Data center computer room planning method, device, equipment and storage medium

Abstract

The embodiments of this specification provide a data center computer room planning method, device, equipment and storage medium, the method comprising: obtaining parameter values ​​of multiple target parameters of a data center to be built, and forming a parameter value set; inputting the parameter value set into a PUE prediction model to obtain a PUE prediction value of the data center to be built; when the PUE prediction value is not lower than the PUE threshold, adjusting the parameter values ​​in the parameter value set, and re-predicting based on the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; determining the parameter value set corresponding to the PUE prediction value lower than the PUE threshold as a target set; determining the computer room planning data of the data center to be built based on the target set. The embodiments of this specification can reduce the PUE value of a data center.

Description

Technical Field

[0001] The present invention relates to the field of big data processing technology, and in particular to a data center computer room planning method, device, equipment and storage medium. Background Art

[0002] Power Usage Effectiveness (PUE) is an indicator for evaluating the energy efficiency of a data center. It is the ratio of all energy consumed by a data center to the energy consumed by IT equipment (or IT load). Therefore, reducing the PUE value of a data center is of great significance for improving the energy efficiency level of a data center. In the prior art, the solution to reducing the PUE value of a data center is mainly to reduce the PUE value of a data center by optimizing the operating parameters of the data center computer room after the data center computer room is built and put into operation.

[0003] In fact, the site selection, area layout, selection of air conditioning equipment and UPS system equipment of the data center computer room have a great impact on the PUE value of the data center. Therefore, the reasonable planning of the data center computer room is also of great significance to reducing the PUE value of the data center. At present, the planning of the data center computer room is generally determined based on human experience, and the planned data center computer room is often difficult to achieve the expected energy efficiency level. Therefore, after the data center computer room is built and put into operation, it is necessary to reduce the PUE value of the data center by optimizing the operating parameters of the data center computer room. Therefore, how to reasonably plan the data center computer room to reduce the PUE value of the data center has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The purpose of the embodiments of this specification is to provide a data center computer room planning method, device, equipment and storage medium to reduce the PUE value of the data center.

[0005] To achieve the above objectives, on the one hand, an embodiment of this specification provides a data center computer room planning method, including:

[0006] Obtain parameter values ​​of multiple target parameters of the data center to be built, and form a parameter value set;

[0007] Inputting the parameter value set into a PUE prediction model to obtain a PUE prediction value of the data center to be built;

[0008] When the PUE prediction value is not lower than the PUE threshold, adjusting the parameter values ​​in the parameter value set, and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold;

[0009] Determine a parameter value set corresponding to the PUE prediction value lower than the PUE threshold as a target set;

[0010] The computer room planning data of the data center to be built is determined according to the target set.

[0011] In an embodiment of the present specification, the adjusting the parameter value in the parameter value set includes:

[0012] Determining a target parameter importance ranking of the plurality of target parameters;

[0013] The parameter values ​​in the parameter value set are adjusted according to the importance ranking of the target parameters.

[0014] In an embodiment of the present specification, the target parameter importance ranking includes:

[0015] The multiple target parameters are sorted from large to small according to the degree of influence on the PUE prediction value.

[0016] In an embodiment of the present specification, adjusting the parameter values ​​in the parameter value set and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold includes:

[0017] Adjusting the parameter value corresponding to the first-ranked target parameter in the target parameter importance ranking;

[0018] Inputting the adjusted parameter value set into the PUE prediction model to obtain a new PUE prediction value;

[0019] When the new PUE prediction value is not lower than the PUE threshold, adjusting the parameter value corresponding to the first-ranked target parameter again, and re-predicting according to the re-adjusted parameter value set;

[0020] The process is repeated repeatedly until a PUE prediction value lower than the PUE threshold is obtained.

[0021] In an embodiment of the present specification, the step of adjusting the parameter value corresponding to the first-ranked target parameter in the target parameter importance ranking includes:

[0022] Determining an adjustment direction according to the influence of the first-priority target parameter on the PUE prediction value;

[0023] Within the value range of the first-priority target parameter, the parameter value corresponding to the first-priority target parameter is adjusted according to the adjustment direction and the set adjustment range.

[0024] In an embodiment of the present specification, the adjusting the parameter values ​​in the parameter value set and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold further includes:

[0025] When a PUE prediction value lower than the PUE threshold is not obtained after adjusting the parameter value corresponding to the first-rank target parameter to the limit value and re-predicting, adjusting the parameter value corresponding to the second-rank target parameter in the target parameter importance ranking, and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold;

[0026] When a PUE prediction value lower than the PUE threshold is not obtained after adjusting the parameter value corresponding to the second-rank target parameter to the limit value and re-predicting, adjusting the parameter value corresponding to the third-rank target parameter in the target parameter importance ranking, and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold;

[0027] The process is repeated repeatedly until a PUE prediction value lower than the PUE threshold is obtained.

[0028] In one embodiment of the present specification, the PUE prediction model is trained in the following manner:

[0029] Acquire historical data of multiple built data centers; the historical data includes parameter value measurement values ​​of multiple target parameters of the multiple built data centers, and PUE measurement values ​​of the multiple built data centers;

[0030] The parameter value measurement value is used as an input variable, and the PUE measurement value is used as a target output variable to train a machine learning model to obtain the PUE prediction model.

[0031] In one embodiment of the present specification, the machine learning model includes a random forest model.

[0032] In an embodiment of the present specification, determining the target parameter importance ranking of the multiple target parameters includes:

[0033] Selecting a parameter value from the set of parameter values;

[0034] Replace the parameter value with a random number within the corresponding value range;

[0035] Determine a difference index value between a PUE prediction value obtained by using the parameter value set after replacement and a PUE prediction value obtained by using the parameter value set before replacement, as a value of the degree of influence of the target parameter corresponding to the parameter value on the PUE prediction value;

[0036] Respectively determine the influence degree values ​​of the target parameters corresponding to the remaining parameter values ​​in the parameter value set on the PUE prediction value;

[0037] The plurality of influence degree values ​​are sorted, and the target parameter importance sorting of the plurality of target parameters is determined according to the sorting result.

[0038] In one embodiment of the present specification, the difference index value includes a mean square error increment.

[0039] In one embodiment of the present specification, the multiple target parameters include at least two of the following parameters of the data center or a combination thereof:

[0040] UPS system resource utilization;

[0041] Thermal environment temperature of the equipment room;

[0042] Utilization rate of air conditioning resources in the computer room;

[0043] The power density of the computer room; and

[0044] The average annual temperature of the computer room location.

[0045] In an embodiment of the present specification, determining the computer room planning data of the data center to be built according to the target set includes:

[0046] Input the UPS system resource utilization value in the target set and the total power of the IT equipment in the data center to be built into the formula Obtaining UPS system parameters of the data center to be built;

[0047] Among them, η1 is the UPS system resource utilization value, P IT is the total power of IT equipment in the data center to be built, P is the total rated power of the UPS system in the data center to be built, and φ is the power factor of the UPS system in the data center to be built.

[0048] In an embodiment of the present specification, determining the computer room planning data of the data center to be built according to the target set includes:

[0049] The power density of the computer rooms in the target set and the total power of the IT equipment in the data center to be built are input into the formula ρ=P IT / S, obtaining the indoor usable area of ​​the computer room of the data center to be built;

[0050] Among them, ρ is the power density of the computer room, P IT is the total power of IT equipment in the data center to be built, and S is the indoor usable area of ​​the computer room in the data center to be built.

[0051] In an embodiment of the present specification, determining the computer room planning data of the data center to be built according to the target set includes:

[0052] Input the resource utilization rate of the computer room air conditioner in the target set and the total power of the IT equipment in the data center to be built into the formula Obtaining the sensible cooling capacity of the air conditioner in the computer room of the data center to be built;

[0053] Among them, AUE is the room air conditioning resource utilization rate, P IT is the total power of IT equipment in the data center to be built, k is the environmental heat dissipation coefficient, S is the indoor usable area of ​​the computer room in the data center to be built, and Q is the sensible cooling capacity of the air conditioner in the computer room of the data center to be built.

[0054] In an embodiment of the present specification, determining the computer room planning data of the data center to be built according to the target set includes:

[0055] The site selection range of the computer room of the data center to be built is determined according to the annual average temperature of the locations of the computer rooms in the target set.

[0056] On the other hand, the embodiment of this specification also provides a data center room planning device, including:

[0057] A parameter acquisition module, used to acquire parameter values ​​of multiple target parameters of the data center to be built and form a parameter value set;

[0058] A PUE prediction module, used to input the parameter value set into a PUE prediction model to obtain a PUE prediction value of the data center to be built;

[0059] A parameter adjustment module, configured to adjust the parameter values ​​in the parameter value set when the PUE prediction value is not lower than the PUE threshold, and re-predict according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold;

[0060] A set determination module, used to determine a parameter value set corresponding to the PUE prediction value lower than the PUE threshold as a target set;

[0061] The planning determination module is used to determine the computer room planning data of the data center to be built according to the target set.

[0062] On the other hand, an embodiment of the present specification further provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein when the computer program is executed by the processor, the instructions of the above method are executed.

[0063] On the other hand, an embodiment of the present specification further provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor of a computer device, the instructions of the above method are executed.

[0064] It can be seen from the technical solutions provided by the above embodiments of this specification that the embodiments of this specification can input the parameter values ​​of multiple target parameters of the data center to be built as a parameter value set into the PUE prediction model, so as to obtain the PUE prediction value of the data center to be built. When the PUE prediction value is not lower than the PUE threshold, the parameter values ​​in the parameter value set can be adjusted, and re-predicted according to the adjusted parameter value set, so that after several adjustments and predictions, a PUE prediction value lower than the PUE threshold can be finally obtained; then the parameter value set corresponding to the PUE prediction value lower than the PUE threshold is used as the target set; in this way, the computer room planning data of the data center to be built can be determined according to the target set. Therefore, the data center computer room planning method of the embodiments of this specification reduces the PUE value of the data center. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings:

[0066] Figure 1 A flowchart of a data center computer room planning method in some embodiments of this specification is shown;

[0067] Figure 2 A training diagram of a PUE model in some embodiments of this specification is shown;

[0068] Figure 3 It shows a structural block diagram of a data center computer room planning device in some embodiments of this specification;

[0069] Figure 4 The structure block diagram of the computer device of some embodiments of the present specification is shown.

[0070] [Description of Reference Numerals]

[0071] 31. Parameter acquisition module;

[0072] 32. PUE prediction module;

[0073] 33. Parameter adjustment module;

[0074] 34. Set determination module;

[0075] 35. Planning and determination module;

[0076] 402. Computer equipment;

[0077] 404, processor;

[0078] 406. Memory;

[0079] 408, driving mechanism;

[0080] 410, input / output module;

[0081] 412. Input devices;

[0082] 414. Output device;

[0083] 416. Presentation equipment;

[0084] 418. Graphical user interface;

[0085] 420, network interface;

[0086] 422. Communication link;

[0087] 424. Communication bus. DETAILED DESCRIPTION

[0088] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0089] refer to Figure 1 As shown, the data center room planning method of some embodiments of this specification may include the following steps:

[0090] S101. Obtain parameter values ​​of multiple target parameters of a data center to be built, and form a parameter value set.

[0091] S102: Input the parameter value set into a PUE prediction model to obtain a PUE prediction value of the data center to be built.

[0092] S103: When the PUE prediction value is not lower than the PUE threshold, adjust the parameter values ​​in the parameter value set, and re-predict according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold.

[0093] S104: Determine a parameter value set corresponding to the PUE prediction value lower than the PUE threshold as a target set.

[0094] S105. Determine computer room planning data of the data center to be built according to the target set.

[0095] In the embodiments of the present specification, the parameter values ​​of multiple target parameters of the data center to be built can be input into the PUE prediction model as a parameter value set, so as to obtain the PUE prediction value of the data center to be built. When the PUE prediction value is not lower than the PUE threshold, the parameter values ​​in the parameter value set can be adjusted, and re-predicted based on the adjusted parameter value set, so that after several adjustments and predictions, a PUE prediction value lower than the PUE threshold can be finally obtained; then the parameter value set corresponding to the PUE prediction value lower than the PUE threshold is used as the target set; in this way, the computer room planning data of the data center to be built can be determined based on the target set. Therefore, the data center computer room planning method of the embodiments of the present specification reduces the PUE value of the data center.

[0096] In the embodiments of this specification, a data center (such as an enterprise data center, etc.) generally refers to a specific IT equipment network that collaborates globally to transmit, accelerate, display, calculate, and store data information. Among them, a data center to be built refers to a data center waiting to be planned and built. Correspondingly, the built data center mentioned below refers to a data center that has been completed and put into operation.

[0097] Research shows that the site selection, area layout, selection of air-conditioning equipment in the data center computer room (generally refers to the computer room, power supply system, air-conditioning equipment, etc. in the data center in addition to IT equipment (computers, servers, network equipment, etc.)) and the selection of uninterruptible power supply (UPS) system equipment have a great impact on the PUE value of the data center, which are explained below.

[0098] 1. Site selection for the computer room

[0099] Climate conditions vary from region to region. Temperature is an important factor affecting the energy consumption of the computer room air conditioning system. The higher the temperature, the greater the energy consumption of the air conditioning system. Therefore, the annual average temperature of the computer room location can be used as an evaluation parameter for the computer room site selection.

[0100] 2. Computer Room Layout

[0101] Reasonable layout of the IT equipment in the data center according to the planned computer room area can avoid the waste of computer room area, help to form a good computer room airflow organization, and reduce the heat dissipation of the computer room environment. Therefore, the power density of the computer room and the utilization rate of the computer room air conditioning resources can be used as evaluation parameters for the layout of the computer room.

[0102] 3. Computer room wall

[0103] Computer rooms generally use concrete block walls as much as possible, and avoid using external windows and glass curtain walls to reduce the heat exchange between the computer room and the external environment. Therefore, the computer room air conditioning resource utilization rate can be used as an evaluation parameter for the computer room wall.

[0104] 4. Air conditioning configuration of computer room

[0105] The computer room air conditioner should be reasonably configured according to the heat dissipation and distribution of IT equipment in the data center, the heat dissipation of the computer room environment, etc., including the cooling capacity of a single unit, the air conditioner redundancy method, etc. Therefore, the computer room air conditioner resource utilization rate can be used as an evaluation parameter for the configuration of the computer room air conditioner.

[0106] 5. UPS system configuration

[0107] The efficiency of the UPS system directly affects the amount of power consumption. When the UPS system has a reasonable load capacity and a high power factor, the reactive power and corresponding losses in the UPS system can be greatly reduced. The low efficiency of the UPS system itself will inevitably increase the overall energy consumption. Therefore, the UPS system resource utilization rate can be used as an evaluation parameter for the UPS system configuration.

[0108] 6. Thermal environment temperature of the equipment room

[0109] The importance of the thermal environment temperature of the computer room cannot be ignored when designing the computer room. The temperature must comply with the temperature range of the computer room of different levels specified in the data center design specifications, and at the same time meet the temperature requirements of the normal operation of IT equipment in the computer room. For energy saving in the computer room, appropriately increasing the thermal environment temperature of the computer room can reduce the energy consumption of the computer room air conditioning system.

[0110] Based on the above analysis, the above evaluation parameters are the main factors affecting the PUE value of the data center. Therefore, in some embodiments of this specification, multiple target parameters can be selected as: UPS system resource utilization rate, computer room thermal environment temperature, computer room air conditioning resource utilization rate, computer room power density and annual average temperature of the computer room location.

[0111] For ease of understanding, the following embodiments of this specification are described using the above five parameters as target parameters. It should be pointed out that this is only an example analysis to illustrate the main factors affecting the PUE value of the data center. In other embodiments of this specification, the selection of multiple target parameters can be appropriately increased or decreased according to actual needs. For example, in one embodiment of this specification, two or more of the above five parameters can be selected as multiple target parameters. For another example, since the factors affecting the PUE value of the data center may also include other non-main factors (i.e., factors that have a relatively small impact on PUE), in another embodiment of this specification, one or more other parameters of the data center can be added to the above five parameters as multiple target parameters. Therefore, this specification does not make a sole limitation on multiple target parameters.

[0112] It has been explained in the above background technology that the planning of the data center room is generally determined based on human experience, and the planned data center room is often difficult to achieve the desired energy efficiency level. Therefore, it is necessary to adjust the planning so that the data center room can be constructed according to the adjusted relevant parameters, and a data center that meets the PUE threshold requirements can be directly obtained. Therefore, the initial data of the room planning planned based on human experience can be used as the object to be processed. Thus, the parameter values ​​of multiple target parameters of the data center to be built can be obtained according to the object to be processed, and a parameter value set can be formed.

[0113] For example, in an embodiment of the present specification, taking the above five parameters as target parameters as an example, the parameter values ​​of the multiple target parameters are obtained as follows:

[0114] 1. UPS system resource utilization

[0115] For UPS systems, the symbol of energy saving is efficiency. The larger the load capacity of the UPS system, the higher its efficiency. Therefore, the UPS system resource utilization rate η1 can be used to evaluate the efficiency of the UPS system in the computer room. The UPS system resource utilization rate η1 can be expressed as: Among them, η1 is the UPS system resource utilization value, P IT is the total power of IT equipment in the data center to be built, P is the total rated power of the UPS system in the data center to be built, and φ is the power factor of the UPS system in the data center to be built. IT , P and φ, and then enter the formula Thus, the UPS system resource utilization rate of the data center to be built can be calculated.

[0116] 2. Utilization rate of air conditioning resources in computer rooms

[0117] The computer room air conditioner is usually configured with redundancy. The cooling capacity planned for the computer room area should be greater than the used cooling capacity. The computer room air conditioner resource utilization rate (AUE) is an indicator used to evaluate the efficiency of the computer room air conditioner. The higher the computer room air conditioner resource utilization rate, the lower the energy consumption of the computer room air conditioner system. The computer room air conditioner resource utilization rate can be expressed as: Among them, AUE is the room air conditioning resource utilization rate, P IT is the total power of IT equipment in the data center to be built, k is the environmental heat dissipation coefficient (an empirical value, ranging from 0.1 to 0.18Kw / m2), S is the indoor usable area of ​​the computer room of the data center to be built, and Q is the sensible cooling capacity of the air conditioner in the computer room of the data center to be built. IT , k, S, and Q, and then enter the formula Thus, the utilization rate of the air conditioning resources in the computer room of the data center to be built can be calculated.

[0118] 3. Annual average temperature of the computer room location

[0119] Due to the different climate conditions in different cities, especially the temperature, the energy consumption of the air conditioning system is closely related to the outdoor heat dissipation environment. The lower the outdoor temperature, the lower the energy consumption of the air conditioning system. Therefore, the annual average temperature of the computer room location can be used to evaluate the outdoor heat dissipation environment of the computer room. One or more computer room location options can be given in the object to be processed, and the annual average temperature of each computer room location option is determined, based on which the annual average temperature of the computer room location of the data center to be built can be determined.

[0120] 4. Power density of the computer room

[0121] The power density of the computer room can be used to evaluate the power density of the computer room. The power density of the computer room can be expressed as: ρ = P IT / S. Among them, ρ is the power density of the computer room, P IT is the total power of IT equipment in the data center to be built, and S is the indoor usable area of ​​the computer room in the data center to be built. P can be obtained according to the object to be processed. IT , S, and then enter the formula ρ=P IT / S, so as to obtain the power density of the computer room of the data center to be built.

[0122] 5. Thermal environment temperature of the equipment room

[0123] The object to be processed generally provides a recommended computer room ambient temperature as the computer room thermal ambient temperature.

[0124] The PUE prediction model in some embodiments of this specification may be a pre-trained model. For example, in one embodiment of this specification, the PUE prediction model is trained in the following manner:

[0125] (1) Obtain historical data of multiple built data centers; the historical data may include parameter value measurement values ​​of multiple target parameters of the multiple built data centers, and PUE measurement values ​​of the multiple built data centers. Please note that the measurement values ​​in the embodiments of this specification include not only the values ​​directly measured by the measuring device, but also the values ​​indirectly obtained by calculating and processing the measured values.

[0126] For example, a joint-stock bank has currently built and put into operation 21 provincial data centers. From the historical data of these 21 provincial data centers, the bank can obtain the measured values ​​of five target parameters, including UPS system resource utilization rate, computer room thermal environment temperature, computer room air conditioning resource utilization rate, computer room power density, and annual average temperature of the computer room location, as well as the PUE measurement values ​​of these 21 provincial data centers.

[0127] (2) Using the parameter value measurement value as the input variable and the PUE measurement value as the target output variable, a machine learning model is trained to obtain the PUE prediction model.

[0128] Among them, the machine learning model can be selected according to actual needs. For example, in Figure 2 In the exemplary embodiment shown, the parameter value measurements of the five target parameters of the above-mentioned 21 provincial data centers can be used as input variables, and the PUE measurement values ​​of the above-mentioned 21 provincial data centers can be used as target output variables to train the preset random forest model, so that a PUE prediction model that meets the preset conditions (such as accuracy, recall rate, etc.) can be trained: PUE = RF (η1, ρ, T, AUE, t). Among them, RF represents the random forest algorithm, PUE is the power utilization efficiency, η1 is the UPS system resource utilization rate, ρ is the power density of the computer room, T is the thermal environment temperature of the computer room, AUE is the utilization rate of the air conditioning resources in the computer room, and t is the annual average temperature. On this basis, the data center computer room planning can be carried out for each data center to be built according to the PUE prediction model.

[0129] In order to improve planning processing efficiency, the parameter values ​​in the parameter value set may be adjusted in the following manner:

[0130] (1) Determine the target parameter importance ranking of the multiple target parameters.

[0131] The target parameter importance ranking may be: multiple target parameters are ranked from large to small according to the degree of influence on the PUE prediction value.

[0132] In some embodiments of the present specification, determining the target parameter importance ranking of the multiple target parameters may include the following:

[0133] 1) Select a parameter value from the set of parameter values.

[0134] 2) Replace the parameter value with a random number within the corresponding value range.

[0135] 3) Determine the difference index value between the PUE prediction value obtained using the set of parameter values after replacement and the PUE prediction value obtained using the set of parameter values before replacement, so as to obtain the influence degree value of the target parameter corresponding to this parameter value on the PUE prediction value. In an embodiment of this specification, the difference index value may be, for example, the error increment, the mean square error increment, etc.

[0136] In an embodiment of this specification, in order to obtain a more accurate influence degree value. When obtaining the PUE prediction value obtained using the set of parameter values after replacement, multiple calculations may be repeated (for example, 10 times) to obtain the average value of the PUE prediction value obtained using the set of parameter values after replacement, and determine the difference index value between this average value and the PUE prediction value obtained using the set of parameter values before replacement.

[0137] For example, assume that the set of parameter values is {η1 = 0.7, ρ = 0.4, T = 25°C, AUE = 0.5, t = 15°C}, and select a parameter value η1 = 0.7 from it. Replace η1 = 0.7 with η1 = s (s represents a random number, and 0 < s < 1). Assume that when obtaining the PUE prediction value obtained using the set of parameter values after replacement, the calculation is repeated 10 times, and the values of s for each time are as follows:

[0138] η1 = 0.1;

[0139] η1 = 0.6;

[0140] η1 = 0.3;

[0141] η1 = 0.6;

[0142] η1 = 0.4;

[0143] η1 = 0.8;

[0144] η1 = 0.7;

[0145] η1 = 0.5;

[0146] η1 = 0.9;

[0147] η1 = 0.2.

[0148] Then the corresponding sets of parameter values are:

[0149] {η1 = 0.1, ρ = 0.4, T = 25°C, AUE = 0.5, t = 15°C};

[0150] {η1=0.6, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0151] {η1=0.3, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0152] {η1=0.6, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0153] {η1=0.4, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0154] {η1=0.8, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0155] {η1=0.7, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0156] {η1=0.5, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0157] {η1=0.9, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0158] {η1=0.2, ρ=0.4, T=25℃, AUE=0.5, t=15℃};

[0159] By inputting the above 10 parameter value sets into the PUE prediction model respectively, 10 PUE prediction values ​​can be predicted accordingly: PUE1 to PUE10. Assuming that the mean value of PUE1 to PUE10 is x, and the PUE prediction value obtained by using the parameter value set before η1 replacement is y, the mean square error increment a1 between x and y can be calculated, so that a1 can be used as the influence degree value of η1 on the PUE prediction value.

[0160] 4) respectively determining the influence degree values ​​of the target parameters corresponding to the remaining parameter values ​​in the parameter value set on the PUE prediction value.

[0161] For example, taking the above parameter value set {η1=0.7, ρ=0.4, T=25℃, AUE=0.5, t=15℃} as an example, after obtaining the influence value a1 of η1 on the PUE prediction value, based on the same method, the influence values ​​a2, a3, a4, a5 of ρ, T, AUE and t on the PUE prediction value can be obtained respectively.

[0162] 5) Sorting the multiple impact degree values, and determining the target parameter importance ranking of the multiple target parameters according to the sorting result.

[0163] For example, taking the above parameter value set {η1=0.7,ρ=0.4,T=25℃,AUE=0.5,t=15℃} as an example, the influence of η1,ρ,T,AUE and t on the predicted PUE value are a1, a2, a3, a4, a5 respectively, and a1>a2>a3>a4>a5. According to the influence on the predicted PUE value, the following ranking of the importance of target parameters can be obtained (here is just an example): η1→ρ→T→AUE→t.

[0164] (2) Adjusting the parameter values ​​in the parameter value set according to the importance ranking of the target parameters.

[0165] In some embodiments of the present specification, adjusting the parameter values ​​in the parameter value set and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold may include the following steps:

[0166] 1) Adjust the parameter value corresponding to the first-ranked target parameter in the target parameter importance ranking.

[0167] In one embodiment of the present specification, adjusting the parameter value corresponding to the first-rank target parameter in the target parameter importance ranking may include: determining an adjustment direction according to the nature of the impact of the first-rank target parameter on the PUE prediction value; and adjusting the parameter value corresponding to the first-rank target parameter according to the adjustment direction and the set adjustment range within the value range of the first-rank target parameter.

[0168] For example, in the above order of η1→ρ→T→AUE→t, η1 is the first-order target parameter (the value range of η1 is 0<η1<1). Since the influence of η1 on the predicted PUE value is: η1 is negatively correlated with the predicted PUE value, that is, when other conditions remain unchanged, the larger η1 is, the smaller the predicted PUE value is. Therefore, in order to reduce the predicted PUE value, η1 should be increased, that is, the adjustment direction is to increase. Assuming the adjustment amplitude is 0.1, η1 can be increased from η1=0.7 to η1=0.8.

[0169] 2) Input the adjusted parameter value set into the PUE prediction model to obtain a new PUE prediction value.

[0170] 3) When the new PUE prediction value is not lower than the PUE threshold, the parameter value corresponding to the first priority target parameter is adjusted again, and re-predicted based on the re-adjusted parameter value set.

[0171] 4) Repeat the above steps until a PUE prediction value lower than the PUE threshold is obtained.

[0172] In an embodiment of the present specification, adjusting the parameter values ​​in the parameter value set and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold may also include:

[0173] 5) When the parameter value corresponding to the first-priority target parameter is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to the second-priority target parameter in the target parameter importance ranking is adjusted, and re-predicted based on the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold.

[0174] For example, in the above-mentioned ordering of η1→ρ→T→AUE→t, when the parameter value corresponding to η1 is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to ρ can be adjusted to the limit value and re-predicted to obtain a PUE prediction value lower than the PUE threshold.

[0175] 6) When the parameter value corresponding to the second-rank target parameter is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to the third-rank target parameter in the target parameter importance ranking is adjusted, and re-predicted based on the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold.

[0176] For example, in the above-mentioned ordering of η1→ρ→T→AUE→t, when the parameter value corresponding to ρ is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to T can be adjusted to the limit value and re-predicted to obtain a PUE prediction value lower than the PUE threshold.

[0177] 7) Repeat the above steps until a PUE prediction value lower than the PUE threshold is obtained.

[0178] Since each target parameter is sorted from large to small according to the degree of influence on the PUE prediction value in the target parameter importance sorting, when the parameter values ​​in the parameter value set are adjusted in this sorting, the target parameter with the greatest influence on the PUE prediction value will be adjusted first, so that the PUE prediction value obtained after adjustment can be quickly reduced, thus greatly improving the planning and processing efficiency of the data center computer room.

[0179] In some embodiments of the present specification, when a certain parameter value set is input into a PUE prediction model and a PUE prediction value lower than a PUE threshold is obtained, the parameter value set is the target set. For example, when a parameter value set {η1=0.9, ρ=0.9, T=26℃, AUE=0.5, t=15℃} is input into a PUE prediction model and a PUE prediction value lower than a PUE threshold is obtained, the parameter value set {η1=0.9, ρ=0.9, T=26℃, AUE=0.5, t=15℃} is the target set.

[0180] In some embodiments of the present specification, determining the computer room planning data of the to-be-built data center according to the target set may include:

[0181] Input the UPS system resource utilization value in the target set and the total power of the IT equipment in the data center to be built into the formula Obtaining UPS system parameters of the data center to be built;

[0182] The power density of the computer rooms in the target set and the total power of the IT equipment in the data center to be built are input into the formula ρ=P IT / S, obtaining the indoor usable area of ​​the computer room of the data center to be built;

[0183] Input the resource utilization rate of the computer room air conditioner in the target set and the total power of the IT equipment in the data center to be built into the formula Obtaining the sensible cooling capacity of the air conditioner in the computer room of the data center to be built;

[0184] Determine the site selection range of the computer room of the data center to be built according to the annual average temperature of the computer room location in the target set;

[0185] The thermal environment temperature of the computer room in the target set is used as the thermal environment temperature of the computer room of the data center to be built.

[0186] In this way, the automatic planning and processing of the data center room is realized. After the data center to be built is completed and put into operation, PUE online optimization can be performed, which can help to further reduce the PUE value of the data center and improve its energy efficiency level.

[0187] Although the process flows described above include multiple operations that occur in a particular order, it should be clearly understood that the processes may include more or fewer operations, which may be performed sequentially or in parallel (eg, using parallel processors or a multi-threaded environment).

[0188] Corresponding to the above-mentioned data center room planning method, this specification also provides an embodiment of a data center room planning device. Figure 3As shown, in some embodiments of this specification, the data center room planning device may include:

[0189] The parameter acquisition module 31 may be used to acquire parameter values ​​of multiple target parameters of the data center to be built and form a parameter value set;

[0190] The PUE prediction module 32 may be used to input the parameter value set into a PUE prediction model to obtain a PUE prediction value of the data center to be built;

[0191] The parameter adjustment module 33 may be configured to adjust the parameter values ​​in the parameter value set when the PUE prediction value is not lower than the PUE threshold, and re-predict according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold;

[0192] The set determination module 34 may be used to determine a set of parameter values ​​corresponding to the PUE prediction value lower than the PUE threshold as a target set;

[0193] The planning determination module 35 may be used to determine the computer room planning data of the data center to be built according to the target set.

[0194] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0195] In the embodiments of the present specification, the parameter values ​​of multiple target parameters of the data center to be built can be input into the PUE prediction model as a parameter value set, so as to obtain the PUE prediction value of the data center to be built. When the PUE prediction value is not lower than the PUE threshold, the parameter values ​​in the parameter value set can be adjusted, and re-predicted based on the adjusted parameter value set, so that after several adjustments and predictions, a PUE prediction value lower than the PUE threshold can be finally obtained; then the parameter value set corresponding to the PUE prediction value lower than the PUE threshold is used as the target set; in this way, the computer room planning data of the data center to be built can be determined based on the target set. Therefore, the data center computer room planning method of the embodiments of the present specification reduces the PUE value of the data center.

[0196] The embodiments of this specification also provide a computer device. Figure 4As shown, in some embodiments of the present specification, the computer device 402 may include one or more processors 404, such as one or more central processing units (CPUs) or graphics processing units (GPUs), and each processing unit may implement one or more hardware threads. The computer device 402 may also include any memory 406, which is used to store any kind of information such as code, settings, data, etc. In a specific implementation, a computer program on the memory 406 and that can be run on the processor 404, when the computer program is run by the processor 404, can execute instructions according to the above method. Non-limiting, for example, the memory 406 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any memory can use any technology to store information. Further, any memory can provide volatile or non-volatile retention of information. Further, any memory can represent a fixed or removable component of the computer device 402. In one case, when the processor 404 executes an associated instruction stored in any memory or a combination of memories, the computer device 402 can perform any operation of the associated instruction. The computer device 402 also includes one or more drive mechanisms 408 for interacting with any storage, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0197] The computer device 402 may also include an input / output module 410 (I / O) for receiving various inputs (via input device 412) and for providing various outputs (via output device 414). A specific output mechanism may include a presentation device 416 and an associated graphical user interface 418 (GUI). In other embodiments, the input / output module 410 (I / O), input device 412, and output device 414 may not be included, and the computer device 402 may be used as a computer device in a network. The computer device 402 may also include one or more network interfaces 420 for exchanging data with other devices via one or more communication links 422. One or more communication buses 424 couple the components described above together.

[0198] The communication link 422 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 422 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0199] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products of some embodiments of the present specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processor to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processor generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0200] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processor to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0201] These computer program instructions can also be loaded onto a computer or other programmable data processor so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0202] In a typical configuration, a computer device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0203] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0204] Computer readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computer device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0205] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, the embodiments of this specification may take the form of complete hardware embodiments, complete software embodiments or embodiments combining software and hardware. Moreover, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0206] The present specification embodiments may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present specification embodiments may also be practiced in distributed computing environments where tasks are performed by remote processors connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0207] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. In the description of this specification, the description of the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of this specification. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, in the absence of contradiction, a person skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0208] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A data center computer room planning method, characterized in that: include: Obtain parameter values ​​of multiple target parameters of the data center to be built, and form a parameter value set; Inputting the parameter value set into a power usage efficiency (PUE) prediction model to obtain a PUE prediction value of the data center to be built; When the PUE prediction value is not lower than the PUE threshold, adjust the parameter values ​​in the parameter value set, and re-predict according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; determine the parameter value set corresponding to the PUE prediction value lower than the PUE threshold as a target set; Determining the computer room planning data of the data center to be built according to the target set; Wherein, adjusting the parameter values ​​in the parameter value set includes: Determining the target parameter importance ranking of the multiple target parameters, including: selecting a parameter value from the parameter value set; replacing the parameter value with a random number within a corresponding value range; determining a difference index value between a PUE prediction value obtained using the replaced parameter value set and a PUE prediction value obtained using the parameter value set before the replacement, as a value of the influence degree of the target parameter corresponding to the parameter value on the PUE prediction value; respectively determining the influence degree values ​​of the target parameters corresponding to the remaining parameter values ​​in the parameter value set on the PUE prediction value; sorting the multiple influence degree values, and determining the target parameter importance ranking of the multiple target parameters according to the sorting result; The parameter values ​​in the parameter value set are adjusted according to the target parameter importance ranking; specifically comprising: when the parameter value corresponding to the first-rank target parameter is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to the second-rank target parameter in the target parameter importance ranking is adjusted, and re-predicted according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; when the parameter value corresponding to the second-rank target parameter is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to the third-rank target parameter in the target parameter importance ranking is adjusted, and re-predicted according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; and the process is repeated recursively until a PUE prediction value lower than the PUE threshold is obtained.

2. The data center room planning method according to claim 1, characterized in that: The target parameter importance ranking includes: The multiple target parameters are sorted from large to small according to the degree of influence on the PUE prediction value.

3. The data center room planning method according to claim 1, characterized in that: The adjusting the parameter values ​​in the parameter value set and re-predicting according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold includes: Adjusting the parameter value corresponding to the first-ranked target parameter in the target parameter importance ranking; Inputting the adjusted parameter value set into the PUE prediction model to obtain a new PUE prediction value; When the new PUE prediction value is not lower than the PUE threshold, adjusting the parameter value corresponding to the first-ranked target parameter again, and re-predicting according to the re-adjusted parameter value set; The process is repeated repeatedly until a PUE prediction value lower than the PUE threshold is obtained.

4. The data center room planning method according to claim 3, characterized in that: The adjusting the parameter value corresponding to the first-ranked target parameter in the target parameter importance ranking includes: Determining an adjustment direction according to the influence of the first-priority target parameter on the PUE prediction value; Within the value range of the first-priority target parameter, the parameter value corresponding to the first-priority target parameter is adjusted according to the adjustment direction and the set adjustment range.

5. The data center room planning method according to claim 1, characterized in that: The PUE prediction model is trained in the following way: Acquire historical data of multiple built data centers; the historical data includes parameter value measurement values ​​of multiple target parameters of the multiple built data centers, and PUE measurement values ​​of the multiple built data centers; The parameter value measurement value is used as an input variable, and the PUE measurement value is used as a target output variable to train a machine learning model to obtain the PUE prediction model.

6. The data center room planning method according to claim 5, characterized in that: The machine learning model includes a random forest model.

7. The data center room planning method according to claim 1, characterized in that: The difference index value includes a mean square error increment.

8. The data center room planning method according to claim 1, characterized in that: The plurality of target parameters include at least two of the following parameters of the data center or a combination thereof: UPS system resource utilization; Thermal environment temperature of the equipment room; Utilization rate of air conditioning resources in the computer room; The power density of the computer room; and The average annual temperature of the computer room location.

9. The data center room planning method according to claim 8, characterized in that: The determining the computer room planning data of the data center to be built according to the target set includes: Input the UPS system resource utilization value in the target set and the total power of the IT equipment in the data center to be built into the formula Obtaining UPS system parameters of the data center to be built; Among them, η1 is the UPS system resource utilization value, P IT is the total power of IT equipment in the data center to be built, P is the total rated power of the UPS system in the data center to be built, and φ is the power factor of the UPS system in the data center to be built.

10. The data center room planning method according to claim 8, characterized in that: The determining the computer room planning data of the data center to be built according to the target set includes: The power density of the computer rooms in the target set and the total power of the IT equipment in the data center to be built are input into the formula ρ=P IT / S, obtaining the indoor usable area of ​​the computer room of the data center to be built; Where ρ is the power density of the computer room, P IT is the total power of IT equipment in the data center to be built, and S is the indoor usable area of ​​the computer room in the data center to be built.

11. The data center room planning method according to claim 8, characterized in that: The determining the computer room planning data of the data center to be built according to the target set includes: Input the resource utilization rate of the computer room air conditioner in the target set and the total power of the IT equipment in the data center to be built into the formula Obtaining the sensible cooling capacity of the air conditioner in the computer room of the data center to be built; Among them, AUE is the room air conditioning resource utilization rate, P IT is the total power of IT equipment in the data center to be built, k is the environmental heat dissipation coefficient, S is the indoor usable area of ​​the computer room in the data center to be built, and Q is the sensible cooling capacity of the air conditioner in the computer room of the data center to be built.

12. The data center room planning method according to claim 1, characterized in that: The determining the computer room planning data of the data center to be built according to the target set includes: The site selection range of the computer room of the data center to be built is determined according to the annual average temperature of the locations of the computer rooms in the target set.

13. A data center computer room planning device, characterized in that: include: A parameter acquisition module, used to acquire parameter values ​​of multiple target parameters of the data center to be built and form a parameter value set; A PUE prediction module, used to input the parameter value set into a PUE prediction model to obtain a PUE prediction value of the data center to be built; A parameter adjustment module, configured to adjust the parameter values ​​in the parameter value set when the PUE prediction value is not lower than the PUE threshold, and re-predict according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; A set determination module, used to determine a parameter value set corresponding to the PUE prediction value lower than the PUE threshold as a target set; A planning determination module, used to determine the computer room planning data of the data center to be built according to the target set; Wherein, adjusting the parameter values ​​in the parameter value set includes: Determining the target parameter importance ranking of the multiple target parameters, including: selecting a parameter value from the parameter value set; replacing the parameter value with a random number within a corresponding value range; determining a difference index value between a PUE prediction value obtained using the replaced parameter value set and a PUE prediction value obtained using the parameter value set before the replacement, as a value of the influence degree of the target parameter corresponding to the parameter value on the PUE prediction value; respectively determining the influence degree values ​​of the target parameters corresponding to the remaining parameter values ​​in the parameter value set on the PUE prediction value; sorting the multiple influence degree values, and determining the target parameter importance ranking of the multiple target parameters according to the sorting result; The parameter values ​​in the parameter value set are adjusted according to the target parameter importance ranking; specifically comprising: when the parameter value corresponding to the first-rank target parameter is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to the second-rank target parameter in the target parameter importance ranking is adjusted, and re-predicted according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; when the parameter value corresponding to the second-rank target parameter is adjusted to the limit value and re-predicted, if a PUE prediction value lower than the PUE threshold is not obtained, the parameter value corresponding to the third-rank target parameter in the target parameter importance ranking is adjusted, and re-predicted according to the adjusted parameter value set to obtain a PUE prediction value lower than the PUE threshold; and the process is repeated recursively until a PUE prediction value lower than the PUE threshold is obtained.

14. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that: When the computer program is executed by the processor, the computer program executes the instructions of the method according to any one of claims 1 to 12.

15. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor of a computer device, the computer program executes the instructions of the method according to any one of claims 1 to 12.

Images