Information processing system

Abstract

This information processing system is characterized in that: a predicted value of the amount of increase in power consumption is calculated in a second control device (16) for each of facility groups comprising a plurality of power facilities (11a, 11b) and a plurality of cooling facilities (12a, 12b) that are respectively combined according to an arrangement in a server room (30), from operation information of the power facilities and the cooling facilities in operation; an information processing device having the smallest increase in power consumption is selected in a first control device (15) from among information processing devices that satisfy a processing condition for allocating a new job; a facility group having the minimum amount of increase in power consumption as a result of an increase in power in the selected information processing device is determined in a coordination device (19) from among the facility groups including the selected information processing device; and the new job is allocated to the information processing device in the facility group having the minimum power consumption increase amount. This enables overall power saving management of the power facilities (11a, 11b), the cooling facilities (12a, 12b), and information processing devices (1-1 to 1-n).

Description

Information Processing Systems

[0001] The present disclosure relates to an information processing system.

[0002] With the development of information and communications technology, as exemplified by cloud computing, there is a demand for more advanced and diverse information processing devices, which are part of the infrastructure. Meanwhile, in order to prevent global warming, it is becoming increasingly important to reduce not only the power consumed by information processing devices, but also the overall power consumption, including the power consumed by the equipment that supplies power to each information processing device and the equipment that cools each information processing device. For example, in Patent Document 1, when a workload is assigned to each information processing device, the sum of the power consumption of the information processing device, the power supply equipment, and the air conditioning equipment is calculated for each assignment combination, and all possible assignment combinations for the total number of information processing devices are calculated, and the combination that minimizes the total power consumption is extracted.

[0003] JP 2009-252056 A

[0004] However, in many existing information processing systems, the reduction of the power consumed by the information processing devices, the reduction of the power consumption of the equipment that supplies power to the information processing devices, and the reduction of the cooling equipment that cools the information processing devices are all performed independently, and no processing equipment is provided for comprehensive power reduction. One reason for this is that the vendors that buy and sell information processing devices are different from the vendors that buy and sell power equipment and cooling equipment, and so there has been no thought given to actively managing the reduction of comprehensive power consumption for the entire system.

[0005] Although Patent Document 1 discloses comprehensive power management, it is not a power management configuration that can be incorporated into existing information processing systems. Furthermore, Patent Document 1 searches all combinations for each workload of the information processing device, which requires an enormous amount of time to derive an optimal solution. In addition, the technology of Patent Document 1 has the problem that it is difficult to calculate an optimal solution for power saving from the perspective of power and cooling equipment because the search is not primarily focused on power saving on the power and cooling equipment side.

[0006] The present disclosure has been made to solve the above-mentioned problems, and aims to provide an information processing system that can maximize power saving effects by allocating workloads to information processing devices based on the amount of power that maximizes efficiency on the power and cooling equipment side, even in existing information processing systems in which the information processing devices and the power and cooling equipment are controlled independently of each other.

[0007] The information processing system disclosed herein has power equipment that supplies power to information processing devices in a server room and cooling equipment that cools the server room, and is equipped with a first control device that controls the information processing devices, a second control device that controls the power equipment and cooling equipment, and a coordination device that coordinates the first control device and the second control device, and is characterized in that the second control device calculates a predicted value of the increase in power consumption for each equipment group formed by combining multiple power equipment and multiple cooling equipment according to the arrangement in the server room from operational information of the power equipment and cooling equipment that are in operation, and the first control device selects the information processing device with the smallest increase in power consumption among the information processing devices that satisfy the processing conditions for allocating a new job, and the coordination device determines, from among the equipment groups having the selected information processing device, the equipment group with the smallest increase in power consumption due to the increase in power of the selected information processing device, and allocates the new job to the information processing device in the smallest equipment group.

[0008] In addition, the information processing system disclosed herein has power equipment that supplies power to information processing devices in a server room and cooling equipment that cools the server room, and is equipped with a first control device that controls the information processing devices, a second control device that controls the power equipment and cooling equipment, and a coordination device that coordinates the first control device and the second control device, and is characterized in that the second control device obtains operational information on a rack-by-rack basis within an equipment group formed by combining multiple power equipment and multiple cooling equipment according to the layout within the server room, and the first control device groups live migration groups for each information processing device that can be live migrated based on aggregation conditions, and the coordination device links the information processing devices in the rack with the information processing devices grouped into the live migration group, selects a source information processing device and a destination information processing device based on the power consumption of the linked information processing devices, and executes job migration.

[0009] In addition, the information processing system disclosed herein has power equipment that supplies power to information processing devices in a server room and cooling equipment that cools the server room, and is equipped with a first control device that controls the information processing devices, a second control device that controls the power equipment and cooling equipment, and a coordination device that coordinates the first control device and the second control device, and obtains operational information on an equipment group basis formed by combining multiple power equipment and multiple cooling equipment according to the layout in the server room from the second control device, and the first control device groups live migration groups for each information processing device that can be live migrated based on aggregation conditions, and the coordination device links the information processing devices in the equipment group with the information processing devices grouped into the live migration group, selects a source information processing device and a destination information processing device based on the power consumption of the linked information processing devices, and executes job migration.

[0010] The information processing system of the present disclosure also has power equipment that supplies power to information processing devices in a server room and cooling equipment that cools the server room, and is equipped with a first control device that controls the information processing devices, a second control device that controls the power equipment and cooling equipment, and a linking device that links the first control device and the second control device, and is characterized in that the second control device calculates the increase in power consumption for each equipment group that is selected and combined from multiple information processing devices, multiple power equipment, and multiple cooling equipment depending on the layout in the server room from operational information of the power equipment and cooling equipment that are in operation, and the linking device instructs the first control device to move the job to the information processing device in the equipment group with the smallest increase in power consumption.

[0011] According to the information processing system of the present disclosure, even in existing information processing systems in which the power control of information processing devices and the power control of power and cooling equipment are controlled independently, by connecting a collaborative device, it is possible to allocate workloads to information processing devices based on the amount of power that maximizes the efficiency of the information processing devices and the power and cooling equipment, thereby maximizing the power saving effect.

[0012] 1 is a block configuration diagram of an information processing system according to a first embodiment. FIG. 2 is a diagram illustrating an example of hardware of a first control device, a second control device, and a linkage device of the information processing system according to the first embodiment. FIG. 3 is a functional block diagram of the first control device, the second control device, and the linkage device of the information processing system according to the first embodiment. FIG. 4 is a diagram illustrating an example of the layout of a server room, power equipment, and cooling equipment of the information processing system according to the first embodiment. FIG. 5 is a flowchart illustrating job allocation in the information processing system according to the first embodiment. FIG. 6 is a diagram illustrating an example of the layout of equipment groups in the information processing system according to the first embodiment. FIG. 7 is a diagram illustrating an increase in power consumption for each equipment group relative to an increase in power for each information processing device constituting the information processing system according to the first embodiment. FIG. 8 is a diagram illustrating selection of equipment groups when allocating jobs to information processing devices constituting the information processing system according to the first embodiment. FIG. 9 is a diagram illustrating the cooling efficiency of a rack in the information processing system according to the first embodiment. FIG. 10 is a diagram illustrating the cooling efficiency of an information processing device in a rack in the information processing system according to the first embodiment. FIG. 11 is a functional block diagram of the first control device, the second control device, and the linkage device of the information processing system according to the second embodiment. FIG. 12 is a flowchart illustrating job aggregation in the information processing system according to the second embodiment. FIG. 13 is a diagram illustrating equipment groups for illustrating job aggregation in the information processing system according to the second embodiment. 10 is a diagram illustrating the relationship between the power consumption of each rack of an information processing system according to a second embodiment and the target value and upper limit value. FIG. 11 is a diagram illustrating the execution of live migration between information processing devices in the information processing system according to the second embodiment. FIG. 12 is a functional block diagram of a first control device, a second control device, and a cooperating device in an information processing system according to a third embodiment. FIG. 13 is a flowchart illustrating job aggregation in an information processing system according to the third embodiment. FIG. 14 is a diagram illustrating the relationship between the power consumption of each facility group of an information processing system according to the third embodiment and the target value and upper limit value. FIG. 15 is a diagram illustrating the execution of live migration between information processing devices in an information processing system according to the third embodiment. FIG. 16 is a diagram illustrating the effect of job aggregation in an information processing system according to the third embodiment. FIG. 17 is a functional block diagram of a first control device, a second control device, and a cooperating device in an information processing system according to a fourth embodiment.10 is a flowchart illustrating job aggregation in an information processing system according to embodiment 4. FIG. 11 is a diagram illustrating the relationship between the number of operating information processing devices and the amount of increase in power consumption in an information processing system according to embodiment 4. FIG. 12 is a diagram illustrating execution of live migration in an information processing system according to embodiment 4.

[0013] Hereinafter, preferred embodiments of the information processing system according to the present application will be described with reference to the drawings. Note that the same reference numerals are used to designate the same contents and corresponding parts, and detailed descriptions thereof will be omitted. Similarly, in the following embodiments, redundant descriptions of components with the same reference numerals will be omitted.

[0014] Embodiment 1. Figure 1 is a block diagram of an information processing system 10 according to embodiment 1. The information processing system 10 includes information processing devices 1-1 to 1-n, power facilities 11a and 11b that supply power to the information processing devices 1-1 to 1-n, cooling facilities 12a and 12b that cool the information processing devices 1-1 to 1-n, a first control device 15 connected to the information processing devices 1-1 to 1-n, a second control device 16 connected to the power facilities 11a and 11b and the cooling facilities 12a and 12b, and a linking device 19 that links the first control device 15 and the second control device 16. Here, it is assumed that the first control device 15 and the second control device 16 are devices that exist in an existing information processing system, and the linking device 19 is a device that is connected to these existing devices to achieve comprehensive power savings.

[0015] Jobs, which are workloads, are assigned to the information processing devices 1-1 to 1-n. A job refers to a set of tasks for which instructions are given to an information processing device, such as "browse the web." One job may be assigned to each of the information processing devices 1-1 to 1-n, multiple jobs may be assigned, or no jobs may be assigned to each of the information processing devices 1-1 to 1-n.

[0016] The first control device 15 is connected to the information processing devices 1-1 to 1-n, has location and operation information of the information processing devices 1-1 to 1-n, and calculates the power consumption of each of the information processing devices 1-1 to 1-n. It also allocates jobs to the information processing devices 1-1 to 1-n. It may also perform live migration, which will be described later. The second control device 16 calculates the power loss of the power equipment 11a and 11b and the cooling power of the cooling equipment 12a and 12b. The coordinating device 19 performs processing up to instructing the first control device 15 to allocate jobs to the information processing devices 1-1 to 1-n based on the power consumption calculated by the first control device 15 and the power loss and cooling power calculated by the second control device 16. It also performs live migration, which will be described later, in cooperation with the first control device 15.

[0017] The information processing devices 1-1 to 1-n, the power equipment 11b, and the cooling equipment 12b are disposed in a server room 30. The power monitoring device 13, the cooling monitoring device 14, the first control device 15, the second control device 16, and the linking device 19 are connected to a general-purpose network (hereinafter referred to as a general-purpose NW) 17 and transmit and receive data. The linking device 19 may be directly connected to the first control device 15 and the second control device 16 via a dedicated line. The power monitoring device 13 collects power information and operational data of the power equipment within the server room 30, and the cooling monitoring device 14 collects temperature information and operational data of the cooling equipment within the server room 30. The data from the power monitoring device 13 and the cooling monitoring device 14 is transmitted to the first control device 15 and the second control device 16 via the general-purpose NW 17. A management device 18 may be connected to the general-purpose NW 17, which displays information from the power monitoring device 13 and the cooling monitoring device 14 and information from the first control device 15 and the second control device 16. The management device 18 is composed of, for example, a computer and a display, and extracts data from the multiple information processing systems 10, 10a, and 10b connected to the general-purpose NW 17, analyzes and displays the transitions and trends in power saving of each information processing system, and accumulates this data. When an information processing system is added in the future, the accumulated data, such as information processing devices and power, the arrangement of cooling equipment, and the priority of equipment for power reduction, can be shared to efficiently obtain a power-saving information processing system.

[0018] The information processing devices 1-1 to 1-n are server devices, storage devices, network (NW) devices, etc., and the server devices include general-purpose servers, dedicated servers, mainframes, parallel computers, supercomputers, embedded computers, personal computers, etc. The storage devices include magnetic disks, solid-state disks, optical disks, tapes, etc. The NW devices include routers, switches, hubs, etc., and also peripheral devices such as printers, copiers, modems, and displays. The information processing devices 1-1 to 1-n are stored in multiple racks 101.

[0019] The power equipment 11a is installed outside the server room 30 and includes, for example, a transformer, an uninterruptible power supply (UPS), an air circuit breaker (ACB), a molded case circuit breaker (MCCB), a miniature circuit breaker (MCB), a vacuum circuit breaker (VCB), etc. The power equipment 11b is installed inside the server room 30 and includes, for example, a distribution board, a power distribution board, a power receiving board, a direct digital controller (DDC), and a power supply for the rack 101 on which the information processing devices 1-1 to 1-n are mounted. However, the classification of the models of the power equipment 11a and 11b is not limited to this. Three-phase power, namely, U-phase, V-phase, and W-phase, is supplied to the information processing devices 1-1 to 1-n via the power equipment 11a and 11b.

[0020] The cooling equipment 12a is installed outside the server room 30 and includes, for example, a refrigerator, a cooling tower, an intake and exhaust port, a fan, a duct, refrigerant piping, a liquid cooling device, a power panel, an air handling unit (AHU), etc. The cooling equipment 12b is installed inside the server room 30 and includes, for example, an air handling unit (AHU), an intake and exhaust port, a fan, a duct, refrigerant piping, a direct digital controller (DDC), a variable air volume control device (VAV), a multi-remote control, a chiller, a local cooling device for the front and rear of a fan coil rack row or a rack, etc. However, the classification of the models of the cooling equipment 12a and 12b is not limited to this.

[0021] The power monitoring device 13 includes a power sensor, a current / voltage sensor, a leakage sensor, etc., which are installed inside or outside the server room 30. The cooling monitoring device 14 includes a temperature sensor, a humidity sensor, a flow rate and flow direction sensor, etc., which are installed inside or outside the server room 30.

[0022] The first control device 15 is configured as a computer, and the second control device is configured as a computer or a programmable logic controller (PLC). The control method can be implemented as software such as middleware, an application, embedded control software, or firmware. The coordinating device 19 selects and receives operational information, power consumption, etc. from the second control device 16, and selects and receives aggregation target values, live migration groups, etc. from the first control device 15, and performs the processing described below. Furthermore, to stabilize transmission and reception with the first control device 15 and the second control device 16, the coordinating device 19 may, for example, determine whether transmission and reception between each control device 15 and 16 is necessary. The coordinating device 19 is configured as a computer or a PLC.

[0023] FIG. 2 shows an example of hardware for the first control device 15, the second control device 16, and the linkage device 19. The device is composed of a processor 100 and a storage device 200. Although not shown, the storage device includes a volatile storage device such as a random access memory and a non-volatile auxiliary storage device such as a flash memory. A hard disk auxiliary storage device may be used instead of the flash memory. The processor 100 executes a program input from the storage device 200 to perform the control described below. The program is input to the processor 100 from the auxiliary storage device via the volatile storage device. The processor 100 may output data such as calculation results to the volatile storage device of the storage device 200, or may store the data in the auxiliary storage device via the volatile storage device. Data input / output to / from the processor 100 and the storage device 200 may be performed via a wired or wireless connection via a transmission / reception device 300.

[0024] FIG. 3 is an example of a functional block diagram of the first control device 15, the second control device 16, and the linkage device 19 according to the first embodiment. The first control device 15 includes information processing device processing information acquisition means (hereinafter, processing information acquisition means) 151, information processing device selection means for job processing (hereinafter, information processing device selection means) 152, power consumption calculation means for the selected information processing device (hereinafter, power consumption calculation means) 153, and information processing device job allocation execution means (hereinafter, allocation execution means) 154. The second control device 16 includes equipment group-based operation information acquisition means (hereinafter, operation information acquisition means) 161, equipment group power consumption increase amount calculation means (hereinafter, power consumption increase amount calculation means) 162, and operation information change means 163. The linkage device 19 includes equipment group, rack, and information processing device determination means (hereinafter, determination means) 191. The means 151 to 154, 161 to 163, and 191 shown in the figure may be realized as hardware or a program.

[0025] 4 is a diagram showing an example of the connection arrangement of power equipment 11a, 11b and cooling equipment 12a, 12b arranged inside and outside a server room 30. In FIG. 4, the server room 30 is viewed from above, and multiple racks 101 (30 in the figure) are arranged inside the server room 30, and each rack 101 houses information processing devices. The racks 101 are connected to a PDU 111 for each of the multiple racks to supply power to each information processing device. The PDU 111 is connected to a transformer 113 via a UPS 112.

[0026] A plurality of air conditioners 121 (three on each side in the figure) are arranged on opposing walls of the server room 30, and can cool the entire rack 101. The air conditioners 121 also require power and are therefore connected to a transformer 113 via a UPS 112. The air conditioners 121 are further connected to a refrigerator 122 by refrigerant piping for cooling, and the refrigerator 122 is connected to a cooling tower 123 that generates refrigerant. Job allocation and job aggregation will now be described for the operation of the information processing system 10, which includes the information processing devices, power facilities 11a and 11b, and cooling facilities 12a and 12b, as arranged in Figure 4.

[0027] <Allocation of a New Job> When a new job A is started in the information processing system, to which information processing device is the new job A allocated will be described using Fig. 5. Fig. 5 is a diagram showing the control flow when allocating job A, in which the processing enclosed by the dashed line is mainly performed by the first control device 15 on the information processing device side, the processing enclosed by the dashed line is mainly performed by the second control device 16 on the power and cooling equipment side, and the processing enclosed by the solid line is mainly performed by the coordinating device 19. It goes without saying that the coordinating device 19 exchanges data with the first control device 15 and the second control device 16 and performs processing in cooperation with them.

[0028] The control flow of the second control device 16 on the power and cooling equipment side will be described. First, the operation information acquisition means 161 determines an equipment group (1-1 in FIG. 5). An equipment group is a unit group for calculating the power consumption on the equipment side and is determined by the physical arrangement of the interconnected power equipment 11a, 11b and cooling equipment 12a, 12b. For example, FIG. 6 shows an example of an equipment group determined from the perspective of the flow of cooling air based on the arrangement of the PDU 111 and air conditioners 121 in the server room 30 of FIG. 4. In FIG. 6, equipment groups a to f indicated by solid oval regions are possible. For example, the air conditioner 121a of equipment group a cools the PDU 111a as well as the five racks 101a in the leftmost row and the five racks 101b in the second row from the left in the server room 30. Furthermore, the air conditioner 121b of equipment group b cools the PDU 111a and also cools the five racks 101a in the leftmost row in the server room 30. If the server room 30 has an underfloor air outlet structure, and the cooling air from the air conditioner 121c passes under the floor and is blown out from under the floor near the PDU 111a, it is also possible to consider an equipment group g consisting of dashed line g-1 and dashed line g-2. The PDU 111a and air conditioner 121a of equipment group a are connected to a UPS and transformer outside the server room 30, and these pieces of equipment may be included in the group. The same applies to the other equipment groups.

[0029] In the determined plurality of equipment groups a to g, current operation information of the cooling equipment 12a, 12b and the power equipment 11a, 11b is acquired by the operation information acquisition means 161 from the power monitoring device 13 and the cooling monitoring device 14, etc. (1-1 in FIG. 5). The operation information of the cooling equipment includes, for example, the cooling efficiency curves of air conditioners such as AHUs, the cooling efficiency curves of chillers and cooling towers, and the cooling efficiency inside the server room 30. The operation information of the power equipment also includes the conversion efficiency of the UPS and transformer for the U, V, and W phases, the three-phase load imbalance loss, the wiring loss, etc.

[0030] Next, the power consumption increase calculation means 162 calculates a predicted value of the increase y in power consumption of the power equipment and cooling equipment for each equipment group in response to the power increase x of any information processing device based on the current operation information (1-2 in FIG. 5). The power consumption increase y can be expressed by the following formula (1): y = ζelectric,i(x) + ζ ａｉｒｃｏｎ,ｉ (x) Equation (1) In Equation (1), ζ ｅｌｅｃｔｒｉｃ represents the increase in power consumption of the power equipment due to the power increase x of the information processing device, and is calculated based on any one or a combination of the U, V, and W phase transformer conversion loss, UPS conversion loss, three-phase load imbalance loss, wiring loss, etc., acquired as operational information. ａｉｒｃｏｎ represents the increase in power consumption of the cooling equipment due to the power increase x of the information processing equipment, and is calculated based on any one or a combination of the power consumption of the air conditioner, the power consumption of the refrigerator, the power consumption of the cooling tower, the cooling efficiency in the server room, etc. obtained as operational information. i represents the number (number) of information processing equipment in the equipment group.

[0031] The relationship between the power increase amount x calculated by equation (1) and the power consumption increase amount y is shown in Figure 7. This shows how much the power consumption of each equipment group increases due to the increased power caused by the operation of a specific information processing device. Note that the relationship between the power amount and the power consumption amount may be shown by using the power amount including the power increase amount instead of the power increase amount, and the power consumption amount including the power increase amount instead of the power consumption increase amount.

[0032] Meanwhile, the first control device 15 acquires current processing information of the information processing devices in each equipment group using the processing information acquisition means 151 (2-1 in FIG. 5). This processing information includes the CPU type, clock frequency, number of cores, and communication environment of each information processing device. Then, to process new job A, the minimum requirements are determined for each information processing device, including the current workload status of the CPU (including whether it is running, idle, or stopped), the clock frequency, number of cores, and communication environment required for processing (2-2 in FIG. 5). Based on these minimum requirements, the information processing device selection means 152 selects candidates P, Q, and R as information processing devices capable of processing job A. The candidates P, Q, and R have similar specifications, for example, in terms of network specifications or information processing device specifications (2-3 and 2-4 in FIG. 5). The power consumption x of the candidates P, Q, and R calculated by the power consumption calculation means 153 of the first control device 15 is transmitted to the cooperating device 19 (2-5 in FIG. 5). The decision means 191 of the collaborative device 19 sorts the data in ascending order of increase amount, and selects the candidate P whose power consumption increase amount y is the smallest relative to the power increase amount x of the candidates P, Q, and R, as shown in Figure 8. Furthermore, if an information processing device corresponding to candidate P exists in both equipment groups a and b, that is, if an information processing device of candidate P that has the same clock frequency, number of cores, communication environment, etc. and can process job A exists in both equipment groups a and b, the decision means 191 compares the power consumption increase amount (●) of equipment group a with the power consumption increase amount (×) of equipment group b based on the power increase amount when candidate P processes the job, as shown in Figure 8, and determines that the equipment group a that can minimize the power consumption increase should be used for the processing (3-1 in Figure 5). Note that the equipment group with the smallest power consumption may be selected by including the power consumption of each candidate information processing device in the power consumption increase amount (y).

[0033] If there are multiple candidate P information processing devices in the equipment group a determined by the determination means 191, for example, if the candidate P information processing device is present in each of racks 1 to 3, the determination means 191 further calculates the cooling efficiency of the candidate P information processing device and the cooling efficiency of the rack in which the candidate P information processing device is stored, and ultimately determines the information processing device to be used for processing (3-2 in FIG. 5). In other words, the determination means 191 of the linkage device 19 determines the rack and information processing device positions that minimize equipment efficiency from the multiple candidate information processing devices in the equipment group. Details are explained below.

[0034] First, the cooling efficiency of the rack in which the information processing device of candidate P is stored will be described. Fig. 9A is a perspective view of equipment group a in Fig. 6. The cooling efficiency γ for each of the racks 101a and 101b corresponding to the PDU 111a and the air conditioner 121a is ｒａｃｋ is, for example, the air blowing loss γ ａｉｒ , and the airflow loss γ around the racks 101a and 101b ｆｌｏｗ＿ｒａｃｋ The cooling efficiency is expressed as the magnitude of loss, so the calculated γ ｒａｃｋ The smaller the value of γ, the higher the cooling efficiency. ｒａｃｋ = γ ａｉｒ +γ ｆｌｏｗ＿ｒａｃｋ Equation (2) When the calculated cooling efficiency of the racks is represented as a graph in FIG. 9B, rack 3 has the best cooling efficiency, so rack 3 is selected (means 191 in FIG. 3, 3-2 in FIG. 5).

[0035] If all the information processing devices in the rack 3 shown in FIG. 10A satisfy the minimum requirements for the information processing device of the candidate P, the cooling efficiency γ ｓｅｒｖｅｒ is, for example, the air blowing loss γ ａｉｒ , airflow loss around the information processing device γ ｆｌｏｗ＿ｓｅｒｖ The cooling efficiency is expressed as the magnitude of loss, so the calculated γ ｓｅｒｖｅｒ The smaller the value of γ, the higher the cooling efficiency. ｓｅｒｖｅｒ = γ ａｉｒ +γ ｆｌｏｗ＿ｓｅｒｖ10(b), the calculated cooling efficiency of the information processing devices is shown as the graph, and it is determined that the information processing device 3 in rack 3 is selected as the information processing device to process job A (means 191 in FIG. 3, 3-2 in FIG. 5). The determined items are transmitted to the first control device 15.

[0036] In accordance with the instruction received from the coordinating device 19, the allocation execution means 154 of the first control device 15 allocates job A to the information processing device 3 of rack 3 (2-6 in FIG. 5). At the same time, since the allocation of job A causes a change in the control of the power equipment 11a, 11b and the cooling equipment 12a, 12b, this change is transmitted from the coordinating device to the second control device 16, and is fed back to the operation information by the operation information change means 163 (3-3 in FIG. 5).

[0037] As described above, in the information processing system of the first embodiment, by connecting the cooperating device 19 to the first control device 15 and the second control device 16, it is possible to allocate jobs to information processing devices in an equipment group that can minimize power consumption based on the increase in power consumption of the power and cooling equipment of the equipment group, thereby suppressing the increase in power consumption of the power and cooling equipment. This enables highly energy-efficient information processing.

[0038] Embodiment 2. <Method of Consolidating Allocated Jobs by Live Migration (within an Equipment Group)> An example of a method of consolidating jobs allocated to each information processing device by live migration according to embodiment 1 will be described with reference to FIGS. 11 and 12 . First, it will be described that live migration, in which jobs are migrated and consolidated within each equipment group, can be performed by connecting a cooperating device 19 to the first control device 15 and the second control device 16. Here, live migration refers to the transfer of a job to another information processing device without stopping the information processing device currently executing the information processing of the job. To perform live migration, virtualization technology is used that abstracts the CPU and storage devices (memory, disks, etc.) and enables software integration or division without being bound by physical limitations. Note that this embodiment may be used when, after an allocated job is completed, deviations from an ideal operating state with minimal power loss in the equipment become significant.

[0039] 11 is an example of a functional block diagram of the first control device 15, the second control device 16, and the coordinating device 19 according to the second embodiment. The first control device 15 includes an aggregation target value and upper limit value selection means (hereinafter, "selection means") 251, a live migration group selection means 252, and a live migration execution means 253. The second control device 16 includes a power consumption target value and upper limit determination means (hereinafter, "determination means") 261, an operation information acquisition means 262, and an operation information change means 263. The coordinating device 19 includes a rack priority order determination means 291, a power consumption amount calculation means 292, an information processing device priority order determination means 293, a live migration group linking means (hereinafter, "linking means") 294, a rack migration feasibility determination means 295, and a migration destination information processing device selection means 296. Each of these means may be implemented as hardware or a program.

[0040] 12 is a diagram showing the control flow of this embodiment, in which the processing enclosed by the dashed line is mainly performed by the first control device 15 on the information processing device side, the processing enclosed by the broken line is mainly performed by the second control device 16 on the power and cooling equipment side, and the processing enclosed by the solid line is mainly performed by the linkage device 19. It goes without saying that the linkage device 19 exchanges data with the first control device 15 and the second control device 16 and performs processing in cooperation with them.

[0041] First, the target value and upper limit of power consumption for each rack are determined by the determining means 261 of the second control device 16 based on the current operation information of each equipment group (4-1 in FIG. 12). This power consumption may include the power consumption of each information processing device in the rack.

[0042] In this embodiment, as shown in FIG. 13, racks 1 to 10 in the same equipment group a as in the first embodiment will be used as an example. The operation information for the equipment group is the same as that described in the first embodiment. The rack priority determination means 291 of the linkage device 19 assigns priorities to racks 1 to 10 as shown in FIG. 14 (FIGS. 12, 6-1). The highest priority is assumed to be a highly efficient rack, and the lowest priority is assumed to be a low-efficiency rack. For example, a rack whose power consumption exceeds the upper limit (rack 2 in FIG. 14) has the lowest priority, followed by a rack whose power consumption does not meet the target value and has a large difference from the target value (rack 1 in FIG. 14). Furthermore, a rack whose power consumption is above the target value but below the upper limit, such as rack 9 in FIG. 14, is selected as the rack to which the job is to be transferred, but does not have the highest priority.

[0043] The selection means 251 of the first control device 15 determines the aggregation conditions and aggregation target values ​​(5-1 in FIG. 12). The aggregation target value is a target value when aggregating information processing devices with low job processing workloads, and is determined to aggregate workloads so that they fall within the target value band of the aggregation destination or do not fall below that target value. The aggregation conditions include CPU load rate, memory usage rate, and IO load, and if there are multiple aggregation conditions, the aggregation target values ​​are set individually for each aggregation condition. For example, a CPU load rate may be specified as a range, such as 40% to 60%.

[0044] Furthermore, the live migration group selection unit 252 groups the live migration groups for the information processing devices for each rack (5-2 in FIG. 12). As an example of grouping, the range in which live migration is possible may be determined based on the physical situation and environment, such as whether the storage of the information processing devices is shared or compatible, or the information processing devices may be divided into groups based on the purpose of use, such as DNS (Domain Name System) servers, FTP (File Transfer Protocol) servers, and Web servers.

[0045] The linking means 294 of the coordinating device 19 links this live migration grouping to each information processing device in each rack of each equipment group (6-2 in FIG. 12). If the rack with the lowest priority determined by the rack priority determination means 291 is rack 2 exceeding the upper limit of power consumption (6-1 in FIG. 12), the power consumption calculation means 292 calculates the amount of power consumption reduction required to meet the target power consumption value for each rack (6-4 in FIG. 12). Furthermore, among the information processing devices in the lowest-priority rack 2, the information processing device priority determination means 293 prioritizes them to determine the source information processing device to which the job is to be migrated by live migration (6-5 in FIG. 12), and the job is migrated from the source information processing device in order of priority (5-3 in FIG. 12). Prioritization methods include descending order of power consumption or descending order of processing time. Priority may also be restricted by the number of jobs migrated or the amount of power consumption.

[0046] Furthermore, if the lowest-level rack is rack 1, which has a large difference from the target value, the rack migration capability determination means 295 of the collaboration device 19 determines whether the workload of the operating or idle information processing devices in rack 1 can be migrated to another rack (6-3 in FIG. 12). Furthermore, among the information processing devices in the lowest-level rack 1, the information processing device priority determination means 293 assigns priorities to determine the source information processing device for moving jobs by live migration (6-5 in FIG. 12). Priorities can be assigned in descending order of power consumption or longest processing time, for example. Priorities may also be restricted by the number of workloads to be migrated or the amount of power consumption. If jobs from operating or idle information processing devices in rack 1 cannot be migrated to another rack, the lowest-level rack 1 may be changed to the highest-level rack (6-8 in FIG. 12).

[0047] The power consumption calculation means 292 of the coordinating device 19 calculates the power consumption up to the target value for the highest-priority, highly efficient rack among racks 1 to 10, e.g., rack 3 in FIG. 14 (see 6-4 in FIG. 12). The destination information processing device selection means 296 determines whether there is a destination information processing device in rack 3 that can process jobs with the calculated power consumption or a power consumption close to that calculated. The destination information processing device selection means 296 determines whether there is a destination information processing device that is grouped in the same live migration group A as the source information processing device x in rack 2 (see 6-7 in FIG. 12). If it is determined that information processing device y corresponds to the destination information processing device, the live migration execution means 253 of the first control device 15 executes live migration to migrate the job from information processing device x to information processing device y, as shown in FIG. 15 (see 5-4 in FIG. 12). If a destination information processing device cannot be found in rack 3, the highest priority of rack 3 may be lowered by one (see 6-9 in FIG. 12). In this case, the priority may be changed to the rack 10 with the second highest priority, and the above-described determination up to the execution of live migration may be performed again.

[0048] After the live migration of jobs from information processing device x to information processing device y is completed, the job consolidation causes changes in the control of the power equipment 11a, 11b and the cooling equipment 12a, 12b, and these changes are fed back to the operational information (see 4-3 and 4-4 in FIG. 12). Furthermore, the lowest-order rack 2 may be changed to the highest-order rack due to the job consolidation (see 6-8 in FIG. 12). This change lowers the priority of the previous highest-order rack by one (see 6-9 in FIG. 12).

[0049] In a similar manner, live migration may be executed to migrate a job from a source information processing device v in rack 1 to a destination information processing device w grouped in the same live migration group B.

[0050] Furthermore, jobs from the source information processing device may be divided and migrated, or the divided jobs may be consolidated and migrated. Possible conditions for ending live migration include the point at which job migration has been completed within all equipment groups, the point at which there are no jobs left in the source information processing device, or the point at which live migration has been performed a predetermined number of times within the same equipment group. However, live migration may be performed continuously and no end point may be determined. Jobs may also be migrated across equipment groups. After job consolidation through live migration, information processing devices that are no longer needed may be shut down, and the operation of the equipment that is no longer needed may also be shut down accordingly.

[0051] As described above, in the information processing system of embodiment 2, by connecting the linking device 19 to the first control device 15 and the second control device 16, it is possible to aggregate jobs within an equipment group by live migration, thereby making it possible to reduce power consumption for each equipment group.

[0052] Third Embodiment <Method of consolidating allocated jobs by live migration (between facility groups)> Next, a method of performing live migration of jobs between facility groups will be described with reference to Fig. 16. Note that this embodiment may be used when, after an allocated job is completed, deviation from an ideal operating state with little power loss of the facility becomes significant.

[0053] FIG. 16 is an example of a functional block diagram of the first control device 15, the second control device 16, and the coordinating device 19 according to the second embodiment. The same reference numerals are used to denote units having the same functions as those in the second embodiment. The first control device 15 includes an aggregation target value and upper limit value selection unit (hereinafter, "selection unit") 251, a live migration group selection unit 252, and a live migration execution unit 253. The second control device 16 includes a power consumption target value and upper limit determination unit (hereinafter, "determination unit") 261, an operational information acquisition unit 262, and an operational information change unit 263. The coordinating device 19 includes a group priority determination unit 391, a power consumption amount calculation unit 392, an information processing device priority determination unit 393, a live migration group linking unit (hereinafter, "linking unit") 394, a group migration feasibility determination unit 395, and a migration destination information processing device selection unit 396. Each of these units may be implemented as hardware or a program.

[0054] 17 is a diagram showing the control flow of this embodiment, in which the processing enclosed by the dashed line is mainly performed by the first control device 15 on the information processing device side, the processing enclosed by the dashed line is mainly performed by the second control device 16 on the power and cooling equipment side, and the processing enclosed by the solid line is mainly performed by the linkage device 19. It goes without saying that the linkage device 19 exchanges data with the first control device 15 and the second control device 16 and performs processing in cooperation with them.

[0055] The target value and upper limit value of power consumption for each equipment group are determined by the determining means 261 of the second control device 16 based on the current operation information of each equipment group (7-1 in FIG. 17). This power consumption may include the power consumption of each information processing device in the equipment group.

[0056] Here, the equipment groups a to g described in the first embodiment will be used as an example. The operational information for each equipment group is the same as that described in the first embodiment. The group priority determination means 391 of the collaboration device 19 assigns priorities to the equipment groups a to g as shown in FIG. 18 (FIG. 17, 9-1). The highest priority is assumed to be a highly efficient equipment group, and the lowest priority is assumed to be a low-efficiency equipment group. For example, an equipment group whose power consumption exceeds the upper limit (equipment group b in FIG. 18) has the lowest priority, followed by an equipment group whose power consumption does not meet the target value and has a large difference from the target value (equipment group a in FIG. 18). Furthermore, an equipment group whose power consumption is above the target value but below the upper limit is selected as an equipment group to which a job is transferred, but does not fall under the highest priority equipment group.

[0057] The selection means 251 of the first control device 15 determines the aggregation conditions and aggregation target values ​​(8-1 in FIG. 17). The aggregation target value is a target value when aggregating information processing devices with low job processing workloads, and is determined to aggregate workloads so that they fall within the target value band of the aggregation destination or do not fall below that target value. The aggregation conditions include CPU load rate, memory usage rate, and IO load, and if there are multiple aggregation conditions, the aggregation target value is set for each aggregation condition. For example, a CPU load rate may be specified as a range, such as 40 to 60%.

[0058] Furthermore, the live migration group selection means 252 groups the live migration groups of information processing devices for each facility group (8-2 in FIG. 17). As an example of grouping, the range in which live migration is possible may be determined based on the physical situation and environment, such as whether the storage of the information processing devices in each facility group is shared or compatible, or the information processing devices may be divided into groups based on their intended use, such as DNS (Domain Name System) servers, FTP (File Transfer Protocol) servers, and Web servers.

[0059] The linking means 394 of the coordinating device 19 links the live migration group to each information processing device in each equipment group (9-2 in FIG. 17). If the group priority determination means 391 determines that the equipment group with the lowest priority is equipment group b, which exceeds the upper limit of power consumption (9-1 in FIG. 17), the power consumption calculation means 392 calculates the amount of power consumption reduction required to meet the target power consumption value for each equipment group (9-4 in FIG. 17). Furthermore, the information processing device priority determination means 393 prioritizes the information processing devices in the lowest equipment group b to determine the source information processing device to which the job is to be migrated by live migration (9-5 in FIG. 17), and the job is migrated from the source information processing device in order of priority (8-3 in FIG. 17). Prioritization methods include descending order of power consumption or descending order of processing time. Note that the priority may be restricted by the number of jobs migrated or the amount of power consumption.

[0060] Furthermore, if the lowest-level equipment group is equipment group a, which has a large difference from the target value, the group migration possibility determination means 395 determines whether the workload of the information processing devices in equipment group a that are in operation or idle can be migrated to another equipment group (9-3 in FIG. 17). Furthermore, among the information processing devices in the lowest-level equipment group, the information processing device priority determination means 393 assigns priorities to determine the source information processing device for moving jobs by live migration (9-5 in FIG. 17). Priorities can be assigned in descending order of power consumption or longest processing time, for example. Priority restrictions may be imposed based on the number of workload migrations or the amount of power consumption. Furthermore, if jobs of information processing devices in equipment group a that are in operation or idle cannot be migrated to another equipment group, equipment group a may be changed to the highest-level equipment group (9-8 in FIG. 17).

[0061] For the equipment group with the highest priority and highest efficiency among the equipment groups a to g, such as equipment group c in FIG. 18, the power consumption calculation means 392 calculates the power consumption up to the target value (9-4 in FIG. 17). Among the information processing devices in equipment group c that can process jobs with the calculated power consumption or a power consumption close to that calculated, the destination information processing device selection means 396 determines whether there is a destination information processing device grouped in the same live migration group A as the source information processing device x in the above-mentioned equipment group b (9-7 in FIG. 17). As shown in FIG. 19, if it is determined that information processing device y corresponds to the destination information processing device, live migration is performed to migrate the job from information processing device x to information processing device y (8-3 in FIG. 17). If a destination information processing device cannot be found within equipment group c, the highest priority of equipment group c may be lowered by one (9-9 in FIG. 17). In this case, the priority may be changed to equipment group f, which has the second highest priority, and the above-mentioned determination of whether to execute live migration may be performed.

[0062] After the execution of live migration, which transfers jobs from information processing device x to information processing device y, is completed, the job aggregation causes control changes in the power equipment 11a, 11b and the cooling equipment 12a, 12b, and this change is fed back to the operation information (7-3, 7-4 in FIG. 17). Furthermore, the lowest-level equipment group may be changed to the highest-level equipment group due to the job aggregation (9-8 in FIG. 17). This change lowers the priority of the previous highest-level equipment group by one (9-9 in FIG. 17).

[0063] In a similar manner, live migration may be executed to migrate a job from a source information processing device v of equipment group a to a destination information processing device w of equipment group c grouped in the same live migration group B.

[0064] Furthermore, jobs from the source information processing device may be divided and migrated, or the divided jobs may be consolidated and migrated. Possible conditions for ending live migration include the point at which job migration has been completed in all equipment groups, the point at which there are no more jobs in the source information processing device, or the point at which a job has been executed a predetermined number of times within the same equipment group, but live migration may be performed continuously and no end point may be determined. After job consolidation through live migration, information processing devices that are no longer needed may be shut down, and the operation of the equipment that is no longer needed may also be shut down accordingly.

[0065] As described above, in the information processing system of the third embodiment, by connecting the cooperating device 19 to the first control device 15 and the second control device 16, jobs can be consolidated between facility groups by live migration, enabling power consumption to be reduced on a facility group basis. For example, it is possible to consolidate jobs from the 13 operating racks shown in FIG. 20(a) to the five operating racks shown in FIG. 20(b). This prevents low-load information processing devices from being scattered and reduces power loss in the facility.

[0066] Fourth Embodiment <Method of Allocating and Aggregating Jobs by Live Migration (Zero-Based)> Next, a method of performing so-called zero-based live migration will be described with reference to Figures 21 and 22, in which jobs are allocated and aggregated by assuming the amount of power consumption of an equipment group relative to an aggregation target value in a state in which no jobs are being processed, without taking into account the current job processing status, using a configuration in which a cooperating device 19 is connected to the first control device 15 and the second control device 16. This method combines, for example, the job allocation described in the first embodiment with the job aggregation described in the second or third embodiment.

[0067] FIG. 21 is an example of a functional block diagram of the first control device 15, the second control device 16, and the coordinating device 19 according to the fourth embodiment. The same reference numerals are used to denote units having the same functions as those in the second embodiment. The first control device 15 includes an aggregation target value and upper limit value selection means (hereinafter, "selection means") 251, a live migration group selection means 252, and a live migration execution means 253. The second control device 16 includes an allocation destination information processing device order setting means (hereinafter, "order setting means") 461, a power consumption increase amount calculation means for each facility group (hereinafter, "power consumption increase amount calculation means") 462, and an operation information change means 263. The coordinating device 19 includes an information processing device priority determination means 491, a live migration group linking means 492, and a migration source information processing device selection means (hereinafter, "selection means") 493. Each of these means may be implemented as hardware or a program. FIG. 23 will be described later together with the description of FIG. 22.

[0068] 22 is a diagram showing the control flow of this embodiment, in which the processing enclosed by the dashed line is mainly performed by the first control device 15 on the information processing device side, the processing enclosed by the dashed line is mainly performed by the second control device 16 on the power and cooling equipment side, and the processing enclosed by the solid line is performed by the linkage device 19. It goes without saying that the linkage device 19 exchanges data with the first control device 15 and the second control device 16 and performs processing in cooperation with them.

[0069] First, the selection means 251 of the first control device 15 determines the aggregation conditions and aggregation target value (11-1 in FIG. 22). The aggregation target value is a target value when aggregating information processing devices with low job processing workloads, and is determined to aggregate workloads so that they fall within the target value band of the aggregation destination or do not fall below that target value. The aggregation conditions include CPU load rate, memory usage rate, and IO load, and when there are multiple aggregation conditions, the aggregation target value is set individually for each aggregation condition. For example, a CPU load rate may be specified as a range, such as 40 to 60%. In this embodiment, the aggregation target value is set to 40% CPU load rate.

[0070] Furthermore, the live migration group selection means 252 of the first control device 15 groups the live migration groups of information processing devices for each facility group (11-2 in FIG. 22). As an example of grouping, the range in which live migration is possible may be determined based on the physical situation and environment, such as whether the storage of the information processing devices in each facility group is shared or compatible, or the information processing devices may be divided into groups based on their intended use, such as DNS (Domain Name System) servers, FTP (File Transfer Protocol) servers, and Web servers.

[0071] The order setting means 461 of the second control device 16 determines in advance for each equipment group the optimal order of the information processing devices to be assigned that minimizes power consumption for the set aggregation target value of 40% CPU load factor (10-1 in FIG. 22). When determining the order, consideration is given to the cooling efficiency of the air conditioning according to the physical arrangement of the information processing devices, three-phase load balancing, and power distribution loss of the power equipment.

[0072] The power consumption increase calculation means 462 of the second control device 16 calculates the value of the power consumption increase of the equipment group including the power equipment and cooling equipment when the information processing equipment are operated sequentially in accordance with a predetermined order of assignment destination information processing equipment (10-2 in FIG. 22). The power consumption increase may include the power consumption of the information processing equipment. The increase in power consumption of the power equipment and cooling equipment may refer to equation (1) described in the first embodiment. An example of the relationship between the number of operating information processing equipment and the power consumption increase of the equipment group is shown in FIG. 23.

[0073] The priority determination means 491 compares the power consumption increase values ​​of each equipment group, prioritizes the equipment groups and information processing devices in ascending order of power consumption increase, and sequentially sets them as the destination information processing devices (12-1 in FIG. 22). For example, if equipment group b operates one information processing device and equipment group c operates zero devices, the power consumption increase is compared between equipment group b when the second device is operating and equipment group c when the first device is operating.

[0074] The selection unit 493 selects a source information processing device from the same live migration group that approaches the aggregate target value of 40% CPU load factor for the destination information processing device with the highest priority and the smallest predicted power consumption increase value (12-2 in FIG. 22). Live migration is performed from the selected source information processing device to the destination information processing device with the highest priority (11-4 in FIG. 22).

[0075] 24 shows that for a destination information processing device x with the smallest increase in power consumption in equipment group b and the highest priority, a source information processing device y with a CPU load factor of 30%, close to the consolidation target value, is selected from the same live migration group A (see FIG. 24(a)), and live migration is performed to consolidate jobs (see FIG. 24(b)). By sequentially performing such a procedure, it is possible to consolidate jobs from, for example, the 13 operating racks shown in FIG. 20(a) already explained to five operating racks as shown in FIG. 20(b) more quickly than with the method shown in embodiment 3.

[0076] If it is not possible to select a source information processing device from the same live migration group that approaches the aggregation target value of 40% CPU load rate, the priority of all information processing devices in the equipment group to which the information processing device with the highest priority belongs can be set to the lowest, and the prioritization process can be performed again (12-4 in Figure 22).

[0077] Consolidation of jobs through live migration based on predicted values ​​results in control changes for the information processing devices, power equipment 11a, 11b, and cooling equipment 12a, 12b. It is determined whether operation under these changed control conditions is in line with the expected operating conditions. For example, it is determined based on actual measurement data whether overcooling, insufficient cooling, three-phase load imbalance, equipment efficiency, etc. are occurring (10-3, 10-4 in FIG. 22).

[0078] If the result of performing live migration exceeds the upper limit of the aggregation target value of the destination information processing device, a group of equipment that can be migrated from the same live migration group may be set and migration may be performed again. Also, in order to satisfy the aggregation target value of the destination information processing device, jobs of the source information processing device may be divided and live migration may be performed to the destination information processing device. Also, multiple destination information processing devices may be selected and live migration may be performed. Note that, although the amount of increase in power consumption is compared in the above, the amount of power consumption may be compared instead of the amount of increase in power consumption.

[0079] Possible conditions for ending live migration include the point at which job migration has finished within all equipment groups, the point at which there are no more jobs in the source information processing device, or the point at which live migration has been performed a predetermined number of times within the same equipment group, but live migration may be performed continuously without determining an end point.Furthermore, after job consolidation through live migration, information processing devices that are no longer needed may be shut down, and the operation of equipment that is no longer needed may also be stopped.

[0080] As described above, in the information processing system of the fourth embodiment, by connecting the cooperating device 19 to the first control device 15 and the second control device 16, it is possible to simultaneously allocate and aggregate jobs and reduce power consumption by assuming the power consumption amount in an ideal state of the equipment group relative to the aggregation target value when no jobs are being processed, without taking into account the current job processing status. This prevents low-load information processing devices from being scattered, and makes it possible to reduce power loss in the equipment while minimizing time loss.

[0081] Although various exemplary embodiments and examples are described in this disclosure, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless variations not illustrated are anticipated within the scope of the technology disclosed in this specification. For example, this includes cases where at least one component is modified, added, or omitted, or where at least one component is extracted and combined with components of another embodiment.

[0082] 1-1 to 1-n: information processing devices, 10, 10a, 10b: information processing systems, 11a, 11b: power equipment, 12a, 12b: cooling equipment, 13: power monitoring device, 14: cooling monitoring device, 15: first control device, 16: second control device, 17: general-purpose network, 18: management device, 19: linkage device, 30: server room, 100: processor, 101, 101a, 101b: rack, 111, 111a: PDU, 112: UPS, 113: transformer, 121, 121a, 121b, 121c: air conditioner, 122: refrigerator, 123: cooling tower, 200: storage device.

Claims

1. An information processing system having power equipment that supplies power to information processing devices in a server room and cooling equipment that cools the server room, comprising a first control device that controls the information processing devices, a second control device that controls the power equipment and the cooling equipment, and a coordinating device that coordinates the first control device and the second control device, wherein the second control device calculates a predicted increase in power consumption for each equipment group formed by combining a plurality of the power equipment and a plurality of the cooling equipment in accordance with the arrangement in the server room from operational information of the power equipment and the cooling equipment that are in operation, and the first control device selects the information processing device with the smallest increase in power consumption among the information processing devices that satisfy the processing conditions for allocating a new job, and among the equipment groups having the selected information processing device, the equipment group with the smallest increase in power consumption due to the power increase of the selected information processing device is determined by the coordinating device, and the information processing system allocates the new job to the information processing device in the smallest equipment group.

2. An information processing system having power equipment for supplying power to information processing devices in a server room and cooling equipment for cooling the server room, comprising a first control device for controlling the information processing devices, a second control device for controlling the power equipment and the cooling equipment, and a coordinating device for coordinating the first control device and the second control device, wherein operation information on a rack-by-rack basis within an equipment group formed by combining a plurality of the power equipment and a plurality of the cooling equipment is obtained from the second control device according to the arrangement within the server room, and the first control device groups live migration groups for each information processing device that can be live migrated based on aggregation conditions, and the coordinating device links the information processing devices in the rack with the information processing devices grouped into the live migration group, selects a source information processing device and a destination information processing device based on the power consumption of the linked information processing devices, and executes job migration.

3. An information processing system having power equipment for supplying power to information processing devices in a server room and cooling equipment for cooling the server room, comprising a first control device for controlling the information processing devices, a second control device for controlling the power equipment and the cooling equipment, and a coordinating device for coordinating the first control device and the second control device, wherein operation information for each equipment group formed by combining a plurality of the power equipment and a plurality of the cooling equipment according to the arrangement in the server room is obtained from the second control device, and the first control device groups live migration groups for each information processing device capable of live migration based on aggregation conditions, and the coordinating device links the information processing devices in the equipment group with the information processing devices grouped into the live migration group, selects a source information processing device and a destination information processing device based on the power consumption of the linked information processing devices, and executes job migration.

4. An information processing system having power equipment that supplies power to information processing devices in a server room and cooling equipment that cools the server room, comprising a first control device that controls the information processing devices, a second control device that controls the power equipment and the cooling equipment, and a coordination device that coordinates the first control device and the second control device, wherein the second control device calculates an increase in power consumption for each equipment group formed by selecting and combining a plurality of the information processing devices, a plurality of the power equipment, and a plurality of the cooling equipment in accordance with the layout in the server room from operational information of the power equipment and the cooling equipment while in operation, and the coordination device instructs the first control device to move jobs to the information processing devices in the equipment group with the smallest increase in power consumption.

5. The information processing system according to claim 2, wherein the information processing device to which the job is to be moved and the information processing device from which the job is to be moved have the same physical environment or the same purpose of use.