Load migration energy-saving calculation method, device and equipment and readable storage medium

Abstract

The invention provides a load migration energy-saving calculation method, device and equipment and a readable storage medium, and the method comprises the steps: obtaining a corresponding relation between a hardware occupancy rate and power consumption as a power consumption curve according to a preset algorithm; obtaining service operation parameters and hardware performance parameters of the emigration server and each immigration server, and calculating expected power consumption of each immigration server according to the service operation parameters and the hardware performance parameters; and obtaining expected energy-saving data for executing load migration according to the current power consumption and the expected power consumption of each immigration server. According to the technical scheme, the corresponding relation between the hardware occupancy rate and the power consumption of each server in the server cluster is accurately obtained, the power consumption curve is formed, then the power consumption changes of the servers before and after load migration are scientifically evaluated, and effective management and optimization of the energy consumption of the server cluster are achieved.

Description

Technical Field

[0001] The present specification relates to the field of communication technology, and in particular to a load migration energy-saving calculation method, device, equipment and readable storage medium. Background Art

[0002] The energy consumption of data centers mainly comes from the continuous operation of servers, storage devices and network devices. According to statistics, the utilization rate of servers is generally low, averaging between 5-15%, while the utilization rate of processors is around 10-20%. Even when servers are idle, they still consume a lot of electricity. A standard server consumes 30-40% of its maximum power even when completely idle. This phenomenon leads to a huge waste of energy and increases the operating costs of data centers.

[0003] To solve this problem, the industry has been exploring ways to improve the energy efficiency of data centers. One effective means is to reasonably allocate running tasks to the devices in the data center to ensure that CPU resources are fully utilized, thereby shutting down idle devices to reduce energy consumption.

[0004] At present, there are some technical solutions that attempt to achieve energy saving through server load migration. For example, when device A runs e-commerce, telecommunications and other services during the day, the processor utilization rate is high; while at night, the business volume decreases and the utilization rate decreases. At the same time, device B may run offline services such as computing services, statistical reports, and data processing at night, and its processor utilization rate is higher at night. By migrating the services on device B to device A, it is possible to ensure that the processor utilization rate remains at a high level throughout the day, while shutting down device B to save energy.

[0005] Although the above solutions can achieve energy saving to a certain extent, they have certain limitations in calculating energy saving. Existing technical solutions usually only consider the lowest power during server operation or the current power during load migration to estimate energy saving, but do not consider the decrease in the power of the original server and the increase in the power of the target server after load migration. This calculation method cannot accurately reflect the actual energy saving effect. Summary of the invention

[0006] In view of this, the present specification provides a load migration energy saving calculation method, device, electronic device, and readable storage medium to improve the above-mentioned problem of inaccurate energy saving calculation.

[0007] The specific technical solutions are as follows:

[0008] The present specification provides a load migration energy-saving calculation method, which is applied to a server management device in a server cluster, wherein the server cluster also includes a plurality of servers, and the method includes: selecting a server to be managed in the server cluster as a target server, obtaining hardware occupancy data and power consumption data of the target server within a preset time period, obtaining the corresponding relationship between the hardware occupancy and power consumption of the target server as a power consumption curve according to a preset algorithm, and looping through this step until the power consumption curves of all servers to be managed are obtained; selecting a server to be managed as a migration-out server, and selecting at least one server to be managed as a migration-in server, obtaining business operation parameters and hardware performance parameters of the migration-out server and each migration-in server, and evaluating the expected hardware occupancy of each migration-in server after the business load currently running on the migration-out server is migrated to the migration-in server based on the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each migration-in server based on the expected hardware occupancy and power consumption curve; obtaining the current power consumption of the migration-out server and each migration-in server, and obtaining expected energy-saving data for executing load migration based on the current power consumption of the migration-out server and each migration-in server and the expected power consumption of each migration-in server.

[0009] As a technical solution, the hardware occupancy data and power consumption data of the target server within a preset time period are obtained, and the corresponding relationship between the hardware occupancy rate and the power consumption of the target server is obtained as a power consumption curve according to a preset algorithm, including: obtaining the hardware occupancy data and power consumption data of the target server within a preset time period, eliminating abnormal data according to preset rules, using the least squares method to perform curve fitting, and obtaining the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve.

[0010] As a technical solution, the method comprises: selecting a server to be managed as a migration server, selecting at least one server to be managed as a migration server, obtaining the business operation parameters and hardware performance parameters of the migration server and each migration server, and evaluating the expected hardware occupancy rate of each migration server after the business load currently running on the migration server is migrated to the migration server according to the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each migration server according to the expected hardware occupancy rate and power consumption curve, including: selecting a server to be managed as a migration server, selecting two servers to be managed as migration servers, obtaining the business operation parameters and hardware performance parameters of the migration server and the two migration servers, and evaluating the expected hardware occupancy rate and corresponding expected power consumption of the two migration servers under different allocation methods after the business load currently running on the migration server is allocated and migrated to the two migration servers according to the business operation parameters and hardware performance parameters in different allocation methods.

[0011] As a technical solution, the current power consumption of the outgoing server and each incoming server is obtained, and the expected energy saving data for executing load migration is obtained based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, including: the expected energy saving data for executing load migration is obtained based on the expected power consumption corresponding to the allocation method with the smallest total expected power consumption values ​​of the two incoming servers under different allocation methods.

[0012] As a technical solution, the current power consumption of the outgoing server and each incoming server is obtained, and based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, the expected energy saving data for executing load migration is obtained, including: if the expected energy saving data is greater than a preset threshold, load migration is executed, and after the load migration is completed, the outgoing server is shut down.

[0013] The present specification also provides a load migration energy-saving calculation device, which is applied to a server management device in a server cluster, wherein the server cluster also includes a plurality of servers, and the device includes: a first module, which is used to select a server to be managed in the server cluster as a target server, obtain hardware occupancy data and power consumption data of the target server within a preset time period, obtain the corresponding relationship between the hardware occupancy and power consumption of the target server as a power consumption curve according to a preset algorithm, and execute this step repeatedly until the power consumption curves of all servers to be managed are obtained; a second module, which is used to select a server to be managed as a migration-out server, and select at least one server to be managed as a migration-in server, obtain the business operation parameters and hardware performance parameters of the migration-out server and each migration-in server, and evaluate the expected hardware occupancy of each migration-in server after the business load currently running on the migration-out server is migrated to the migration-in server according to the business operation parameters and hardware performance parameters, and calculate the expected power consumption of each migration-in server according to the expected hardware occupancy and power consumption curve; a third module, which is used to obtain the current power consumption of the migration-out server and each migration-in server, and obtain the expected energy-saving data for executing load migration according to the current power consumption of the migration-out server and each migration-in server and the expected power consumption of each migration-in server.

[0014] As a technical solution, the hardware occupancy data and power consumption data of the target server within a preset time period are obtained, and the corresponding relationship between the hardware occupancy rate and the power consumption of the target server is obtained as a power consumption curve according to a preset algorithm, including: obtaining the hardware occupancy data and power consumption data of the target server within a preset time period, eliminating abnormal data according to preset rules, using the least squares method to perform curve fitting, and obtaining the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve.

[0015] As a technical solution, the method comprises: selecting a server to be managed as a migration server, selecting at least one server to be managed as a migration server, obtaining the business operation parameters and hardware performance parameters of the migration server and each migration server, and evaluating the expected hardware occupancy rate of each migration server after the business load currently running on the migration server is migrated to the migration server according to the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each migration server according to the expected hardware occupancy rate and power consumption curve, including: selecting a server to be managed as a migration server, selecting two servers to be managed as migration servers, obtaining the business operation parameters and hardware performance parameters of the migration server and the two migration servers, and evaluating the expected hardware occupancy rate and corresponding expected power consumption of the two migration servers under different allocation methods after the business load currently running on the migration server is allocated and migrated to the two migration servers according to the business operation parameters and hardware performance parameters in different allocation methods.

[0016] As a technical solution, the current power consumption of the outgoing server and each incoming server is obtained, and the expected energy saving data for executing load migration is obtained based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, including: the expected energy saving data for executing load migration is obtained based on the expected power consumption corresponding to the allocation method with the smallest total expected power consumption values ​​of the two incoming servers under different allocation methods.

[0017] As a technical solution, the current power consumption of the outgoing server and each incoming server is obtained, and based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, the expected energy saving data for executing load migration is obtained, including: if the expected energy saving data is greater than a preset threshold, load migration is executed, and after the load migration is completed, the outgoing server is shut down.

[0018] This specification also provides an electronic device, including a processor and a readable storage medium, wherein the readable storage medium stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the aforementioned load migration energy-saving calculation method.

[0019] This specification also provides a readable storage medium, which stores machine executable instructions. When the machine executable instructions are called and executed by a processor, the machine executable instructions prompt the processor to implement the aforementioned load migration energy-saving calculation method.

[0020] The above technical solution provided in this specification brings at least the following beneficial effects:

[0021] By accurately obtaining the corresponding relationship between the hardware occupancy rate and power consumption of each server in the server cluster, a power consumption curve is formed, and then the power consumption changes of the server before and after load migration are scientifically evaluated to achieve effective management and optimization of the energy consumption of the server cluster. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the implementation methods of this specification or the technical solutions in the prior art, the drawings required for use in the implementation methods of this specification or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some implementation methods recorded in this specification. For ordinary technicians in this field, other drawings can also be obtained based on these drawings of the implementation methods of this specification.

[0023] Figure 1 is a flow chart of a load migration energy saving calculation method in one embodiment of this specification;

[0024] Figure 2 is a structural diagram of a load migration energy-saving computing device in one embodiment of the present specification;

[0025] Figure 3 It is a hardware structure diagram of an electronic device in one implementation mode of this specification.

[0026] Reference numerals: first module 21 , second module 22 , third module 23 . DETAILED DESCRIPTION

[0027] The terms used in the embodiments of this specification are only for the purpose of describing specific embodiments, rather than limiting this specification. The singular forms of "a", "said" and "the" used in this specification and claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to any or all possible combinations of one or more associated listed items.

[0028] It should be understood that although the terms first, second, third, etc. may be used to describe various information in the embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, in addition, the word "if" used may be interpreted as "at the time of" or "when" or "in response to determining".

[0029] With the advent of the digital age, data centers are becoming increasingly important as key infrastructure supporting services and applications in all walks of life. However, while data centers provide efficient services, they also face the problem of low resource utilization. According to statistics, the utilization of IT equipment inside data centers, especially servers, is generally not high. Generally, server utilization is only 5% to 15%, and processor utilization is only maintained at around 10% to 20%. This means that a large amount of computing resources are idle. Even so, these devices still consume considerable energy, and even when there is no load at all, they can consume 30% to 40% of their maximum power.

[0030] To address this problem, an effective strategy to reduce data center energy consumption is to reasonably allocate tasks to devices in the data center, ensure that CPU resources can be fully utilized, and shut down idle devices in a timely manner. For example, in some cases, device A may be responsible for running services such as e-commerce and telecommunications services. During the day, these services are in high demand, resulting in high CPU utilization. At night, due to reduced business demand, CPU utilization also decreases. At the same time, device B may be responsible for computing-intensive tasks, statistical data reporting, or offline services such as data processing. Such tasks are often executed at night according to user habits, making the CPU utilization of device B higher at night and lower during the day. Based on this difference in usage patterns, it is possible to consider migrating the services on device B to device A to ensure that the CPU utilization can be maintained at a high level throughout the day, and shut down device B after the migration is completed, so as to achieve the purpose of energy saving and emission reduction.

[0031] In some technical solutions, there are certain limitations on how to evaluate and calculate the energy-saving benefits brought about by such load migration. For example, some solutions only estimate the energy-saving benefits over a period of time based on the lowest power of the server during operation or the current power when the load migration occurs. A significant drawback of this method is that it does not take into account the change in the power of the device itself after the load migration, and the possible increase in the power of the device carrying the load after the migration, so the calculated energy-saving benefits are not accurate enough.

[0032] In view of the above problems, this specification proposes a new energy-saving calculation scheme, which accurately calculates the energy-saving effect of server shutdown after load migration by establishing a first-order polynomial relationship model between the total power and CPU usage of servers in the data center. This scheme not only takes into account the power changes after server load migration, but also combines the actual power consumption of the equipment in different states, thereby providing a more accurate energy saving evaluation method.

[0033] The specific technical solution is described below.

[0034] In one embodiment, the present specification provides a load migration energy-saving calculation method, which is applied to a server management device in a server cluster, wherein the server cluster also includes a plurality of servers, and the method includes: selecting a server to be managed in the server cluster as a target server, obtaining hardware occupancy data and power consumption data of the target server within a preset time period, obtaining the corresponding relationship between the hardware occupancy and power consumption of the target server as a power consumption curve according to a preset algorithm, and looping through this step until the power consumption curves of all servers to be managed are obtained; selecting a server to be managed as an outgoing server, and selecting at least one other server to be managed as an incoming server, obtaining business operation parameters and hardware performance parameters of the outgoing server and each incoming server, and evaluating the expected hardware occupancy of each incoming server after the business load currently running on the outgoing server is migrated to the incoming server based on the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each incoming server based on the expected hardware occupancy and power consumption curve; obtaining the current power consumption of the outgoing server and each incoming server, and obtaining expected energy-saving data for performing load migration based on the current power consumption of the outgoing server and each incoming server, and the expected power consumption of each incoming server.

[0035] Specifically, Figure 1 , including the following steps:

[0036] Step S11, select a server to be managed in the server cluster as the target server, obtain the hardware occupancy data and power consumption data of the target server within a preset time period, obtain the corresponding relationship between the hardware occupancy and power consumption of the target server as the power consumption curve according to the preset algorithm, and repeat this step until the power consumption curves of all servers to be managed are obtained.

[0037] A server in the server cluster is selected as the target server. The hardware occupancy data and power consumption data of the target server in the preset time period are obtained by the centralized management software working in conjunction with HDM2 (Hardware Device Management2). Specifically, the centralized management software will collect information such as the number of CPUs, main frequency, and number of cores of the device when taking over HDM2. The HDM2 system regularly collects the CPU usage in the operating system, and the centralized management software synchronizes the CPU usage, fan mode, and current total power of all HDM2 systems in the data center every 30 seconds through the Redfish interface. In order to ensure the consistency and accuracy of the data, it is necessary to continuously collect 24 hours of CPU usage, device power, and fan mode data, totaling 2880 records. Then, these data are grouped by CPU usage, and the box plot method is used to filter abnormal data. Finally, the relationship curve between the hardware occupancy rate and power consumption of the target server, that is, the power consumption curve, is fitted by the least squares method. This process needs to be repeated for each server until the power consumption curves of all servers to be managed are obtained.

[0038] Step S12, select a server to be managed as the outgoing server, select at least one other server to be managed as the incoming server, obtain the business operation parameters and hardware performance parameters of the outgoing server and each incoming server, and evaluate the expected hardware occupancy of each incoming server after the business load currently running on the outgoing server is migrated to the incoming server based on the business operation parameters and hardware performance parameters, and calculate the expected power consumption of each incoming server based on the expected hardware occupancy and power consumption curve.

[0039] Select one server from the servers to be managed as the outgoing server, and select at least one server as the incoming server. Then, obtain the business operation parameters and hardware performance parameters of the outgoing server and each incoming server. Based on these parameters, we can evaluate the expected hardware occupancy of each incoming server after the business load currently running on the outgoing server is migrated to the incoming server. The expected hardware occupancy is determined based on the business load on the outgoing server and the hardware configuration of the incoming server. For example, if there are two businesses running on the outgoing server, the computing power consumed by each business can be expressed by the formula H = N*F*C*U, where H represents computing power, N represents the number of CPUs, F represents CPU frequency, C represents the number of CPU cores, and U represents CPU usage. After migrating these businesses to the incoming server, the expected hardware occupancy will be recalculated based on the hardware configuration of the incoming server. Once the expected hardware occupancy is determined, the expected power consumption of each incoming server can be calculated based on the power consumption curve obtained previously.

[0040] Step S13, obtaining the current power consumption of the outgoing server and each incoming server, and obtaining expected energy saving data for executing load migration according to the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server.

[0041] Get the actual power consumption data of the current migrated-out server and each migrated-in server. Combined with the calculated expected power consumption, the expected energy saving data after the load migration can be calculated. For example, suppose that the CPU usage rate of the current business of device A is 20%, and the total power consumption is 260 watts; the CPU usage rate of the current business of device B is 30%, and the total power consumption is 340 watts. If the business of device A is migrated to device B, the CPU usage rate of device B increases to 50%, and the total power consumption increases to 420 watts, then the power consumption saved after device A is shut down will be 260 watts plus 340 watts minus 420 watts, that is, 180 watts. If calculated on a daily basis, the energy saved is 180 watts multiplied by 24 hours, which is equal to 4.32 kWh.

[0042] The above implementation provides a method for accurately calculating the energy-saving benefits of server load migration and shutdown. Compared with the existing technology, this solution more finely reflects the actual energy consumption changes, which helps data centers to achieve energy conservation and emission reduction more effectively.

[0043] In one embodiment, a server to be managed is selected from a server cluster as a target server. Within a preset time period, the hardware occupancy data (such as CPU and memory usage) and power consumption data of the target server are collected. The collected data is processed using a preset algorithm (such as the least squares method) to establish a corresponding relationship between the hardware occupancy and the power consumption to form a power consumption curve. The above steps are repeated for all servers to be managed in the server cluster to obtain the power consumption curve of each server.

[0044] Select one server from the server cluster as the outgoing server and at least one server as the incoming server. Collect the business operation parameters (such as the current business load) and hardware performance parameters (such as the number of CPUs, main frequency, and number of cores) of the outgoing server and each incoming server. Based on the business operation parameters and hardware performance parameters, evaluate the expected hardware occupancy rate after the business load currently running on the outgoing server is migrated to each incoming server. Calculate the expected power consumption of each incoming server using the expected hardware occupancy rate and corresponding power consumption curve of each incoming server.

[0045] Collect the current power consumption data of the migrated-out server and each migrated-in server. Calculate the expected energy saving data after the load migration based on the current power consumption of the migrated-out server, the current power consumption of each migrated-in server, and the expected power consumption.

[0046] Assume that there are three servers in the server cluster: Server A, Server B, and Server C. Server A runs e-commerce business during the day, and the CPU usage and power consumption data are as follows: CPU usage: 20%, power consumption: 260W; CPU usage: 30%, power consumption: 300W; CPU usage: 40%, power consumption: 340W.

[0047] By fitting the least square method, we can get the power consumption curve of Server A as follows:

[0048] PA=50UA+200PA=50UA+200.

[0049] Server B and Server C run computing services at night. The power consumption data is as follows:

[0050] Server B: CPU usage 20%, power consumption: 220 W; CPU usage 30%, power consumption: 260 W; CPU usage 40%, power consumption: 300 W.

[0051] Fitting power consumption curve: PB=40UB+180PB=40UB+180.

[0052] Server C: CPU usage 20%, power consumption: 230 W; CPU usage 30%, power consumption: 270 W; CPU usage 40%, power consumption: 310 W.

[0053] Fitting power consumption curve: PC=45UC+185PC=45UC+185.

[0054] During the peak period, migrate the services of Server A to Server B and Server C. Assume that after the migration, the CPU usage of Server B increases to 50% and the CPU usage of Server C increases to 60%.

[0055] Expected power consumption of Server B: PB′=40×50%+180=310W PB′=40×50%+180=310W;

[0056] Expected power consumption of Server C: PC′=45×60%+185=337WPC′=45×60%+185=337W.

[0057] Total power consumption before migration:

[0058] Pbefore=PA+PB+PC=260+220+230=710WPbefore=PA+PB+PC=260+220+230=710W;

[0059] Total power consumption after migration:

[0060] Pafter=PB′+PC′=310+337=647WPafter=PB′+PC′=310+337=647W;

[0061] Expected energy savings:

[0062] ΔP=Pbefore-Pafter=710-647=63W ΔP=Pbefore-Pafter=710-647=63W.

[0063] Therefore, through load migration, an energy saving of 63W is expected, which significantly reduces the energy consumption of the data center.

[0064] In one embodiment, the hardware occupancy data and power consumption data of the target server within a preset time period are obtained, and according to a preset algorithm, the corresponding relationship between the hardware occupancy rate and the power consumption of the target server is obtained as a power consumption curve, including: obtaining the hardware occupancy data and the power consumption data of the target server within a preset time period, eliminating abnormal data according to preset rules, using the least squares method to perform curve fitting, and obtaining the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve.

[0065] In one embodiment, the method comprises: selecting a server to be managed as an outgoing server, selecting at least one other server to be managed as an incoming server, obtaining business operation parameters and hardware performance parameters of the outgoing server and each incoming server, evaluating the expected hardware occupancy of each incoming server after the business load currently running on the outgoing server is migrated to the incoming server based on the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each incoming server based on the expected hardware occupancy and power consumption curve, including: selecting a server to be managed as an outgoing server, selecting two other servers to be managed as incoming servers, obtaining business operation parameters and hardware performance parameters of the outgoing server and the two incoming servers, and evaluating the expected hardware occupancy and corresponding expected power consumption of the two incoming servers under different allocation methods after the business load currently running on the outgoing server is allocated and migrated to the two incoming servers based on the business operation parameters and hardware performance parameters in different allocation methods.

[0066] In one embodiment, the current power consumption of the outgoing server and each incoming server is obtained, and the expected energy saving data for executing load migration is obtained based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, including: the expected energy saving data for executing load migration is obtained based on the expected power consumption corresponding to the allocation method with the smallest total expected power consumption values ​​of the two incoming servers under different allocation methods.

[0067] In one embodiment, the current power consumption of the outgoing server and each incoming server is obtained, and based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, expected energy saving data for executing load migration is obtained, including: if the expected energy saving data is greater than a preset threshold, load migration is executed, and after the load migration is completed, the outgoing server is shut down.

[0068] In one embodiment, Figure 2 The present specification also provides a load migration energy-saving calculation device, which is applied to a server management device in a server cluster, wherein the server cluster also includes a plurality of servers, and the device includes: a first module, which is used to select a server to be managed in the server cluster as a target server, obtain hardware occupancy data and power consumption data of the target server within a preset time period, obtain the corresponding relationship between the hardware occupancy and power consumption of the target server as a power consumption curve according to a preset algorithm, and execute this step repeatedly until the power consumption curves of all servers to be managed are obtained; a second module, which is used to select a server to be managed as a migration-out server, and select at least one server to be managed as a migration-in server, obtain the business operation parameters and hardware performance parameters of the migration-out server and each migration-in server, and evaluate the expected hardware occupancy of each migration-in server after the business load currently running on the migration-out server is migrated to the migration-in server according to the business operation parameters and hardware performance parameters, and calculate the expected power consumption of each migration-in server according to the expected hardware occupancy and power consumption curve; a third module, which is used to obtain the current power consumption of the migration-out server and each migration-in server, and obtain the expected energy-saving data for executing load migration according to the current power consumption of the migration-out server and each migration-in server and the expected power consumption of each migration-in server.

[0069] In one embodiment, the hardware occupancy data and power consumption data of the target server within a preset time period are obtained, and according to a preset algorithm, the corresponding relationship between the hardware occupancy rate and the power consumption of the target server is obtained as a power consumption curve, including: obtaining the hardware occupancy data and the power consumption data of the target server within a preset time period, eliminating abnormal data according to preset rules, using the least squares method to perform curve fitting, and obtaining the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve.

[0070] In one embodiment, the method comprises: selecting a server to be managed as an outgoing server, selecting at least one other server to be managed as an incoming server, obtaining business operation parameters and hardware performance parameters of the outgoing server and each incoming server, evaluating the expected hardware occupancy of each incoming server after the business load currently running on the outgoing server is migrated to the incoming server based on the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each incoming server based on the expected hardware occupancy and power consumption curve, including: selecting a server to be managed as an outgoing server, selecting two other servers to be managed as incoming servers, obtaining business operation parameters and hardware performance parameters of the outgoing server and the two incoming servers, and evaluating the expected hardware occupancy and corresponding expected power consumption of the two incoming servers under different allocation methods after the business load currently running on the outgoing server is allocated and migrated to the two incoming servers based on the business operation parameters and hardware performance parameters in different allocation methods.

[0071] In one embodiment, the current power consumption of the outgoing server and each incoming server is obtained, and the expected energy saving data for executing load migration is obtained based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, including: the expected energy saving data for executing load migration is obtained based on the expected power consumption corresponding to the allocation method with the smallest total expected power consumption values ​​of the two incoming servers under different allocation methods.

[0072] In one embodiment, the current power consumption of the outgoing server and each incoming server is obtained, and based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, expected energy saving data for executing load migration is obtained, including: if the expected energy saving data is greater than a preset threshold, load migration is executed, and after the load migration is completed, the outgoing server is shut down.

[0073] The device implementation is the same as or similar to the corresponding method implementation, and will not be described in detail here.

[0074] In one embodiment, this specification provides an electronic device, including a processor and a readable storage medium, wherein the readable storage medium stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the aforementioned load migration energy saving calculation method. From the hardware level, the hardware architecture diagram can be seen in Figure 3 shown.

[0075] In one embodiment, the present specification provides a readable storage medium, wherein the readable storage medium stores machine executable instructions, and when the machine executable instructions are called and executed by a processor, the machine executable instructions prompt the processor to implement the aforementioned load migration energy-saving calculation method.

[0076] Here, the readable storage medium can be any electronic, magnetic, optical or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, the readable storage medium can be: RAM (Radom Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as optical disk, DVD, etc.), or similar storage medium, or a combination thereof.

[0077] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which may be in the form of a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email transceiver, a game console, a tablet computer, a wearable device or a combination of any of these devices.

[0078] 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.

[0079] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may be in 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 code.

[0080] This specification is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of 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 processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device 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.

[0081] Moreover, these computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating 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.

[0083] It will be appreciated by those skilled in the art that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may be in the form of a computer program product implemented on one or more computer-usable storage media (which may include but are not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The above description is only an embodiment of the present specification and is not intended to limit the present specification. For those skilled in the art, the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification shall be included in the scope of the claims of the present specification.

Claims

1. A load migration energy saving calculation method, characterized in that: A server management device applied to a server cluster, wherein the server cluster further comprises a plurality of servers, wherein the method comprises: Select a server to be managed in the server cluster as the target server, obtain the hardware occupancy data and power consumption data of the target server within a preset time period, obtain the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as the power consumption curve according to the preset algorithm, and repeat this step until the power consumption curves of all servers to be managed are obtained; Select a server to be managed as the outgoing server, select at least one other server to be managed as the incoming server, obtain the business operation parameters and hardware performance parameters of the outgoing server and each incoming server, evaluate the expected hardware occupancy rate of each incoming server after the business load currently running on the outgoing server is migrated to the incoming server based on the business operation parameters and hardware performance parameters, and calculate the expected power consumption of each incoming server based on the expected hardware occupancy rate and power consumption curve; The current power consumption of the outgoing server and each incoming server is obtained, and the expected energy saving data of executing the load migration is obtained according to the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server.

2. The method according to claim 1, characterized in that The step of obtaining the hardware occupancy rate data and the power consumption data of the target server within a preset time period, and obtaining the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve according to a preset algorithm, includes: The hardware occupancy rate and power consumption data of the target server within a preset time period are obtained, abnormal data are eliminated according to preset rules, and the least squares method is used for curve fitting to obtain the corresponding relationship between the hardware occupancy rate and power consumption of the target server as a power consumption curve.

3. The method according to claim 1, characterized in that The method comprises: selecting a server to be managed as a migration server, selecting at least one server to be managed as a migration server, obtaining business operation parameters and hardware performance parameters of the migration server and each migration server, evaluating the expected hardware occupancy rate of each migration server after the business load currently running on the migration server is migrated to the migration server according to the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each migration server according to the expected hardware occupancy rate and the power consumption curve, including: A server to be managed is selected as the out-of-port server, and two other servers to be managed are selected as the in-port servers. The business operation parameters and hardware performance parameters of the out-of-port server and the two in-port servers are obtained. According to the business operation parameters and hardware performance parameters, the expected hardware occupancy rate and corresponding expected power consumption of the two in-port servers under different allocation methods are evaluated after the business load currently running on the out-of-port server is allocated and migrated to the two in-port servers in different allocation methods.

4. The method according to claim 3, characterized in that The step of obtaining the current power consumption of the outgoing server and each incoming server, and obtaining expected energy saving data for executing load migration according to the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, includes: According to the expected power consumption corresponding to the allocation method with the smallest total expected power consumption values ​​of the two migrated servers under different allocation methods, the expected energy saving data of executing the load migration is obtained.

5. The method according to claim 1, characterized in that The step of obtaining the current power consumption of the outgoing server and each incoming server, and obtaining expected energy saving data for executing load migration according to the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, includes: If the expected energy-saving data is greater than a preset threshold, load migration is performed, and after the load migration is completed, the migrated server is shut down.

6. A load migration energy-saving computing device, characterized in that: A server management device applied to a server cluster, wherein the server cluster also includes a plurality of servers, and the device includes: The first module is used to select a server to be managed in the server cluster as a target server, obtain the hardware occupancy data and power consumption data of the target server within a preset time period, obtain the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve according to a preset algorithm, and repeat this step until the power consumption curves of all servers to be managed are obtained; The second module is used to select a server to be managed as a migration server, select at least one server to be managed as a migration server, obtain the business operation parameters and hardware performance parameters of the migration server and each migration server, evaluate the expected hardware occupancy rate of each migration server after the business load currently running on the migration server is migrated to the migration server according to the business operation parameters and hardware performance parameters, and calculate the expected power consumption of each migration server according to the expected hardware occupancy rate and power consumption curve; The third module is used to obtain the current power consumption of the outgoing server and each incoming server, and obtain the expected energy saving data for executing load migration based on the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server.

7. The device according to claim 6, characterized in that The step of obtaining the hardware occupancy rate data and the power consumption data of the target server within a preset time period, and obtaining the corresponding relationship between the hardware occupancy rate and the power consumption of the target server as a power consumption curve according to a preset algorithm, includes: The hardware occupancy rate and power consumption data of the target server within a preset time period are obtained, abnormal data are eliminated according to preset rules, and the least squares method is used for curve fitting to obtain the corresponding relationship between the hardware occupancy rate and power consumption of the target server as a power consumption curve.

8. The device according to claim 6, characterized in that The method comprises: selecting a server to be managed as a migration server, selecting at least one server to be managed as a migration server, obtaining business operation parameters and hardware performance parameters of the migration server and each migration server, evaluating the expected hardware occupancy rate of each migration server after the business load currently running on the migration server is migrated to the migration server according to the business operation parameters and hardware performance parameters, and calculating the expected power consumption of each migration server according to the expected hardware occupancy rate and the power consumption curve, including: A server to be managed is selected as the out-of-port server, and two other servers to be managed are selected as the in-port servers. The business operation parameters and hardware performance parameters of the out-of-port server and the two in-port servers are obtained. According to the business operation parameters and hardware performance parameters, the expected hardware occupancy rate and corresponding expected power consumption of the two in-port servers under different allocation methods are evaluated after the business load currently running on the out-of-port server is allocated and migrated to the two in-port servers in different allocation methods.

9. The device according to claim 8, characterized in that The step of obtaining the current power consumption of the outgoing server and each incoming server, and obtaining expected energy saving data for executing load migration according to the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, includes: According to the expected power consumption corresponding to the allocation method with the smallest total expected power consumption values ​​of the two migrated servers under different allocation methods, the expected energy saving data of executing the load migration is obtained.

10. The device according to claim 6, characterized in that The step of obtaining the current power consumption of the outgoing server and each incoming server, and obtaining expected energy saving data for executing load migration according to the current power consumption of the outgoing server and each incoming server and the expected power consumption of each incoming server, includes: If the expected energy-saving data is greater than a preset threshold, load migration is performed, and after the load migration is completed, the migrated server is shut down.

11. An electronic device, characterized in that: include: A processor and a readable storage medium, wherein the readable storage medium stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement any method described in claims 1-5.

12. A readable storage medium, characterized in that: The readable storage medium stores machine executable instructions, and when the machine executable instructions are called and executed by a processor, the machine executable instructions prompt the processor to implement any one of the methods of claims 1-5.

Images