divided into time periods. For example, the first column, i.e., user ID "1" and response ID "1," indicates that the power consumption 210 required to execute the processing of the consumption pattern from 0:00 to 0:30 is 2 kWh, and the power consumption 220 required to execute the processing of the job queue during the same time period is 2 kWh; the power consumption 210 required to execute the processing of the consumption pattern from 0:30 to 1:30 is 3 kWh, and the power consumption 220 required to execute the processing of the job queue during the same time period is 3 kWh.

[0030] Similarly, the power consumption 220, 230 for another response ID 202 of the same user ID 201 is also stored. Furthermore, the power consumption 220, 230 for all response IDs 202 of another user ID 201 is also stored. In the column where the user ID 201 is "1" and the response ID 202 is "2", the power consumption 210, 220 has been calculated, but these have not been reflected because the reflected flag 204 is False. In this way, the response result management table 200 is made up of one or more records having each item.

[0031] (Consumption pattern baseline management table) The consumption pattern baseline management table 300 is a table that stores the ID information of each DC user 98 and whether or not the user can respond to operation plan optimization. FIG. 5 is a diagram showing an example of a consumption pattern baseline management table. The consumption pattern baseline management table 300 is composed of one or more records that contain each item of a user ID 301 (corresponding to the user ID 102 in the user list table 100) in which the identification information of the DC user 98 is set, and a consumption pattern 310 in a workload execution plan that considers only the DC user 98's own convenience. The consumption pattern baseline management table 300 indicates that, according to the DC user 98's usual consumption pattern, the power consumption from 0:00 to 0:30 is 2 kWh, the power consumption from 0:30 to 1:00 is 3 kWh, the power consumption from 1:00 to 1:30 is 2 kWh, and the power consumption from 1:30 to 2:00 is 3 kWh. Note that, in the workload execution plan based on the consumption pattern 310, the workload may be executed immediately when execution becomes possible, or the workload may be delayed by being placed in a job queue.

[0032] FIG. 6 illustrates an example of the hardware and functions of a WL execution management server 70. The WL execution management server 70 includes a processing unit 71 (processor) such as a CPU, DSP, GPU, or FPGA; a memory (main storage device) 72 such as ROM or RAM; a storage device 73 such as an HDD or SSD; and a communication device 74 configured with a NIC, wireless communication module, USB module, or serial communication module. The processing unit 71 executes various programs stored in the storage device 73 to manage workload execution. Although not shown here, the WL execution management server 70 may also include input devices such as a mouse and keyboard and output devices such as a display. The storage device 73 stores multiple programs, not shown, including a corresponding cost calculation program 75 and a WL control plan execution program 76. The storage device 73 also manages various information, including a WL management table 500 and a response history table 600, which are referenced and changed during the execution of various programs.

[0033] The corresponding cost calculation program 75 receives a consumption pattern as input, creates a workload execution plan, and calculates the cost to be paid for requesting the DC user 98 to change the workload plan and job execution plan in order to realize the consumption pattern. The function achieved by executing the corresponding cost calculation program 75 corresponds to the "cost calculation unit" of the present invention. The WL control plan execution program 76 controls the execution of the workload in real time based on the workload execution plan. The function achieved by executing the WL control plan execution program 76 corresponds to the "WL control plan execution unit" of the present invention.

[0034] Next, each piece of information stored in the WL execution management server 70 will be described in detail.

[0035] (WL management table) The WL management table 500 is a table that stores information about each workload. Fig. 7 is a diagram showing an example of a WL management table. The WL management table 700 stores a workload ID 501 in which identification information for the workload is set, an estimated power consumption 502 that is expected to be consumed by executing the workload, an execution deadline 503 that indicates the deadline for delaying the execution of the workload, and an execution delay cost 504 that is charged to the DC provider 96 in accordance with the loss suffered by the DC user 98 due to the delay in execution. As shown in Fig. 7, the WL management table 700 is made up of one or more records identified by the workload ID 501.

[0036] The execution delay cost 504 is expressed, for example, as the delay cost when a workload is not executed immediately and the delay cost when the execution deadline is exceeded. This delay cost is a value used in a formula for solving the cooperative operation plan optimization problem, and its unit is yen (￥). Note that the unit is not limited to yen, and other formats may be used as long as the execution delay cost can be expressed as a monotonically increasing function that defines the execution time. The records recorded in each column of the WL management table 700 include workloads for which information is available in advance, such as regularly executed workloads and pre-registered workloads, as well as workloads generated based on predictions for which accurate workload information is not available in advance, such as workloads resulting from website access, etc.

[0037] (Response history table) The response history management table 600 is a table that stores responses sent by the WL (workload) execution management server 70 to the operation planning server 60, and the workload execution plans for those responses. Fig. 8 is a diagram showing an example of the response history management table 600. When the execution management server 70 sends a response to the operation planning server 60, unique identification information, i.e., a response ID 601, is set for that response. The response history management table 600 is made up of one or more records identified by the response ID 601, and each record has the following items: a response cost 602 for the response, a power consumption 610 for the response, and an execution WL 620 that indicates the workload ID to be executed in each time block when realizing the power consumption. For example, in a record where the response ID 301 is "1", the execution management server 70 indicates to the operation plan server 60 that when the operation plan is changed to the execution WL 620 (the execution workload is "1, 3" from 0:00 to 0:30, and the execution workload is "2, 4, 6, 7, 9" from 0:00 to 0:30), the power consumption 610 will be 1 kWh and 5 kWh, respectively, and the cost 602 to be paid to the DC user 98 for changing the execution plan is recorded as 400 yen. Similarly, the history of multiple responses is stored in the response history table 600. Note that the cost of the response ID 601 is displayed as "Inf￥", which means that the cost is infinite and the operation is not possible, and the execution workload is stored as "N / A (Not Applicable)", and the power consumption 610 is stored as 0 kWh.

[0038] The functions of the operation planning server 60 and the WL execution management server 70 described above are realized by the processing devices 61, 71 reading and executing programs stored in the auxiliary storage devices 63, 73. The programs can be distributed by being recorded on a portable recording medium, or downloaded from a program distribution server via a network. All or part of each server 60, 70 may be realized using virtual information processing resources provided using virtualization technology, process space separation technology, or the like, such as a virtual server provided by a cloud system. All or part of the functions provided by each information processing device may be realized by a service provided by a cloud system via an API (Application Programming Interface), for example.

[0039] Next, an example of a consumption pattern of a DC user 98 will be described using the consumption pattern 700 in FIG. 9 . The consumption pattern 700 stores a user ID 701 (corresponding to the user ID 102 in the user list table 100) in which the DC user 98's identification information is set, and the DC user 98's hourly power consumption for each time block of the target period of the operation plan. Here, the consumption pattern 700 is composed of hourly power consumption values ​​710, 711, 712, 713, and so on. While FIG. 7 only shows the power consumption from 0:00 to 2:00, consumption patterns for a longer time period, preferably 24 hours, are stored. When an operation plan is created, this consumption pattern 700 is sent from the operation plan server 60 to the WL execution management server 70 as an execution workload proposal, and the WL execution management server 70 replies with the corresponding costs required to realize the consumption pattern. Once the operation plan is created and the consumption patterns of each DC user 98 are determined, the DC user 98 receives the corresponding costs from the DC provider 96, but is responsible for executing the planned workload to realize the consumption pattern.

[0040] Next, we will explain how to identify early execution of workloads with corresponding costs. Consumption pattern A corresponds to the early execution pattern of consumption pattern B, which means that there is a workload execution plan for consumption pattern B that realizes consumption pattern A and executes all workloads earlier than the execution times in the workload execution plan for consumption pattern B. This determination is made using consumption pattern A, consumption pattern B, and the job queue in consumption pattern B. The conditions for the determination are as follows:

[0041] (1) In the consumption pattern, the power consumption corresponding to the reduction in the job queue (the amount taken out of the job queue and executed) is considered to be for batch jobs, and the rest is considered to be for interactive jobs. For all time blocks, it is determined whether the power consumption of consumption pattern A exceeds the power consumption of interactive jobs in consumption pattern B. This makes it possible to confirm that interactive jobs can be executed at the same time in consumption pattern A.

[0042] (2) For each time t, confirm that the "total power consumption from the start time to time t in consumption pattern A" is greater than the "total power consumption from the start time to time t in consumption pattern B" and less than the "total power consumption and job queue sum from the start time to time t in consumption pattern B." This confirms that the total amount of workload that can be executed up to each time is greater in consumption pattern A than in consumption pattern B, and that consumption pattern A can be realized by changing the execution time of the workload.

[0043] By combining the above conditions (1) and (2), it can be confirmed that consumption pattern A can be realized by using the workload execution plan for consumption pattern B, while continuing to execute interactive jobs as is, and by advancing the execution time of batch jobs.

[0044] FIG. 10 shows an example of consumption pattern fluctuations and cost identification due to early execution. This example will be used to explain the determination of early execution and cost identification. The horizontal axis of the consumption pattern and job key within the boxes shown in (a) to (e) represents the time frame, where, for example, t=1 to t=6 represent a 30-minute interval. For example, time t=1 corresponds to 0:00 to 0:30, time t=2 corresponds to 0:30 to 1:00, time t=3 corresponds to 1:00 to 1:30, and similarly, time t=6 corresponds to 2:30 to 3:00. Assume that the WL execution management server 70 shown in FIG. 1 currently receives the consumption pattern and job queue for response 1 (1201) and the consumption pattern and job queue for response 2 (1202) from the operation planning server 60 (reference).

[0045] Consumption pattern 1 (1203) is a pattern in which the batch job of response 1 (1201) that is executed at time t = 6 and the batch job that has not yet been executed to completion are executed at time t = 5, and this pattern satisfies conditions (1) and (2). Therefore, as shown by dotted line 1211, consumption pattern 1 (1203) corresponds to the early execution pattern of response 1 (1201).

[0046] On the other hand, when response 2 (1202) is compared with consumption pattern 1 (1203), condition (2) is not satisfied, and therefore consumption pattern 1 (1203) is not an early execution pattern of response 2 (1202). In fact, consumption pattern 1 cannot be realized unless the execution of the batch job executed at time t=4 is delayed. Here, consumption pattern 1 (1203) is considered to be the cheapest early execution pattern of response 1 that satisfies the conditions, and it is considered possible to implement it for a total of 5,000 yen, including the cost that the workload provider 96 will pay to the workload user 98 to implement it and the corresponding additional power consumption.

[0047] Consumption pattern 2 (1204) satisfies conditions (1) and (2) for both response 1 (1201) and response 2 (1202). Here, consumption pattern 2 is considered to be an early execution pattern for response 2 (1202), which is the cheapest among those that satisfy the conditions, as indicated by arrows 1212 and 1213, and is considered to be executable at a cost of 1,000 yen.

[0048] Consumption pattern 3 (1205) does not satisfy condition (1) for response 1 (1201) and response 2 (1202). In fact, the power consumption at time t=5 is lower than the power consumption for the interactive jobs at time t=5 for both response 1 (1201) and response 2 (1202), and the interactive jobs cannot be executed. Therefore, consumption pattern 3 (1205) does not fall under the early execution pattern for response 1 (1201) and response 2 (1202). Note that the baseline is not uniquely determined, but is checked for each of response 1, response 2, response 3, etc. Consumption pattern 3 does not satisfy the condition, so it is not identified with any response and is considered to be a new job with a predefined new search cost (cost search ) and estimated costs, etc.

[0049] Next, the processes executed by the supply and demand adjusting server 10, the operation plan server 60, and the WL execution management server 70 when creating a coordinated operation plan will be described.

[0050] <First collaborative operation plan creation process> 11 is a diagram illustrating an example of the first coordinated operation plan creation process. This process is started, for example, at time intervals (for example, at midnight every day) or when requested by a related business operator. In this process, the coordinated operation plan is re-planned at the data center 20 in response to a request from the supply and demand adjustment server 10 of the data center 20. At the data center 20 side that has received the re-planning request from the supply and demand adjustment server 10, the coordinated operation plan is re-created through communication between multiple WL execution management servers 70 and the operation plan server 60.

[0051] First, the operation planning server 60 sends a request to create a consumption pattern that will serve as a baseline to the WL execution management server 70 of each business operator (s1). Upon receiving this creation request, the WL execution management server 70 determines the amount of power consumption when executing a workload, taking into consideration only its own convenience, and sends the consumption pattern 700 shown in Fig. 9 to the operation planning server 60 (s2). The operation planning server 60 registers the received consumption pattern 700 (see Fig. 9) in the consumption pattern baseline management table 300 (see Fig. 2).

[0052] Next, when the operation plan server 60 receives a re-planning request from the supply and demand adjustment plan server 10 (s3), it starts a series of operation plan creation processes in cooperation with the WL execution management server 70. Here, examples of re-planning requests include a demand response request and fluctuations in electricity prices.

[0053] The operation planning server 60 executes the cooperative operation plan optimization problem creation process s11. Here, the information includes the control of each device (81, 82, 83, etc. in FIG. 1) owned by the data center 20, the consumption pattern 700 (see FIG. 9) for each DC user 98, and variables indicating whether the consumption pattern 700 corresponds to the early execution pattern for each response in the response history management table. The operation planning server 60 executes the cooperative operation plan optimization problem creation process s11, which uses this information to create an optimization problem that maximizes or minimizes indicators such as costs including the corresponding costs of the consumption patterns of each DC user 98.

[0054] The operation plan server 60 solves the optimization problem created in s11 and executes the operation plan proposal creation process s12 as the optimal solution. In the operation plan proposal creation process s12, the proposed control plan for the equipment used by the DC business operator 96 and the proposed consumption pattern of each DC user 98 are taken into consideration, and a determination is made as to which early execution pattern of a past response the current pattern corresponds to, or does not correspond to, any of the early execution patterns of a past response. The determination of the early execution pattern can be easily performed by referencing the binary variable introduced in the optimization formula update process s15.

[0055] Based on the output of the operation plan creation process s12, the operation plan server 60 sequentially determines for all users whether the plan of the DC user 98 can be considered to be any of the early execution patterns of past responses (s13). If at least one user does not fall into any of the early execution patterns of past responses (NO in s13), the operation plan server 60 transmits the consumption pattern 700 of each DC user 98, which is one of the outputs of the operation plan creation process s12, to the WL execution management server 70 of the corresponding DC user 98 (s4). At this time, for DC users 98 whose consumption pattern to be transmitted falls into any of the early execution patterns of past responses, the corresponding response ID from the response result management table 200 is also transmitted. Thereafter, a series of processes (corresponding cost calculation) starting from s21 are executed. On the other hand, if all users can be considered to be any of the early execution patterns of past responses (YES in all s13), the operation plan server 60 executes a series of processes starting from s6.

[0056] The WL execution management server 70, which has received the consumption pattern 700 from the operation planning server 60, executes a corresponding cost calculation process s21. In the corresponding cost calculation process s21, the consumption pattern 700 (including the response ID) received from the operation planning server 60 is used as input to create a workload execution plan, and the corresponding cost to be requested of the DC user 98 in order to realize the consumption pattern is calculated. In addition, the WL execution management server 70 assigns unique identification information to the received consumption pattern 700 and the corresponding cost calculated in s21, and transmits a response to the operation planning server 60. The operation planning server 60 adds the received content to the response history table 600 (s5).

[0057] The operation planning server 60 executes a job queue calculation process s14 for the received consumption pattern of the DC user 98. In the job queue calculation process s14, the consumption pattern of the corresponding user registered in the consumption pattern baseline management table 300 (see FIG. 2) is compared, and for each consumption pattern, a lower limit of the power consumption corresponding to the amount of workload waiting to be executed in each time block is calculated. Next, the operation planning server 60 executes an optimization formula update process s15. In the optimization formula update process s15, for each new record registered in the response result management table 200 (see FIG. 2), the operation planning optimization formula is updated so that a consumption pattern that can be realized by accelerating its execution (hereinafter referred to as an early execution pattern) can be used at the corresponding corresponding cost.

[0058] Next, the operation plan server 60 transmits the operation plan that satisfies the termination condition to the supply and demand adjustment server 10 of the electricity retailer 2 (s6). The operation plan server 60 transmits the operation plan that satisfies the termination condition as a confirmed consumption pattern in the operation plan to the WL execution management server 70 of each DC user 98 (s7). Based on the received confirmed consumption pattern, the WL execution management server 70 controls the execution of the workload so as to satisfy the consumption pattern (s22).

[0059] If the received plan satisfies the purpose of supply and demand adjustment (YES in s9), the supply and demand adjustment server 10 terminates the processing; if not (NO in s9), the supply and demand adjustment server 10 returns to s3 and executes the processing from s3 onwards to send a re-planning request again in order to achieve the purpose of supply and demand adjustment.

[0060] Next, the details of the processes in s11, s14, s15, and s21 in FIG. 11 will be described.

[0061] <Cooperative operation plan optimization problem creation process> Fig. 12 is a flowchart explaining the detailed procedure of the cooperative operation plan optimization problem processing s11 shown in Fig. 11. First, the cooperative operation plan optimization problem program 65 (see Fig. 2) repeatedly executes the following processing of s201 and s202 for all DC users 98.

[0062] First, in order to handle the consumption patterns of the DC users 98, a variable representing the power consumption of each time block is introduced (s201). Upper and lower limit constraints are added to the introduced variable (s202). These constraints may be set based on the maximum output of the equipment used by the DC users 98, or, if information is provided by the DC users 98, this information may be used. Next, if the DC provider 96 is also a DC user 98, if there is a workload executed by the DC provider 96 itself, this information is reflected (s204). In s204, the processing device 61 adds the execution time of the workload as a variable from the WL management table 500 of the WL execution management server 70 of the DC provider 96, and adds the WL execution delay cost corresponding to that execution time as the objective.

[0063] Next, the processing device 61 reflects the actions that the device can take and the costs for all controllable devices of the DC business operator 96 (s205). The variables added here may include the power consumed by the device in each time block as well as parameters for control such as turning the device's power on / off. This process can be performed by creating variables, constraints, etc. in advance from the devices owned by the data center and reflecting them. Next, the processing device 61 reflects information such as the power price and DR request received from the supply and demand adjustment server, adds constraints to the consumption pattern of the entire data center 20, and adds the power cost as an objective (s206).

[0064] Next, for each record in the response result management table, "Reflected" is changed to "False" (s207). Finally, the optimization formula update (uniform costing of early execution patterns) process (s15) is executed to terminate the optimization problem creation process for the coordinated operation plan. The optimization formula update process s15 is the same as s11 shown in FIG. 11 and s15 described later with reference to FIG. 14. Note that the steps of the coordinated operation plan optimization problem creation method described here are just an example, and the coordinated operation plan optimization problem creation program 65 may be created to perform processing to maximize an index such as the renewable energy utilization rate, for example.

[0065] <Job queue calculation process> FIG. 13 is a flowchart explaining the detailed procedure of the job queue calculation process s14 shown in FIG. 11. The process of FIG. 13 is performed by the processing device 60 of the operation planning server 60 executing the job queue calculation program 66 shown in FIG. 2. First, the job queue calculation program 66 selects one unprocessed response from the newly received responses (s301). Next, for the response selected in s301, the impact calculation program 66 extracts the consumption pattern baseline of the DC user 98 corresponding to that response from the consumption pattern baseline management table 300, and calculates the difference (X diff,t ) is calculated (s302).

[0066] Next, the job queue calculation program 66 calculates the lower limit (Q bt ) and the queue lower bound (Q st ) is calculated by repeating the following process of s304 from time 0 (s303). c = min(Q bt + X diff,t , Q st )year, Q bt+1 =Q bt + X diff,t + c, Q st+1 = Q st + c (s304) This process is equivalent to selecting the pattern that minimizes the amount of job queued in the newly received response from a total of four patterns, depending on whether the two operations were performed on the baseline side or on the newly received response side, on the assumption that the fluctuation in the consumption pattern is due to one of two operations: delaying the execution of the workload and putting it in the job queue, or taking the workload out of the job queue and executing it early.

[0067] Next, the user ID, response ID, cost, consumption pattern, and calculated job queue of the newly received response are added to the response result management table with False set in the reflection column (s305). Next, the job queue calculation program 66 checks whether the processes of s301 to s305 have been executed for all responses (s306). If the processes of s301 to s305 have been executed for all responses (Yes in s306), the processing device 61 (see FIG. 2) ends the job queue calculation process s14. If there is a response for which the processes of s301 to s305 have not been executed (s307: No), the job queue calculation program 66 returns to the process of s301 to select that response, and repeats the processes of S301 to S306.

[0068] <Optimization formula update process> 14 is a flowchart explaining the detailed procedure of the optimization formula update process s15 shown in FIG. 11. First, the optimization formula update program 68 shown in FIG. 2 selects one record from the response result management table 200 for which the reflected column is set to "False" (s401). Next, for the record selected in s401, the optimization formula update program 68 checks the binary variable flg (see FIG. 5) for determining whether the consumption pattern of the user ID 301 of the record in the optimization formula corresponds to the early execution pattern of the consumption pattern of the record. i,j Introduce (s402).

[0069] Next, from the consumption pattern of the record and the job queue, i,j If the condition is 0, a constraint is added so that the consumption pattern of the user ID of the record in the optimization formula is an early execution pattern of the consumption pattern of the record (s403). As an example, the following process is executed.

[0070] (1) In the consumption pattern of the record, the power consumption of the batch job corresponds to the amount of reduction in the job queue (the amount of power taken out of the job queue and executed), and the rest corresponds to the amount of power consumed by the interactive job. i,j" where "*" is the symbol for multiplication.

[0071] (2) For all time points, "the sum of consumption patterns in the optimization formula up to that time point" is "the sum of consumption patterns of the record up to that time point * flg i,j " is added as a constraint that the value is greater than 0.

[0072] By carrying out the above steps, the match determination shown in FIG. 10 becomes possible.

[0073] Next, flg i,j If the value is 1, the consumption pattern is added to the objective so that it can be used by paying the corresponding cost (s404). As an example, the following process is executed. (a) The cost of the record is set as "cost" and the purpose is "cost*flg i,j " Add. (b) A binary variable flg indicating that the user ID of the record does not fall under the early execution pattern of any response consumption pattern. i If flg is not installed, i Introduce new search costs search The purpose is "cost search *flg i " Add. (c) For the user ID of the record, add the constraint "sum of related flags" = 1 so that exactly one related flag is set.

[0074] By carrying out the above steps, the cost of each consumption pattern in Figure 10 can be calculated.

[0075] Next, the reflected column 204 of the record in the response result management table 200 is changed to True (s405). Next, the optimization formula update program 68 checks whether all records in the response result management table 200 have been reflected. If all records have been reflected (Yes in s406), the optimization formula update process s15 ends. If there are records that have not been reflected (No in s406), the optimization formula update program 68 repeats the processes of s401 to s405 to select those records. Note that the optimization formula update method described here is just an example, and the optimization formula update process program 68 may add constraints and objectives to achieve the same effect. Furthermore, the new search cost described here is just an example, and it may be possible to use a predefined constant as in the example, vary it depending on the number of responses, or estimate the corresponding cost depending on the records in the response result management table 200.

[0076] <Corresponding cost calculation process> 15 is a flowchart illustrating the details of the corresponding cost calculation process s21. First, the corresponding cost calculation program 75 determines whether information indicating that the received consumption pattern is an early execution pattern of a past response has been provided (s501). If a response ID that makes the consumption pattern the early execution pattern has been provided (Yes in s501), the program executes the processes from s511 onward to simply calculate the cost. If a response ID has not been provided (No in s501), the program executes the processes from s551 onward to create a workload execution plan that matches the consumption pattern.

[0077] Next, a series of processes from s511 to s513 will be explained. When a response ID is provided, the corresponding cost calculation program 75 shown in Fig. 6 extracts the record of the corresponding response ID 601 from the response history table 600 (see Fig. 8) and compares its consumption patterns to calculate the power consumption (source time, destination time) (s511). Next, among the source times, the batch job that will have the greatest cost reduction due to execution at the destination time is found and moved (s512). Finally, the corresponding cost is calculated from the total sum of the cost changes due to the batch job movement in s512 and the value of the cost 600 of the corresponding record in the response history table 600, and the process ends (s513).

[0078] Next, the series of processes from s551 to s553 will be described. If a response ID is not provided, the correspondence cost calculation program 75 performs workload execution plan optimization. From the WL management table 500 shown in FIG. 7, the execution time of each workload is added as a variable, and the power cost is added as the objective from the execution delay cost 504 corresponding to the execution time 503 (s551). Next, the correspondence cost calculation program 75 adds a constraint to the sum of the power consumption of the execution WL of each time block so as to satisfy the proposed consumption pattern (s552). Finally, the correspondence cost calculation program 75 solves the created optimization problem and obtains the correspondence cost as the optimal value and the execution WL of each time block as the optimal solution (s553). At this time, if a feasible solution is not found, the correspondence cost "inf" is obtained, indicating that the problem is not feasible. Note that the correspondence cost calculation method described here is merely an example, and the correspondence cost calculation program 75 may be any program that can obtain a correspondence cost and a workload execution plan. For example, the correspondence cost may be calculated taking into account an index such as a renewable energy utilization rate, or a method that does not rely on optimization may be used.

[0079] (Management screen for data center operators) FIG. 16 is a diagram showing an example of a management screen for the DC provider 96 displayed on the DC provider terminal 95 (see FIG. 1). The DC provider management screen 1000 includes a consumption pattern display field 1020 for each DC user 98 for the created workload execution plan, and a cost display field 1030 for plan execution. The consumption pattern display field 1020 displays consumption patterns 1021 to 1023 for the DC provider 96 and each DC user 98 in the current operation plan. The horizontal axis represents time, and the vertical axis represents cost (unit: yen). The cost display field 1030 displays a total 1031 of the total of the response costs for each DC user and the total of the DC provider's control cost and power cost, as well as the response cost for each DC user 98 and the amount of variation 1032 from the consumption pattern baseline.

[0080] (Administration screen for DC users) FIG. 17 is a diagram showing an example of a management screen for a DC user 98 displayed on the DC user terminal 97 (see FIG. 1). The DC user management screen 1100 includes a WL (workload) execution plan display field 1120 for the created plan and a response history display field 1130 for the plan creation. The WL execution plan display field 1120 displays an interactive job execution amount 1121 and a batch job execution amount 1122 for each time block in the current workload execution plan. The vertical axis represents the amount of jobs to be executed, and the horizontal axis represents the time period for creating the operation plan—here, six blocks divided into 30-minute intervals from 12:00 to 15:00. In FIG. 17, interactive jobs are shown as open bars, and batch jobs are shown as solid bars above the interactive jobs. The vertical axis represents the power (kWh) required to execute the jobs. Here, it can be seen that batch jobs are executed between 12:30 and 14:30, and that there is a particularly large amount of batch jobs scheduled to be executed between 13:00 and 14:00.

[0081] The response history display field 1130 displays the corresponding costs 1131 received from the data center provider for the current workload execution plan, and a list 1132 of the consumption patterns presented when the coordinated operation plan was created and the corresponding corresponding costs calculated for them. Here, it is displayed that consumption patterns numbered 1 to 3 were received when the optimal coordinated operation plan was created, and it shows that the consumption pattern shown in the third record 1134 at the bottom was adopted. The corresponding cost of 5,000 yen shown in record 1134 is displayed in the explanation field 1131.

[0082] <Renewable Energy Procurement Plan No. 2 Creation Process> 18 is a diagram illustrating an example of the second coordinated operation plan creation process. This process is started, for example, at time intervals (for example, at midnight every day) or when requested by a related business operator.

[0083] First, the operation planning server 60 sends a request to create a consumption pattern that will serve as a baseline to the WL execution management server 70 of each business operator (s1). Upon receiving this creation request, the WL execution management server 70 sends to the operation planning server 60 a consumption pattern 700 for executing a workload taking into consideration only its own convenience (s2). The operation planning server 60 registers the received consumption pattern in the consumption pattern baseline management table 300 shown in FIG. 5.

[0084] Next, the operation planning server 60 sends a request to create a consumption pattern variation plan to the WL execution management server of each business operator (s9). Upon receiving this creation request, the WL execution management server 70 generates multiple feasible consumption patterns from its own WL management table 500 (see FIG. 6) (s23). As an example, the consumption patterns can be generated as the result of randomly allocating each workload within its execution deadline.

[0085] Next, the WL execution management server 70 creates a workload execution plan using each consumption pattern generated in s23 as input, and executes a corresponding cost calculation process s21 to calculate the corresponding cost to be requested of the DC user in order to realize the consumption pattern. The corresponding cost calculation process s21 is executed according to the procedure shown in Fig. 15, and the WL execution management server 70 assigns unique identification information to each consumption pattern and the corresponding cost calculated in s21, sends a response to the operation plan server 60, and at the same time adds the information to the response history table 600 (see Fig. 6) (s5).

[0086] Next, the operation planning server 60, which has received responses from each WL execution management server 70, executes a job queue calculation process (s14). The job queue calculation process (s14) is executed according to the procedure shown in Fig. 13. A workload execution plan is created as described above, and jobs are executed according to that plan. However, if the need arises to re-plan the workload execution plan due to cooperative operation, the supply and demand adjustment planning server 10 requests each operation planning server 60 to re-plan the workload execution plan (s3).

[0087] When the operation planning server 60 receives a re-planning request from the supply and demand adjustment planning server 10 (s3), it starts a series of operation plan creation processes in cooperation with the WL execution management server 70. Examples of re-planning requests include a demand response request and a review request due to fluctuations in electricity prices. The operation planning server 60 executes a coordinated operation plan optimization problem creation process s11 to create an optimization problem that maximizes or minimizes indicators such as costs, including the corresponding costs of the consumption patterns of each DC user 98, using variables including the control of each device owned by the DC business operator 96, the consumption pattern of each DC user 98, and whether each response in the response history management table 600 (see FIG. 8) corresponds to the early execution pattern. The coordinated operation plan optimization problem creation process s11 is performed according to the procedure shown in the flowchart of FIG. 12.

[0088] The operation plan server 60 solves the optimization problem created in s11, and executes the operation plan creation process s12 to output, as the optimal solution, a control plan for the equipment used by the DC business operator 96, a consumption pattern plan for each DC user 98, and whether the early execution pattern corresponds to any of the past responses, or whether the early execution pattern does not correspond to any of the past responses. However, in the <renewable energy procurement plan second creation process>, a new search is not performed when re-planning, so cost search Set to inf to force the early execution pattern of any past response.

[0089] The operation planning server 60 transmits the operation plan that satisfies the termination conditions to the supply and demand adjustment server 10 (s6), and also transmits the operation plan that satisfies the termination conditions to the WL execution management server 70 of each DC user 98 as a confirmed consumption pattern in the operation plan (s7). If the received operation plan satisfies the purpose of supply and demand adjustment (YES in s9), the supply and demand adjustment server 10 terminates processing; if not (NO in s9), the supply and demand adjustment server 10 returns to s3 to send a re-planning request again to achieve the purpose of supply and demand adjustment, and re-executes the processing from s3 onwards. Based on the confirmed consumption pattern received in s7, each WL execution management server 70 controls the execution of the workload so as to satisfy that consumption pattern (s22).

[0090] As described above, according to the embodiment of the present invention, power interchange between DC operators 96 becomes easier in cases such as when dealing with a supply-demand mismatch due to the introduction of renewable energy. Also, user companies (DC users 98) can create operation plans that are tailored to external factors such as the amount of renewable energy supply. Furthermore, when optimizing operation plans, the DC operator 96 proposes changes to consumption patterns while estimating how much the user companies (DC users 98) can change their operation plans, thereby reducing the number of times the DC operator 96 and DC users 98 have to go through the cycle of adjusting consumption patterns and costs.

[0091] The present invention is not limited to the above-described embodiments and can be implemented using any components within the scope of the present invention. The above-described embodiments and modifications are merely examples, and the present invention is not limited to these contents as long as the features of the invention are not impaired. Furthermore, some of the functions provided by each device in this embodiment may be provided in another device, or functions provided in another device may be provided in the same device. Furthermore, the program configuration described in this embodiment is merely an example. For example, part of a program may be incorporated into another program, or multiple programs may be configured as a single program. [Explanation of symbols]

[0092] 1 Operational planning system 2. Electricity retailers 10. Supply and demand adjustment server 20 Data Centers 40 DER operators 50 General business operators 60 Operational Planning Server 70 WL execution management server 81 Server equipment 90 Integrated DC Management System 91 Plan Viewing Server 95 DC business terminal 96 DC operators 97 DC user terminal 98 DC users

Claims

1. An operation plan creation device in a data center having a plurality of server devices that execute workloads of users, a workload execution management server provided in each of the server devices, and an operation plan server connected to the workload execution management servers, the workload execution management server has a cost calculation unit that manages execution of the workload of the user and calculates a corresponding cost for a given consumption pattern; The operation planning server a cooperative operation plan optimization problem creation unit that creates an optimization problem for maximizing or minimizing the index of the response cost taking into account the consumption patterns of the users; an optimization formula update unit that changes the consumption pattern in the past response history from the user according to a certain rule so that the consumption pattern can be used at a corresponding corresponding cost; and an operation plan proposal calculation unit that generates a proposal for an operation plan for the consumption pattern of each user and the equipment held by the data center operator by finding a solution to the updated optimization equation. An operation plan creation device for a data center, comprising:

2. The operation plan server further a job queue calculation unit that compares the consumption pattern of the user with a consumption pattern baseline of the corresponding user that has been registered in advance, and calculates a lower limit of power consumption corresponding to the amount of a queue of jobs waiting to be executed in each time block for each consumption pattern; The operation plan creation device according to claim 1, characterized in that, each time the optimization formula update unit receives a consumption pattern and a corresponding cost from the user, it updates the operation plan optimization formula so that a pattern that can be realized by accelerating the execution of the job queue calculated from the consumption pattern is set as an early execution pattern and becomes available at the corresponding corresponding cost.

3. The operation planning server has a management table for storing workload information of the user and its forecast information, 3. The operation plan creation device according to claim 2, wherein the collaborative operation plan optimization problem creation unit evaluates the cost of workload execution delays and the required energy costs for the consumption patterns presented by the data center operator, and calculates the costs required to address these.

4. The operation plan creation device described in claim 3, characterized in that if the consumption pattern of each user calculated by the operation plan proposal calculation unit is an early execution pattern of a consumption pattern that has been responded to in the past, the corresponding consumption pattern and the calculated job queue are provided to the workload execution management server, thereby simplifying the calculation in the cost calculation unit.

5. the cooperative operation plan optimization problem creation unit executes optimization of the power interchange plan in response to a request from an electricity retailer having a supply and demand adjustment server that plans power interchange among the data center, other data center operators, and other distributed energy resource operators; 5. The operation plan creation device according to claim 4, wherein, when the optimization is performed, the operation plan for the data center is repeatedly recreated until a termination condition for achieving the adjustment of the power interchange is satisfied.

6. 3. The operation plan creation device according to claim 2, wherein the workload execution management server comprises a workload control plan execution unit that controls workload execution of each of the users based on the operation plan.

7. When a re-planning request is received from the operation plan server, if the user's corresponding cost calculation unit is unavailable or the response speed is insufficient, the operation plan proposal calculation unit: The operation plan creation device according to claim 6, characterized in that an optimal operation plan is created by searching only for consumption patterns that can be realized by early execution of job queues calculated for the registered consumption patterns from the multiple consumption patterns registered in advance and their corresponding costs.

8. 1. A method for creating an operation plan in a data center having a plurality of server devices that execute workloads, a workload execution management server provided in each of the server devices, and an operation planning server connected to the workload execution management servers, comprising: The operation plan server, when creating or changing an operation plan for the created workload, Calculating a consumption pattern for each user of the server device and the corresponding cost; creating an optimization formula for maximizing or minimizing the index of the response cost taking into account the consumption patterns of the users of the server device; updating the optimization formula for making it possible to use consumption patterns in the past response history from the user that are varied according to a certain rule at a corresponding corresponding cost; A method for creating an operation plan in a data center, characterized by creating an operation plan for the consumption pattern of each user and the equipment owned by the data center operator by finding a solution to the updated optimization equation.

9. The workload execution management server comparing the consumption pattern of the user with a consumption pattern baseline of a corresponding user registered in advance; For each consumption pattern, a lower limit of power consumption corresponding to the amount of queued jobs waiting to be executed in each time block is calculated and transmitted to the operation planning server; The operational plan creation method described in claim 8, characterized in that each time the operational planning server receives a consumption pattern and a corresponding cost from the workload execution management server, the operational planning server updates the operational plan optimization formula so that an early execution consumption pattern that can be realized by accelerating the execution of a job queue calculated from the user's consumption pattern becomes available at the corresponding corresponding cost.

10. The workload execution management server The operational plan creation method described in claim 9, characterized in that it stores the user's workload information and its forecast information, evaluates the cost due to delayed workload execution and the required energy cost for the consumption pattern presented by the operational plan server, calculates the cost required to respond, and presents the early execution consumption pattern to the operational plan server.

11. The operational plan creation method described in claim 10, characterized in that if the consumption pattern of each user calculated by the operational plan server is an early execution pattern of a consumption pattern that has been responded to in the past, the corresponding consumption pattern and the calculated job queue are provided to the workload execution management server, thereby simplifying the calculation of the corresponding cost.

12. In response to a request from an electricity retailer having a supply and demand adjustment server that plans electricity interchange between the data center, other data center operators, and other distributed energy resource operators, the operation planning server executes optimization of the electricity interchange plan; Sequentially transmitting the optimized power interchange plan to the plurality of workload execution management servers; receiving response costs from the plurality of workload execution management servers and minimizing the response costs; 12. The operation plan creation method according to claim 11, further comprising the step of repeatedly recreating the operation plan for the data center until a termination condition for achieving the adjustment of the power interchange is satisfied.

13. 13. The operation plan creation method according to claim 12, wherein the workload execution management server controls workload execution of each of the users based on the operation plan.

14. The operation plan creation method described in claim 13, characterized in that when the user's corresponding cost calculation unit is unavailable or the response speed is insufficient at the time of a re-planning request from the operation plan server, an optimal operation plan is created by searching from multiple pre-registered consumption patterns and their corresponding costs, with only consumption patterns that can be realized by early execution of job queues calculated for the registered consumption patterns as the search range.

15. a management system including the operation plan creation device according to any one of claims 1 to 7 and a plan viewing server connected to the operation plan creation device via a network, An operation plan optimization system in a data center, characterized in that the plan viewing server provides the operation plan created by the operation plan creation device in accordance with a request from the user via the network and accepts input from the user regarding the operation plan.

Images