Energy consumption assessment for data center

Abstract

Disclosed in the present disclosure are an energy consumption assessment method and apparatus for a data center, an electronic device, and a storage medium. The method comprises: acquiring energy-related parameters and energy storage cost parameters of devices in a data center and task processing parameters for the data center to process computing tasks, and acquiring unit electricity prices in different time periods provided for the data center; establishing an energy management model of the data center on the basis of the acquired energy-related parameters, energy storage cost parameters, task processing parameters, and unit electricity prices in different time periods; and on the basis of the energy management model and a preset maximum energy consumption cost of the data center, calculating electricity power consumption ranges of the data center in the different time periods.

Description

[0001] Data center energy consumption assessment technology field

[0002]

[0001] The present disclosure relates to the field of computer technology, and more particularly to data center energy consumption assessment.

[0003]

[0002] The low-carbon and low-cost energy consumption of data centers is crucial. As new energy sources, such as wind power and photovoltaics, are integrated into the power system, the gap between peak and off-peak electricity prices has widened. By exploring the energy consumption of data centers, it is possible to absorb the output of their own new energy equipment, thereby improving the low-carbonization of data center energy consumption and reducing the energy costs of data centers.

[0004]

[0003] Therefore, the evaluation of energy consumption in data centers has become a technical problem that needs to be solved urgently.

[0005]

[0004] The present disclosure provides a data center energy consumption evaluation method, device, electronic device and storage medium.

[0006]

[0005] A first aspect of the present disclosure proposes a method for evaluating energy consumption in a data center, the method comprising: obtaining energy-related parameters, energy storage cost parameters, and task processing parameters for computing tasks processed by the data center of each device in the data center, and obtaining unit electricity prices for different time periods provided to the data center; establishing an energy management model for the data center based on the energy-related parameters, the energy storage cost parameters, the task processing parameters, and the unit electricity prices for different time periods, the energy management model comprising an energy cost relationship, an energy constraint relationship, and a resource usage relationship of the data center; and calculating the power consumption range of the data center in the different time periods based on the energy management model and a preset maximum energy consumption value of the data center.

[0007]

[0006] The second aspect of the present disclosure proposes a data center energy consumption evaluation device, which includes: a parameter acquisition module, which is used to obtain energy-related parameters, energy storage cost parameters and task processing parameters of each device in the data center for processing computing tasks, and obtain the unit electricity price provided for the data center in different time periods; a model establishment module, which is used to establish an energy management model of the data center based on the energy-related parameters, the energy storage cost parameters, the task processing parameters and the unit electricity price in different time periods, and the energy management model includes the energy cost relationship, energy constraint relationship and resource usage relationship of the data center; a calculation module, which is used to calculate the power consumption range of the data center in the different time periods based on the energy management model and the preset maximum energy consumption cost of the data center.

[0008]

[0007] A third aspect of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in the first aspect above.

[0009]

[0008] A fourth aspect of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the method described in the first aspect above.

[0010]

[0009] Based on the data center energy consumption evaluation method and device described in the first and second aspects above, the present disclosure has at least the following beneficial effects or advantages: by obtaining energy-related parameters, energy storage cost parameters, and task processing parameters of the data center's computing tasks, as well as unit electricity prices provided to the data center in different time periods, these data are used as model parameters to establish an energy management model. The management model includes a data center energy cost relationship, an energy constraint relationship, and a resource usage relationship, and comprehensively describes the energy consumption characteristics of the data center from different perspectives. By giving the data center an energy cost budget, that is, a given maximum energy cost, the management model is used to solve the data center's power consumption range in different time periods. In other words, each time period corresponds to a power consumption range corresponding to the unit electricity price. This can help the data center understand the power consumption range under the given budget, thereby guiding the data center's electricity purchasing behavior in each time period.

[0011]

[0010] The above description is only an overview of the technical solution of the present disclosure. In order to more clearly understand the technical means of the present disclosure, it can be implemented according to the contents of the description. In order to make the above and other purposes, features and advantages of the present disclosure more obvious and easy to understand, the following specific embodiments of the present disclosure are specifically cited.

[0012]

[0011] The drawings described herein are intended to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are intended to explain the present disclosure and do not constitute an improper limitation of the present disclosure. In the drawings:

[0013]

[0012] FIG1 is a flow chart of an embodiment of a method for evaluating energy consumption in a data center according to an exemplary embodiment;

[0014]

[0013] FIG2 is a schematic diagram showing a subtask dependency relationship of a batch processing task according to an exemplary embodiment;

[0015]

[0014] FIG3 is a schematic diagram showing a framework of energy consumption evaluation of a data center according to an exemplary embodiment;

[0016]

[0015] FIG4 is a schematic structural diagram of a data center energy consumption evaluation device according to an exemplary embodiment;

[0017] FIG5 is a schematic diagram of a hardware structure of an electronic device according to an exemplary embodiment;

[0018]

[0017] FIG6 is a schematic diagram showing the structure of a storage medium according to an exemplary embodiment.

[0019]

[0018] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, like numbers in different drawings represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present disclosure. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

[0020]

[0019] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the present disclosure may employ terms such as first, second, and third to describe various types of information, such information should not be limited to these terms. These terms are merely used to distinguish information of the same type from one another. For example, first information may be referred to as second information, and similarly, second information may be referred to as first information, without departing from the scope of the present disclosure. Depending on the context, the term "if" as used herein may be interpreted as "at the time of," "when," or "in response to a determination."

[0022]

[0021] As mentioned above, with the integration of new energy sources, the gap in electricity prices has widened. By exploring the energy consumption of data centers, we can fully utilize peak and valley electricity prices and absorb the power output of data centers' self-built new energy equipment.

[0023]

[0022] From an energy perspective, data centers need to both power servers and dissipate the heat generated by them. Therefore, data centers are coupled cooling and electricity systems, requiring both electricity and cooling. The equipment involved includes refrigeration equipment, cold storage equipment, power storage equipment, and servers. Furthermore, given the clean nature of new energy sources, data centers may also incorporate wind power equipment and photovoltaic equipment.

[0024]

[0023] From the perspective of computing tasks, the energy consumption characteristics of a data center are related to the computing tasks it processes. If classified according to the response time of the computing tasks, data center computing tasks can be divided into two categories: real-time tasks and batch tasks. Batch tasks, due to their relatively intensive use of computing resources, are insensitive to computing latency and have a certain degree of time flexibility. This also makes the energy consumption characteristics of the data center have a certain degree of time flexibility. In addition, considering the widening gap in electricity prices at different times, the energy consumption of the data center can be evaluated according to electricity prices at different times.

[0025]

[0024] Based on the above analysis, in order to evaluate the energy usage of a data center, the present disclosure obtains energy-related parameters of each device in the data center, energy storage cost parameters, task processing parameters of the data center's computing tasks, and unit electricity prices provided to the data center in different time periods. These data are used as model parameters to establish an energy management model. The management model includes a data center energy cost relationship, an energy constraint relationship, and a resource usage relationship, comprehensively describing the energy usage characteristics of the data center from different perspectives. By giving the data center an energy cost budget, that is, a given maximum energy cost, the management model is used to solve the data center's power usage range in different time periods. In other words, each time period corresponds to a power usage range. This helps the data center understand the power usage range within the given budget, thereby guiding the data center's power purchasing behavior in each time period.

[0026]

[0025] The technical solution of the present disclosure and how the technical solution of the present disclosure solves the aforementioned technical problems are described in detail below using specific embodiments. The several specific embodiments listed may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0027]

[0026] FIG1 is a flow chart of an embodiment of a data center energy consumption evaluation method according to an exemplary embodiment, comprising the following steps.

[0028]

[0027] Step 101: Obtain energy-related parameters, energy storage cost parameters, and task processing parameters of each device in the data center for processing computing tasks, and obtain unit electricity prices for different time periods provided to the data center.

[0029]

[0028] Step 102: Establish an energy management model for the data center based on the acquired energy-related parameters, energy storage cost parameters, task processing parameters, and unit electricity prices in different time periods.

[0030]

[0029] Step 103: Based on the energy management model and the preset maximum energy cost of the data center, calculate the power consumption range of the data center in the different time periods.

[0031]

[0030] In the disclosed embodiments, energy-related parameters and energy storage cost parameters of each device, as well as task processing parameters for processing computing tasks, can be pre-calculated based on the data center's historical energy usage data. The magnitude of these parameters is related to the devices and computing tasks within the data center. Because the devices and / or computing tasks within different data centers vary, the magnitude of these parameters varies across data centers. For example, the server loss coefficient, included in the energy-related parameters, refers to the power usage efficiency of the data center. Its calculation principle is to subtract 1 from the ratio of the data center's total power consumption to the power consumption of all servers within the data center. Therefore, the magnitude of the server loss coefficient is related to the number of servers within the data center.

[0032] The unit electricity price refers to the price per kilowatt-hour of electricity. Different time periods can be divided into 24 time periods with a 1-hour interval. Each time period corresponds to a unit electricity price. The unit electricity prices for different time periods may be the same or different. Therefore, the unit electricity prices for different time periods can be understood as the price per kilowatt-hour of electricity corresponding to each hour of the day set by the electricity sales company.

[0033]

[0032] The energy management model includes a data center energy cost equation, an energy constraint equation, and a resource usage equation. The parameters obtained above are used as known quantities in these equations, although these equations also involve unknown quantities. The energy cost equation describes the energy cost constraint of the data center. Specifically, the total energy cost of the data center is composed of the electricity purchase cost and the state costs corresponding to the initial and final state changes in the room temperature, the initial and final state changes in the energy storage of the power storage device, and the initial and final state changes in the energy storage of the cooling storage device. The energy constraint equation describes the energy constraints of the data center, namely, the need to maintain a balance between the data center's electrical energy and cooling energy. The resource usage equation describes the resource usage of the data center's computing tasks, namely, the need to maintain a balance between the computing resources used by the computing tasks and the computing resources output by the servers.

[0034] The preset maximum energy cost is the energy cost budget for the data center. When calculating the power consumption range for each time period, it is necessary to ensure that the total energy cost in the energy cost equation is less than or equal to the preset maximum energy cost. The power consumption ranges for different time periods refer to the fact that the data center will have a corresponding power consumption range for each time period in the future. The minimum value of the power consumption range represents the minimum power consumption of the data center during the corresponding time period, and the maximum value represents the maximum power consumption of the data center during the corresponding time period.

[0035]

[0034] As mentioned above, assuming that different time periods are 24 time periods divided into 24 time periods with an interval step of 1 hour, the power consumption intervals of different time periods refer to a power consumption interval corresponding to each hour of the data center in a day.

[0036]

[0035] At this point, the evaluation process shown in FIG1 is completed. By obtaining energy-related parameters of each device in the data center, energy storage cost parameters, task processing parameters of the data center's computing tasks, and unit electricity prices provided to the data center in different time periods, these data are used as model parameters to establish an energy management model. The management model includes the data center's energy cost relationship, energy constraint relationship, and resource usage relationship, and comprehensively describes the data center's energy usage characteristics from different perspectives. By giving the data center an energy cost budget, that is, a given maximum energy cost, the management model is used to solve the data center's power consumption range in different time periods. In other words, each unit electricity price in each time period corresponds to a power consumption range. This helps the data center understand the power consumption range under the given budget, thereby guiding the data center's electricity purchasing behavior in each time period.

[0037]

[0036] In some embodiments of the present disclosure, in the process of establishing the energy management model in step 102, an energy cost relationship equation can be established based on the energy storage cost parameter and the unit electricity price in different time periods, an energy constraint relationship equation can be established based on the energy-related parameters, and a resource usage relationship equation can be established based on the task processing parameters.

[0038]

[0037] Energy storage cost parameters are all cost-related parameters and are used as known quantities to establish the energy cost equation. Energy-related parameters are all parameters related to the energy of the data center, such as electrical energy parameters and cooling energy parameters. These parameters are also used as known quantities to establish the energy constraint equation. Task processing parameters are all parameters related to computing task processing and are used as known quantities to establish the resource usage equation.

[0039]

[0038] In some embodiments of the present disclosure, in a process of establishing an energy cost relationship based on energy storage cost parameters and unit electricity prices in different time periods, first, a first-state cost relationship is established using the unit temperature control cost corresponding to the computer room temperature control in the energy storage cost parameters and the computer room temperature difference between the start time period and the end time period. A second-state cost relationship is established using the unit electricity storage cost corresponding to the energy storage change of the electric storage device in the energy storage cost parameters and the energy storage difference of the electric storage device between the start time period and the end time period. A third-state cost relationship is established using the unit cooling cost corresponding to the energy storage change of the cooling device in the energy storage cost parameters and the energy storage difference of the cooling device between the start time period and the end time period. Then, an energy cost relationship is established using the first-state cost relationship, the second-state cost relationship, the third-state cost relationship, the unit electricity prices in different time periods, and the power consumption of the data center in the different time periods.

[0040]

[0039] The aforementioned start time period and end time period are respectively the start time period and the end time period of different time periods, for example, the start time period and the end time period of a day. That is, the temperature difference between the start time period and the end time period of the computer room refers to the temperature change between the beginning and the end of the computer room; the energy storage difference between the start time period and the end time period of the electric storage device refers to the energy storage change between the beginning and the end of the electric storage device; and the energy storage difference between the start time period and the end time period of the cold storage device refers to the energy storage change between the beginning and the end of the cold storage device.

[0041]

[0040] As can be seen from the above description, the energy storage cost parameters include the unit temperature control cost corresponding to the temperature control of the computer room, the unit electricity storage cost corresponding to the energy storage change of the electricity storage device, and the unit cold storage cost corresponding to the energy storage change of the cold storage device.

[0042] The specific expression of the first state cost relationship is as follows:

[0043] SC1=CT (TL) Formula 1

[0044] In the above formula 1, W is the unit temperature control cost corresponding to the computer room temperature control, which is the average value of the product of the power consumed by the refrigeration equipment when the computer room temperature drops by 1 degree Celsius and the electricity price of the power sales company, calculated based on the historical data of the data center; (Kuoweihong) is the computer room temperature difference between the start time period and the end time period; SCI is the state cost corresponding to the initial and final state changes of the computer room temperature.

[0045] The specific expression of the second state cost relationship is as follows:

[0046] In the above formula 2, ^ is the unit electricity storage cost corresponding to the energy storage change of the electricity storage device, which is the average value of the unit electricity price in different time periods set by the electricity sales company; (SOCS-SOC^) is the energy storage difference of the electricity storage device at the start time period and the end time period; SC2 is the state cost corresponding to the energy storage change of the electricity storage device at the beginning and end states.

[0047] The specific expression of the third state cost relationship is as follows: In the above formula 3, SC is the unit cooling cost corresponding to the change in the cooling energy storage device. This parameter is the average of the unit electricity prices for different time periods set by the electricity sales company divided by the refrigeration equipment conversion efficiency; (soa-soc^) is the difference in cooling energy storage device energy storage between the start time period and the end time period; and SC3 is the state cost corresponding to the change in cooling energy storage device energy storage between the initial and final states.

[0048]

[0047] Based on the above formulas 1 to 3, the energy cost relationship of the data center is specifically expressed as follows:

[0049]

[0048] In the above formula 4, a is the unit electricity price in different time periods, R is the power consumption of the data center in different time periods, XiStPt is the electricity purchase cost from the power sales company, t=1, 2, 3-24; TC is the total energy cost of the data center.

[0050]

[0049] In some embodiments of the present disclosure, in the process of establishing an energy constraint relationship based on energy-related parameters, the energy balance relationship is established using the server performance parameters and the refrigeration equipment conversion efficiency included in the energy-related parameters; the computer room temperature-related parameters among the energy-related parameters are used to establish a computer room temperature change relationship; the power storage device performance parameters among the energy-related parameters are used to establish a power storage device energy change relationship; and the cold storage device performance parameters among the energy-related parameters are used to establish a cold storage device energy change relationship.

[0051]

[0050] It can be seen that the above energy constraint relationship includes the energy balance relationship, the computer room temperature change relationship, the power storage device energy change relationship, and the cold storage device energy change relationship.

[0052]

[0051] Furthermore, as mentioned above, a data center is a cooling-electricity coupled system, which has both electricity and cooling demands. Therefore, the above energy balance equation specifically involves an electric energy balance equation, a cooling energy balance equation, a first constraint equation between server power consumption and server heat dissipation power, and a second constraint equation between the cooling power of the refrigeration equipment and the power consumption of the refrigeration equipment.

[0053]

[0052] In a specific embodiment, in the process of establishing an energy balance equation using energy-related parameters including server performance parameters and refrigeration equipment conversion efficiency, the server power consumption coefficient and no-load standby power of each server in the server performance parameters can be used to establish a server power consumption equation. The power balance equation is also established using the server power consumption equation, refrigeration equipment power, storage equipment charging power and storage equipment supply power, new energy equipment output, and data center power consumption. The cooling energy balance equation is also established using the refrigeration equipment cooling power, server heat dissipation power, storage equipment charging power and storage equipment cooling power, and data center required cooling power. The first constraint equation is established using the server loss coefficient in the server performance parameters, and the second constraint equation is established using the refrigeration equipment conversion efficiency.

[0053] The server power consumption equation is specifically expressed as follows:

[0054]

[0054] In the above formula 5, τ is the no-load standby power of server s in the data center, which is the power consumption when the processor utilization rate of the server is 0%, and varies due to differences in server performance; τ is the power consumption coefficient of server s in the data center, which is the difference between the power consumption when the processor utilization rate of the server is 100% and τ. It also varies due to differences in server performance. The serial number of the server, t=1, 2.. ...24.

[0055] Ignoring the network constraints of the server, the power balance equation is specifically expressed as follows:

[0056]

[0056] In the above formula 6, t=1, 2...24; t is the power consumption of the refrigeration equipment; t is the power consumption of the server; t is the charging power of the power storage device; t is the power supply power of the power storage device; t and t are the output of the new energy equipment (including photovoltaic output and wind power output); and t is the power consumption of the data center.

[0057] The cold energy balance equation is specifically expressed as follows:

[0058] C HVA (' c server

[0059]

[0058] In the above formula 7, t=1, 2...24; Vt' is the cooling power of the refrigeration equipment; O is the heat dissipation of the server Cooling power refers to the cooling power that needs to be output to the computer room.

[0060] The first constraint relationship between the server power consumption Ln and the server heat dissipation power Qn is specifically expressed as follows:

[0061]

[0060] In the above formula 8, * is the server loss coefficient, which is the ratio of the total power consumption of the data center to the power consumption of all servers in the data center minus 1.

[0062] The second constraint relationship between the refrigeration power Q of the refrigeration equipment and the power consumption F of the refrigeration equipment is specifically expressed as follows: Provide basic information of products.

[0063]

[0063] Since the cooling power output to the computer room will change the temperature of the computer room, the temperature change relationship of the computer room is specifically expressed as follows:

[0064]

[0064] In the above formula 10, , and , are parameters related to the room temperature. , T is the room temperature constant, specifically the time it takes for the difference between the room temperature of the data center and the outdoor ambient temperature to drop to 0.3679 times the original value. , L is the room equivalent thermal resistance, specifically the temperature drop when the cooling equipment in the data center absorbs 1J of heat. , H is the time interval step, i.e., 1 hour. , , and , are the room temperatures in two adjacent time periods. , , represents the room temperature and the outdoor ambient temperature.

[0065]

[0065] The energy change relationship of the power storage device is specifically expressed as follows:

[0066]

[0066] In the above formula 11, Ψ and Ψ are performance parameters of the energy storage device, where Ψ is the energy leakage coefficient of the energy storage device, which is provided by the basic product information provided on the nameplate of the power generation equipment used in the data center, Ψ is the charge and discharge efficiency of the energy storage device, which is also provided by the basic product information provided on the nameplate of the energy storage device used in the data center. Ψ and Ψ are the stored energy of the energy storage device in two adjacent time periods, Ψ is the charging power of the energy storage device, and Ψ is the supply power of the energy storage device.

[0067] The energy change relationship of the cold storage equipment is specifically expressed as follows:

[0068]

[0068] In the above formula 12, Q and Qt are both performance parameters of the cold storage equipment, Q is the energy leakage coefficient of the cold storage equipment, which is provided by the basic product information provided on the nameplate of the refrigeration equipment used in the data center, Q is the energy charging and discharging efficiency of the cold storage equipment, which is also provided by the basic product information provided on the nameplate of the cold storage equipment used in the data center, Q and Qt are the stored energy of the cold storage equipment in two adjacent time periods, Q is the cold storage equipment charging power, and Qt is the cold storage equipment cooling power.

[0069]

[0069] As can be seen from the above description, the energy-related parameters of the data center include the power consumption coefficient and no-load standby power of the server, the server loss coefficient, the refrigeration equipment conversion efficiency, the computer room temperature constant, the computer room equivalent thermal resistance, the time period step, the energy leakage coefficient of the power storage device, the charge and discharge efficiency of the power storage device, the energy leakage coefficient of the cold storage device, and the charge and discharge efficiency of the cold storage device.

[0070]

[0070] In some embodiments of the present disclosure, with respect to the process of establishing a resource usage relationship based on task processing parameters, resource constraints for the server to process real-time tasks and batch tasks can be established based on the maximum resource usage of the server in the task processing parameters, and a resource balance relationship for the data center to process batch tasks can be established based on the resource usage required for each serial numbered subtask in different categories of batch tasks in the task processing parameters. The different categories of batch tasks are obtained by clustering the batch tasks of the data center based on the attribute characteristics of the batch tasks, and the task processing constraints are established based on the minimum number of batch tasks of each category that need to be completed cumulatively in different time periods in the task processing parameters.

[0071] In this embodiment, the resource usage relational expression includes resource constraint conditions and a resource balance relational expression. Computational tasks in a data center include real-time tasks and batch tasks. Since batch tasks have temporal flexibility and are numerous in data centers, their computing energy consumption accounts for a significant proportion of total computing energy consumption. To reduce the complexity of resource usage relations caused by a large number of batch tasks, this embodiment clusters batch tasks and establishes a resource balance relation based on the resource usage required by each subtask in each batch task category, thereby accurately characterizing the computing flexibility of the data center.

[0072] By clustering batch tasks, batch tasks with similar characteristics can be grouped together. In addition to subtask dependencies, the attribute characteristics of batch tasks may also include resource usage, delay time, and the like. After clustering, batch tasks belonging to the same category have the same subtask dependencies. Each batch task belonging to the same category has the same number of subtasks, and the subtasks in each batch task have the same sequence number. As shown in FIG2 , a schematic diagram of subtask dependencies for batch tasks is shown. Other batch tasks belonging to the same category also have the same subtask dependencies as shown in FIG2 .

[0073]

[0073] Since batch processing tasks have a deadline for completion, each type of batch processing task has a lower limit on the number of subtasks that need to be completed in each time period. Therefore, the above-mentioned task processing constraints are specifically constraints for batch processing tasks. The task processing constraints are specifically expressed as follows: The number of subtasks with sequence number j in the batch task.

[0074]

[0075] Furthermore, since batch tasks have subtask dependencies, it is also necessary to establish constraints on the dependencies of batch tasks, which are specifically expressed as follows: Round down % to the nearest integer, and % is the completion progress of the subtask with sequence number j in the batch task of category i that has been completed cumulatively in the seventh time period.

[0075]

[0077] In an optional embodiment, the maximum resource usage of the server may include a maximum processor usage and a maximum memory usage. In the process of establishing resource constraints for the server to process real-time tasks and batch tasks based on the maximum resource usage of the server in the task processing parameters, the maximum processor usage is used to establish the processor usage constraint of the server, and the maximum memory usage is used to establish the memory usage constraint of the server.

[0076]

[0078] In this embodiment, since the computing task interacts with the server through the container structure running on the server, the processor usage and memory usage of the server are used as the constraint boundaries.

[0077]

[0079] The specific expression of the processor usage constraint condition is as follows:

[0078] In the above formula 15, % is the processor usage rate of server s processing real-time tasks, and % is the processor usage rate of server s processing real-time tasks. High processor usage, usually Black Sea = 1.

[0079]

[0081] The specific expression of the memory usage constraint condition is as follows: Formula 16

[0080]

[0082] In the above formula 16, % is the memory usage rate of server s processing real-time tasks, % is the memory usage rate of server s processing batch tasks, % is the memory usage rate of server s in time period t, % is the maximum memory usage rate, and % is usually 1.

[0081]

[0083] In an optional embodiment, in the process of establishing a resource balance relationship for processing batch tasks in a data center based on the resource usage required by each serial numbered subtask in different categories of batch tasks in the task processing parameters, a processor usage balance relationship is established using the processor usage required by each serial numbered subtask in different categories of batch tasks in the task processing parameters, and a memory usage balance relationship is established using the memory usage required by each serial numbered subtask in different categories of batch tasks in the task processing parameters.

[0082]

[0084] The required resource usage includes processor usage and memory usage.

[0083]

[0085] Specifically, the processor usage balance relationship is specifically expressed as follows: "U" is the processor usage of the server s for processing batch tasks.

[0084] The memory usage balance equation is specifically expressed as follows:

[0085]

[0088] In the above formula 18, % is the memory usage rate required by the subtask with sequence number j in the batch processing task of category i, and is the average memory usage rate of the subtask with sequence number j in the batch processing task of category i recorded by the data center in the historical period; % represents the number of completed subtasks with sequence number j in the batch processing task of category i in the tth time period; and % is the memory usage rate of the server s processing the batch processing task.

[0086]

[0089] As can be seen from the above description, the task processing parameters of the data center include the maximum processor usage and maximum memory usage of the server, the processor usage and memory usage required by the subtasks of each sequence number in different categories of batch processing tasks, and the minimum number of batch processing tasks of each category that need to be completed cumulatively in different time periods.

[0087]

[0090] In some embodiments of the present disclosure, in the process of calculating the power usage intervals for different time periods in step 103, the preset maximum energy cost and the energy cost relationship can be used to determine the energy cost constraint. Then, based on the energy cost constraint, the energy constraint relationship, and the resource usage relationship, the power usage intervals for the data center for different time periods are calculated.

[0091] The energy cost relationship, the energy constraint relationship, and the resource usage relationship are represented by Formulas 1-18. It can be seen that each formula contains at least one unknown quantity, and these unknown quantities are not fixed values. Therefore, under the condition of a given maximum energy cost, the calculation of the power usage intervals for different time periods is an optimization problem.

[0088] According to formula 1-formula 18, the optimization solution is as follows:

[0089] The maximum value of the power consumption interval in different time periods is solved: max ball, £=mL2.24 st TC < a set of 0) - formula (1.8)

[0090]

[0095] Wherein, I' is the preset maximum energy cost.

[0091]

[0096] Based on the embodiment shown in FIG1 above, FIG3 is a schematic diagram of a framework for energy consumption evaluation of a data center according to an exemplary embodiment. First, based on the historical energy consumption data of each device in the data center, energy-related parameters and energy storage cost parameters of each device and task processing parameters for processing computing tasks are obtained, and the unit electricity prices provided to the data center in different time periods are obtained; then, these obtained parameters and unit electricity prices in different time periods are input into the energy management model, and these input data are substituted into the above formulas 1 to 18; finally, using a given maximum energy cost, formulas 1 to 18 are used to solve the power consumption range for each time period to guide the data center's electricity purchasing behavior in each time period.

[0092]

[0097] The execution subject of the embodiment of the present disclosure may be an application, a service, an instance, a functional module in the form of software, a virtual machine (VM), a container or a cloud server, or a hardware device with data processing function.

[0093] (e.g., servers or terminal devices) or hardware chips (e.g., CPUs, GPUs, FPGAs, NPUs, AI accelerator cards, or DPUs). Devices implementing data center energy usage assessment can be deployed on the computing devices of the application provider providing the corresponding service, or on a cloud computing platform that provides computing power, storage, and network resources. The cloud computing platform can provide external services in the form of IaaS (Infrastructure as a Service), PaaS (Platform as a Service), SaaS (Software as a Service), or DaaS (Data as a Service). For example, a cloud computing platform providing SaaS software as a service can utilize its own computing resources to train data center energy usage assessment models or execute data center energy usage assessment modules. The specific application architecture can be built based on service requirements. For example, the platform can provide a model-based construction service to application providers or individuals using platform resources. Furthermore, the platform can invoke the model based on data center energy usage assessment requests submitted by relevant clients or servers, implementing online or offline data center energy usage assessment functions.

[0094]

[0098] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0095]

[0099] Corresponding to the embodiment of the aforementioned data center energy usage evaluation method, the present disclosure also provides an embodiment of a data center energy usage evaluation device.

[0096]

[0100] FIG4 is a schematic diagram showing the structure of a data center energy usage evaluation device according to an exemplary embodiment. The device is used to execute the data center energy usage evaluation method provided in any of the above embodiments. As shown in FIG4, the data center energy usage evaluation device includes:

[0097]

[0101] A parameter acquisition module 510 is used to obtain energy-related parameters of each device in the data center, energy storage cost parameters, and task processing parameters of the data center's computing tasks, and obtain unit electricity prices provided to the data center in different time periods;

[0098]

[0102] A model building module 520 is configured to build an energy management model for the data center based on the energy-related parameters, the energy storage cost parameters, the task processing parameters, and the unit electricity prices in different time periods. The energy management model includes an energy cost equation, an energy constraint equation, and a resource usage equation for the data center.

[0099]

[0103] The calculation module 530 is used to calculate the power consumption range of the data center in the different time periods based on the energy management model and the preset maximum energy consumption cost of the data center.

[0100]

[0104] In an optional embodiment, the model building module 520 is specifically configured to establish the energy cost relationship according to the energy storage cost parameter and the unit electricity price in different time periods; establish the energy constraint relationship according to the energy-related parameters; and establish the resource usage relationship according to the task processing parameters.

[0101]

[0105] In an optional embodiment, each device includes an electric storage device and a cold storage device; the model building module 520 is specifically configured to, in the process of establishing the energy cost relationship according to the energy storage cost parameter and the unit electricity price of the different time periods, establish a first state cost relationship using the unit temperature control cost corresponding to the computer room temperature control in the energy storage cost parameter and the temperature difference between the computer room between the start time period and the end time period, where the start time period and the end time period are respectively the start time period and the end time period of the different time periods; establish a second state cost relationship using the unit electric storage cost corresponding to the energy storage change of the electric storage device in the energy storage cost parameter and the energy storage difference between the electric storage device between the start time period and the end time period; establish a third state cost relationship using the unit cold storage cost corresponding to the energy storage change of the cold storage device in the energy storage cost parameter and the energy storage difference between the cold storage device between the start time period and the end time period; and establish the energy cost relationship using the first state cost relationship, the second state cost relationship, the third state cost relationship, the unit electricity price of the different time periods, and the power consumption of the data center in the different time periods.

[0102]

[0106] In an optional embodiment, the devices further include a server and a refrigeration device, and the energy constraint relationship includes an energy balance relationship, a computer room temperature change relationship, an energy storage device energy change relationship, and a cold storage device energy change relationship; the model establishment module 520 is specifically configured to establish an energy balance relationship using the server performance parameters and the refrigeration device conversion efficiency included in the energy-related parameters in the process of establishing the energy constraint relationship according to the energy-related parameters; establish a computer room temperature change relationship using the computer room temperature-related parameters among the energy-related parameters; establish an energy storage device energy change relationship using the energy storage device performance parameters among the energy-related parameters; and establish a cold storage device energy change relationship using the cold storage device performance parameters among the energy-related parameters.

[0103]

[0107] In an optional embodiment, the energy balance relationship includes an electric energy balance relationship, a cooling energy balance relationship, a first constraint relationship between server power consumption and server heat dissipation power, and a second constraint relationship between refrigeration equipment cooling power and cooling equipment power consumption; the model establishment module 520 is specifically configured to, in the process of establishing the energy balance relationship using the server performance parameters and refrigeration equipment conversion efficiency included in the energy-related parameters, establish a server power consumption relationship using the power consumption coefficient and no-load standby power of each server in the server performance parameters; establish the electric energy balance relationship using the server power consumption relationship, cooling equipment power, energy storage device charging power and energy storage device supply power, new energy device output, and data center power consumption; establish the cooling energy balance relationship using the refrigeration equipment cooling power, server heat dissipation power, cold storage device charging power and cold storage device cooling power, and the required cooling power of the data center; establish the first constraint relationship using the server loss coefficient in the server performance parameters; and establish the first constraint relationship using the refrigeration equipment conversion efficiency. The second constraint relationship is established.

[0104]

[0108] In an optional embodiment, the computing task includes a real-time task and a batch task, the batch task includes multiple subtasks, and each subtask has a unique serial number. The resource usage relationship includes resource constraints, resource balance relationships, and task processing constraints. The model building module 520 is specifically used to establish resource constraints for the server to process real-time tasks and batch tasks according to the maximum resource usage of the server in the task processing parameters during the process of establishing the resource usage relationship according to the task processing parameters; establish a resource balance relationship for the data center to process batch tasks according to the resource usage required by the subtasks with different serial numbers in different categories of batch tasks in the task processing parameters; the different categories of batch tasks are obtained by clustering the batch tasks of the data center according to the attribute characteristics of the batch tasks; and establish task processing constraints according to the minimum number of batch tasks of each category that need to be completed cumulatively in different time periods in the task processing parameters.

[0105]

[0109] In an optional embodiment, the resource constraint condition includes a processor usage constraint condition and a memory usage constraint condition; the maximum resource usage includes a maximum processor usage and a maximum memory usage; the model establishment module 520 is specifically used to establish the server's processor usage constraint condition by using the maximum processor usage in the process of establishing the server's resource constraint condition for processing real-time tasks and batch tasks based on the server's maximum resource usage in the task processing parameters; and to establish the server's memory usage constraint condition by using the maximum memory usage.

[0106]

[0110] In an optional embodiment, the resource balance relationship includes a processor usage balance relationship and a memory usage balance relationship; the resource usage includes processor usage and memory usage. The model establishment module 520 is specifically used to establish the resource balance relationship for processing batch tasks in the data center based on the resource usage required by the subtasks of each sequence number in the batch tasks of different categories in the task processing parameters, using the processor usage required by the subtasks of each sequence number in the batch tasks of different categories in the task processing parameters to establish the processor usage balance relationship; and using the memory usage required by the subtasks of each sequence number in the batch tasks of different categories in the task processing parameters to establish the memory usage balance relationship.

[0107]

[0111] In an optional embodiment, the calculation module 530 is specifically used to determine the energy cost constraint condition using the preset maximum energy cost and the energy cost relationship formula; based on the energy cost constraint condition, the energy constraint relationship formula and the resource usage relationship formula, calculate the power consumption range of the data center in the different time periods.

[0108]

[0112] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0109]

[0113] As for the device embodiments, since they basically correspond to the method embodiments, reference will be made to the partial description of the method embodiments for relevant details. The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the objectives of the disclosed solution. Those skilled in the art can understand and implement the present invention without inventive effort.

[0110]

[0114] The embodiment of the present disclosure also provides an electronic device corresponding to the data center energy consumption evaluation method provided in the above embodiment to execute the above data center energy consumption evaluation method.

[0111] FIG5 is a hardware structure diagram of an electronic device according to an exemplary embodiment. The electronic device includes a communication interface 601, a processor 602, a memory 603, and a bus 604. The communication interface 601, the processor 602, and the memory 603 communicate with each other via the bus 604. The processor 602 can execute the data center energy usage assessment method described above by reading and executing machine-executable instructions corresponding to the control logic of the data center energy usage assessment method in the memory 603. The details of the method are described in the above embodiments and will not be repeated here.

[0112]

[0116] The memory 603 mentioned in the present disclosure may be any electronic, magnetic, optical, or other physical storage device, and may contain stored information, such as executable instructions, data, and the like. Specifically, the memory 603 may be RAM (Random Access Memory), flash memory, a storage drive (such as a hard disk drive), any type of storage disk (such as an optical disk, a DVD, etc.), or similar storage media, or a combination thereof. The system network element and at least one other network element are connected via at least one communication interface 601 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, and the like.

[0113]

[0117] The bus 604 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 603 is used to store programs, and the processor 602 executes the programs after receiving an execution instruction.

[0114]

[0118] The processor 602 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be completed by hardware integrated logic circuits or software instructions in the processor 602. The processor 602 may be a general-purpose processor, including a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It may implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present disclosure. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present disclosure may be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor.

[0115]

[0119] The electronic device provided in the embodiment of the present disclosure and the data center energy consumption evaluation method provided in the embodiment of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0116]

[0120] The present disclosure also provides a computer-readable storage medium corresponding to the data center energy consumption evaluation method provided in the aforementioned embodiment. Please refer to Figure 6, where the computer-readable storage medium is a CD 30 on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the data center energy consumption evaluation method provided in any of the aforementioned embodiments.

[0117]

[0121] It should be noted that examples of the computer-readable storage medium may also 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 optical or magnetic storage media, which are not listed here one by one.

[0118]

[0122] The computer-readable storage medium provided by the above-mentioned embodiment of the present disclosure and the data center energy consumption evaluation method provided by the embodiment of the present disclosure are based on the same inventive concept and have the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0119]

[0123] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0120]

[0124] It should also be noted that the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a..." does not preclude the presence of other identical elements in the process, method, commodity, or apparatus comprising the element.

[0121]

[0125] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present disclosure shall be included in the scope of protection of the present disclosure.

Claims

Claims 1. A data center energy consumption assessment method, comprising: Obtain energy-related parameters, energy storage cost parameters, and task processing parameters for computing tasks processed by the data center for each device in the data center, and obtain unit electricity prices for different time periods provided to the data center; establish an energy management model for the data center based on the obtained energy-related parameters, energy storage cost parameters, task processing parameters, and unit electricity prices for different time periods, wherein the energy management model includes an energy cost relationship, an energy constraint relationship, and a resource usage relationship for the data center; and calculate the power consumption range of the data center in the different time periods based on the energy management model and the preset maximum energy cost of the data center.

2. The method according to claim 1, wherein: The establishing of the energy management model of the data center based on the acquired energy-related parameters, energy storage cost parameters, task processing parameters, and unit electricity prices in different time periods includes: establishing the energy cost relationship equation based on the energy storage cost parameters and the unit electricity prices in different time periods; establishing the energy constraint relationship equation based on the energy-related parameters; and establishing the resource usage relationship equation based on the task processing parameters.

3. The method according to claim 2, wherein: The devices include power storage devices and cold storage devices; establishing the energy cost relationship equation based on the energy storage cost parameters and the unit electricity prices in different time periods includes: establishing a first-state cost relationship equation using a unit temperature control cost corresponding to computer room temperature control in the energy storage cost parameters and a temperature difference between the computer room between a start time period and an end time period, where the start time period and the end time period are, respectively, the start time period and the end time period of the different time periods; establishing a second-state cost relationship equation using a unit power storage cost corresponding to a change in energy storage of the power storage device in the energy storage cost parameters and the energy storage difference between the power storage device between the start time period and the end time period; establishing a third-state cost relationship equation using a unit cold storage cost corresponding to a change in energy storage of the cold storage device in the energy storage cost parameters and the energy storage difference between the cold storage device between the start time period and the end time period; and establishing the energy cost relationship equation using the first-state cost relationship equation, the second-state cost relationship equation, the third-state cost relationship equation, the unit electricity prices in different time periods, and the power consumption of the data center in the different time periods.

4. The method according to claim 2, wherein: The devices further include servers and refrigeration equipment, and the energy constraint equations include an energy balance equation, a computer room temperature change equation, a power storage device energy change equation, and a cooling device energy change equation. The step of establishing the energy constraint relationship based on the energy-related parameters includes: establishing an energy balance relationship using server performance parameters and refrigeration equipment conversion efficiency included in the energy-related parameters; establishing a computer room temperature change relationship using computer room temperature-related parameters among the energy-related parameters; establishing an energy storage equipment energy change relationship using power storage equipment performance parameters among the energy-related parameters; and establishing an energy storage equipment energy change relationship using cold storage equipment performance parameters among the energy-related parameters.

5. The method according to claim 4, wherein: The energy balance relationship includes an electric energy balance relationship, a cooling energy balance relationship, a first constraint relationship between server power consumption and server heat dissipation power, and a second constraint relationship between refrigeration equipment cooling power and cooling equipment power consumption; the server performance parameters and refrigeration equipment conversion efficiency included in the energy-related parameters are used to establish the energy balance relationship, including: using the power consumption coefficient and no-load standby power of each server in the server performance parameters to establish a server power consumption relationship; using the server power consumption relationship, cooling equipment power, energy storage equipment charging power and energy storage equipment supply power, new energy equipment output and power consumption of the data center to establish the electric energy balance relationship; using the cooling power of the refrigeration equipment, server heat dissipation power, cold storage equipment charging power and cold storage equipment cooling power, and the required cooling power of the data center to establish the cooling energy balance relationship; using the server loss coefficient in the server performance parameters to establish the first constraint relationship; and using the refrigeration equipment conversion efficiency to establish the second constraint relationship.

6. The method according to claim 2, wherein: The computing tasks include real-time tasks and batch tasks, and the batch tasks include multiple subtasks, each of which has a unique serial number. The resource usage relationship includes resource constraints, resource balance relationships, and task processing constraints. The resource usage relationship is established based on the task processing parameters, including: establishing resource constraints for the server to process real-time tasks and batch tasks based on the maximum resource usage of the server in the task processing parameters; establishing a resource balance relationship for the data center to process batch tasks based on the resource usage required for subtasks of each serial number in different categories of batch tasks in the task processing parameters; the different categories of batch tasks are obtained by clustering the batch tasks of the data center based on the attribute characteristics of the batch tasks; and establishing task processing constraints based on the minimum number of batch tasks of each category that need to be completed cumulatively in different time periods in the task processing parameters.

7. The method according to claim 6, wherein: The resource constraints include processor usage constraints and memory usage constraints; the maximum resource usage includes maximum processor usage and maximum memory usage. Usage rate; the establishment of resource constraints for the server to process real-time tasks and batch tasks based on the maximum resource usage of the server in the task processing parameters includes: using the maximum processor usage rate to establish a processor usage constraint condition of the server; using the maximum memory usage rate to establish a memory usage constraint condition of the server.

8. The method according to claim 6, wherein: The resource balance equation includes a processor usage balance equation and a memory usage balance equation; the resource usage includes processor usage and memory usage; and establishing a resource balance equation for processing batch tasks in a data center based on the resource usage required for each serial numbered subtask in different categories of batch tasks in the task processing parameters includes: establishing the processor usage balance equation using the processor usage required for each serial numbered subtask in different categories of batch tasks in the task processing parameters; and establishing the memory usage balance equation using the memory usage required for each serial numbered subtask in different categories of batch tasks in the task processing parameters.

9. The method according to claim 1, wherein: The calculation of the power consumption range of the data center in the different time periods based on the energy management model and the preset maximum energy consumption cost of the data center includes: determining the energy cost constraint condition by using the preset maximum energy consumption cost and the energy cost relationship formula; and calculating the power consumption range of the data center in the different time periods based on the energy cost constraint condition, the energy constraint relationship formula and the resource usage relationship formula.

10. A data center energy consumption assessment device, comprising: A parameter acquisition module is configured to acquire energy-related parameters, energy storage cost parameters, and task processing parameters for computing tasks processed by the data center for each device in the data center, thereby obtaining unit electricity prices for the data center in different time periods. A model establishment module is configured to establish an energy management model for the data center based on the energy-related parameters, energy storage cost parameters, task processing parameters, and unit electricity prices for different time periods. The energy management model includes an energy cost equation, an energy constraint equation, and a resource usage equation for the data center. A calculation module is configured to calculate the power consumption ranges of the data center in the different time periods based on the energy management model and a preset maximum energy cost for the data center.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor executes the program to implement the method according to any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, wherein: The program is executed by a processor to implement the method according to any one of claims 1 to 9.

Images