Computing equipment management method and device and computing equipment
Patent Information
- Application Number
- CN202311444914.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-01
- Publication Date
- 2025-05-06
Smart Images
Figure CN119937758A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a computing device management method, apparatus and computing device. Background Art
[0002] With the improvement of processor performance and the pursuit of low carbon in society, the pursuit of processor computing power in the field of computer technology has gradually shifted to the pursuit of energy efficiency. That is, when the processor computing power meets the business needs, the power consumption of the processor should be reduced as much as possible. This requires adjusting the processor computing power according to the business load of the processor, so as to meet the computing power requirements of the business load with the lowest power consumption and achieve a high energy efficiency.
[0003] It is difficult for computing devices to predict the business load of the processor (for example, it is difficult to predict the busy phase, idle phase and other different phases of the business). In addition, some businesses do not allow the consequences of prediction failure (prediction failure may result in insufficient computing power of the processor and no time to process business data). Therefore, it is difficult for computing devices to use appropriate computing power to process business loads, resulting in relatively low energy efficiency. Summary of the invention
[0004] Provided are a computing device management method, device and computing device, which can improve the energy efficiency ratio of a processing unit.
[0005] In a first aspect, a computing device management method is provided, wherein the computing device comprises a display screen and a processing unit cluster, wherein the processing units in the processing unit cluster have computing power, and the computing power of the processing units is adjustable; the method comprises: receiving a computing power display indication, wherein the computing power display indication is issued by a user when the computing device runs a business; in response to the computing power display indication, displaying computing power information of the processing unit cluster on the display screen, wherein the computing power information of the processing unit cluster includes first computing power information of a first processing unit in the processing unit cluster, wherein the first processing unit is used to provide part or all of the computing power required for the computing device to run the business; when receiving a computing power adjustment indication issued by the user, displaying second computing power information of the first processing unit on the display screen, wherein the computing power adjustment indication is used to instruct the computing device to adjust the computing power of the first processing unit, and the second computing power information is the computing power information of the first processing unit after the computing power adjustment. Exemplarily, the first computing power information comprises at least one of the computing power, computing power usage, computing power margin, and energy efficiency ratio of the first processing unit. Exemplarily, the computing power adjustment indication is used to instruct the computing device to execute: changing the operating frequency of the first processing unit, shutting down the first processing unit, and turning on the first processing unit.
[0006] Based on considerations such as energy saving, the industry and users hope that computing devices have a higher energy efficiency ratio, that is, the lower the power consumption of computing devices, the better, while ensuring that the business can run normally. This method can display the computing power information of the processing units in the processing unit cluster, so that users can understand whether there is room for optimization of the computing power usage strategy of the processing unit, such as whether the computing power of the processing unit meets the needs of business operation, whether there is too much redundancy, whether the energy efficiency ratio is too small, etc., so that the computing power of the processing unit can be adjusted in a targeted manner to optimize the computing power usage strategy of the processing unit and improve the energy efficiency ratio of the processing unit.
[0007] For example, the computing device can display the computing power and computing power usage of the processing unit. If the computing power usage of the processing unit accounts for too large a proportion of the computing power of the processing unit, it means that the computing power of the processing unit is difficult to meet the needs of business operation. In this case, the user can increase the computing power of the processing unit through the computing power adjustment instruction, such as increasing the operating frequency of the processing unit. For another example, the computing device can display the computing power and computing power margin of the processing unit. If the computing power margin of the processing unit accounts for too large a proportion of the computing power of the processing unit, it means that there is too much redundancy in the computing power of the processing unit. In this case, the user can reduce the computing power of the processing unit through the computing power adjustment instruction, such as reducing the operating frequency of the processing unit, so as to improve the energy efficiency ratio of the processing unit. And so on, I will not list them one by one here.
[0008] After adjusting the computing power of the processing unit, the method can display the computing power information of the processing unit after the computing power adjustment, so that the user can understand the effect of the computing power adjustment, for example, whether the computing power of the processing unit after the computing power adjustment meets the needs of business operation, whether there is still too much redundancy, the energy efficiency ratio is still too small, etc., so as to facilitate the user to continue or stop adjusting the computing power of the processing unit.
[0009] In this way, through the above method, the user can continuously adjust the computing power of the processing unit, so that the computing power usage strategy of the processing unit is continuously optimized, and the energy efficiency ratio of the processing unit is continuously improved while meeting the computing power requirements of business operations.
[0010] In one possible implementation, the computing power information of the processing unit cluster also includes the computing power information of the second processing unit in the processing unit cluster, wherein the first processing unit is used to provide a portion of the computing power required for the computing device to run the business, and the second processing unit is used to provide another portion of the computing power required for the computing device to run the business; the method also includes: receiving a policy indication issued by a user for the first processing unit and the second processing unit; and setting a business scheduling policy between the first processing unit and the second processing unit based on the policy indication.
[0011] Among them, the computing power information may include computing power margin. The business scheduling strategy is divided into a polling scheduling strategy and a strategy of giving priority to the processing unit with the largest computing power margin. The polling scheduling strategy refers to polling among multiple processing units so that the processing units in the multiple processing units have equal opportunities to execute business. The strategy of giving priority to the processing unit with the largest computing power margin refers to giving priority to using the processing unit with the largest computing power margin among multiple processing units to execute business.
[0012] In this implementation, the computing power information of the first processing unit and the computing power information of the second processing unit are displayed, so that the user can understand the computing power information of the first processing unit and the computing power information of the second processing unit, thereby issuing targeted policy instructions. For example, if the difference between the computing power margin of the first processing unit and the computing power margin of the second processing unit is small, the policy instruction is used to indicate the polling scheduling strategy. Based on the policy instruction, the method sets the service scheduling strategy between the first processing unit and the second processing unit to the polling scheduling strategy. If the difference between the computing power margin of the first processing unit and the computing power margin of the second processing unit is large, the policy instruction is used to indicate the strategy of giving priority to scheduling the processing unit with the largest computing power margin. Based on the policy instruction, the method can set the service scheduling strategy between the first processing unit and the second processing unit to the strategy of giving priority to scheduling the processing unit with the largest computing power margin.
[0013] In this way, users can set business adjustment strategies between processing units to further optimize the computing power utilization strategy of the processing units.
[0014] In a possible implementation, the method further includes: receiving an alarm condition issued by a user for the first processing unit; and issuing an alarm when the first computing power information meets the alarm condition. The alarm condition may include at least one of a preset computing power margin threshold, a preset computing power usage threshold, a preset energy efficiency ratio threshold, etc.
[0015] In this implementation, the user can set an alarm condition. When the computing power information of the processing unit meets the alarm condition, the computing device can issue an alarm to prompt the user to pay attention to the computing power information of the processing unit, so that the user can adjust the computing power of the processing unit in time, thereby realizing timely optimization of the computing power usage strategy of the processing unit.
[0016] In a possible implementation, the method further includes: receiving a limit frequency for the first processing unit issued by a user; and controlling the operating frequency of the first processing unit to be greater than or less than the limit frequency.
[0017] In this implementation, the user can set different limiting frequencies as needed. For example, in order to prevent the first processing unit from consuming too much power, the maximum operating frequency is set for the first processing unit. For another example, in order to prevent the first processing unit from being able to respond to the service in a timely manner, the minimum operating frequency is set for the first processing unit. The computing device controls the operating frequency of the first processing unit according to the limiting frequency, so that the computing power usage strategy of the processing unit meets the needs of the user.
[0018] In a possible implementation, the first processing unit is any one of the following: a processor, a computing core in a multi-core processor, and a computing core group, wherein the computing core group includes at least two computing cores.
[0019] In this implementation, computing power information can be displayed at different granularities, making it convenient for users to adjust computing power at different granularities.
[0020] In a possible implementation, the first computing power information includes computing power information of the first processing unit at multiple historical time points.
[0021] The computing power information at multiple time points reflects the changing trend of the computing power information. This implementation method allows the user to understand the changing trend of the computing power information, so that the user can adjust the computing power of the processing unit more effectively.
[0022] In a second aspect, a computing device management device is provided, wherein the computing device comprises: a display screen and a processing unit cluster, wherein the processing units in the processing unit cluster have computing power, and the computing power of the processing units is adjustable, and the device comprises: a receiving unit, for receiving a computing power display indication, wherein the computing power display indication is issued by a user when the computing device runs a business; a display unit, for responding to the computing power display indication, and displaying computing power information of the processing unit cluster on the display screen, wherein the computing power information of the processing unit cluster includes first computing power information of a first processing unit in the processing unit cluster, and the first processing unit is used to provide part or all of the computing power required for the computing device to run the business; the display unit is also used to: when receiving a computing power adjustment indication issued by the user, display second computing power information of the first processing unit on the display screen, wherein the computing power adjustment indication is used to instruct the computing device to adjust the computing power of the first processing unit, and the second computing power information is the computing power information of the first processing unit after the computing power adjustment.
[0023] In one possible implementation, the computing power information of the processing unit cluster also includes the computing power information of the second processing unit in the processing unit cluster, wherein the first processing unit is used to provide a part of the computing power required for the computing device to run the business, and the second processing unit is used to provide another part of the computing power required for the computing device to run the business; the device also includes: a setting unit; wherein the receiving unit is used to: receive a policy indication issued by a user for the first processing unit and the second processing unit; the setting unit is used to: set a business scheduling policy between the first processing unit and the second processing unit based on the policy indication.
[0024] In a possible implementation, the device also includes: an alarm unit; wherein the receiving unit is used to: receive an alarm condition issued by a user for the first processing unit; the alarm unit is used to: issue an alarm when the first computing power information meets the alarm condition.
[0025] In a possible implementation, the device further includes: a control unit; wherein the receiving unit is used to: receive a limit frequency for the first processing unit issued by a user; and the control unit is used to: control the operating frequency of the first processing unit to be greater than or less than the limit frequency.
[0026] According to a third aspect, a computing device is provided, comprising: a memory for storing an executable program; and a processor for executing the method provided in the first aspect by running the executable program.
[0027] In a fourth aspect, a computer-readable storage medium is provided, comprising computer program instructions. When the computer program instructions are executed by a computing device, the computing device executes the method provided in the first aspect.
[0028] According to a fifth aspect, a computer program product comprising instructions is provided. When the instructions are executed by a computing device, the computing device executes the method provided in the first aspect.
[0029] Among them, the beneficial effects of the second to fifth aspects can be found in the above description of the beneficial effects of the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 A schematic diagram of a system architecture provided in an embodiment of the present application;
[0031] Figure 2 A flowchart of a computing device management method provided in an embodiment of the present application;
[0032] Figure 3 A schematic diagram of the structure of a management device provided in an embodiment of the present application;
[0033] Figure 4AA schematic diagram of a calculation scheme for the computing power of a processing unit provided in an embodiment of the present application;
[0034] Figure 4B A schematic diagram of a calculation scheme for the amount of computing power used by a processing unit provided in an embodiment of the present application;
[0035] Figure 5 A schematic diagram of a user interface provided in an embodiment of the present application;
[0036] Figure 6 A schematic diagram of the structure of a computing device management device provided in an embodiment of the present application;
[0037] Figure 7 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0038] The scheme provided by the embodiment of the present application will be described below in conjunction with the accompanying drawings. In the embodiment of the present application, "plurality" refers to two or more than two. "First", "second", etc. are only used to distinguish similar objects and are not necessarily used to describe a specific order or number of objects.
[0039] To facilitate understanding of the solutions provided by the embodiments of the present application, some technical terms that may be involved in the embodiments of the present application are first introduced before the solutions provided by the embodiments of the present application are introduced.
[0040] Computing power: also known as the computing power or information processing ability of a processing unit, refers to the ability of the processing unit hardware and software to work together to perform certain computing requirements. Among them, the computing power of a processing unit can be expressed as the maximum number of floating-point operations per second (FLOPS), or as the number of reads and writes per second (input / output operations per second, IOPS). The computing power of a processing unit is related to the operating frequency of the processing unit, the chip architecture of the processing unit, the chip layout of the processing unit, etc. The computing power of a processing unit can be adjusted. Among them, the computing power of a processing unit is positively correlated with the operating frequency of the processing unit, and the computing power of the processing unit can be adjusted by adjusting the operating frequency of the processing unit.
[0041] Processing unit: refers to a computer module, component or assembly with computing power for performing one or more computing tasks. A processing unit can be a processor, a computing core in a multi-core processor, or a computing core group in a multi-core processor. Among them, the computing core group is also called a partition, including at least two computing cores located in the same multi-core processor. In addition, the processor as a processing unit can be a single-core processor or a multi-core processor. The processor can be a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC) or a neural-network processing unit (NPU), etc.
[0042] Computing power usage: refers to the computing power currently being used in the processing unit.
[0043] Computing power margin: refers to the unused computing power of a processing unit, i.e., idle computing power. The sum of the computing power margin of a processing unit and the computing power usage of the processing unit equals the computing power of the processing unit.
[0044] Energy efficiency ratio: also known as performance-to-power ratio, is an indicator that measures the amount of computing power a processing unit converts into electrical energy. The energy efficiency ratio is obtained by dividing the computing power used by the processing unit by the power consumption of the processing unit. The common dimension of energy efficiency ratio is IOPS / watt (w).
[0045] Among them, the energy efficiency ratio is positively correlated with the computing power usage, and negatively correlated with the computing power margin and power consumption. And power consumption is positively correlated with computing power. Specifically, the power consumption of the processing unit is positively correlated with the operating frequency of the processing unit, and the computing power of the processing unit is positively correlated with the operating frequency of the processing unit. Therefore, power consumption is positively correlated with computing power. In other words, under the condition of the same computing power usage, the energy efficiency ratio of the processing unit is negatively correlated with the computing power of the processing unit. Therefore, one way to improve the energy efficiency ratio of the processing unit is to reduce the computing power or computing power margin of the processing unit as much as possible while meeting the computing power requirements of the business load, that is, under the condition of the same computing power usage. Among them, the computing power of the processing unit can be reduced by reducing the operating frequency of the processing unit.
[0046] Since it is difficult for computing devices to predict the business load of the processing unit, in order to ensure the execution effect of the business, the computing power of the processing unit is maintained at a high level, which leads to relatively low energy efficiency of the processing unit. Users understand the business operation rules and know when the business is in a busy stage (i.e., the business load is large) and when it is in an idle stage (i.e., the business load is small). However, it is difficult for users to obtain the computing power usage of the processing unit (for example, whether the computing power margin of the processing unit is too large or too small, etc.), so users do not know how to adjust the computing power of the processing unit. In addition, the relevant technology does not provide a function for users to adjust the computing power of the processing unit, so users cannot participate in the adjustment of the computing power of the processing unit.
[0047] The embodiment of the present application provides a computing device management method. The method can display the computing power information of the processing unit so that the user can understand the computing power information of the processing unit. The method can also receive a computing power adjustment instruction issued by the user, and adjust the computing power of the processing unit based on the computing power adjustment instruction, and display the computing power information of the processing unit after the computing power adjustment. As a result, the user can adjust the computing power of the processing unit, and understand whether the adjusted computing power meets the needs of the business load and whether there is room for further adjustment, so that the computing power usage strategy of the processing unit can be continuously optimized, and the energy efficiency ratio of the processing unit can be improved as much as possible while meeting the computing power requirements of the business load.
[0048] Next, the computing device management method provided in the embodiment of the present application is described.
[0049] Figure 1 A system architecture that can be used to implement a computing device management method is shown. Figure 1 As shown, the system architecture includes a user and a computing device 100 .
[0050] The user may be a person who can operate the computing device 100. The user may be an ordinary consumer or an operation and maintenance personnel.
[0051] like Figure 1 As shown, computing device 100 includes display screen 110 , processing unit cluster 120 , and management device 130 .
[0052] Among them, the computing device 100 can be a stand-alone device, such as a mobile phone, a tablet computer, a desktop computer, a laptop computer, a handheld computer, a notebook computer, a vehicle-mounted device, a smart wearable device, etc. The computing device 100 can also be a computing device cluster composed of multiple computing nodes. When the computing device 100 is a computing device cluster, the display screen 110, the processing unit cluster 120 and the management device 130 can be deployed on the same computing node or on different computing nodes. In addition, different processing units in the processing unit cluster 120 can be deployed on the same computing node or on different computing nodes. The embodiment of the present application does not limit the specific implementation form of the computing device 100.
[0053] The display screen 110 can be used to display visual information such as text, images, and videos. In the embodiment of the present application, the display screen 110 can display the computing power information of the processing unit, for example, in the form of a file, an image, or a video. The display screen 110 includes a display panel. The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode or an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), Miniled, MicroLed, Micro-oLed, a quantum dot light emitting diode (QLED), and the like.
[0054] The processing unit cluster 120 may include at least one processing unit, for example, a processing unit 121, a processing unit 122, and the like. The processing units in the processing unit cluster 120 have computing power, and the computing power of the processing units is adjustable. For example, the operating frequency of the processing unit is adjustable, and the computing power of the processing unit can be adjusted by adjusting the operating frequency of the processing unit. Specifically, increasing the operating frequency of the processing unit can increase the computing power of the processing unit; reducing the operating frequency of the processing unit can reduce the computing power of the processing unit. In addition, the specific implementation of the processing unit is also referred to above, and will not be repeated here. In some embodiments, the processing unit 121 and the processing unit 122 may be isomorphic processing units, that is, the structure, maximum operating frequency, and maximum computing power of the processing unit 121 and the processing unit 122 are the same. In some embodiments, the processing unit 121 and the processing unit 122 may be heterogeneous processing units, that is, any one or more of the structure, maximum operating frequency, and maximum computing power of the processing unit 121 and the processing unit 122 are different.
[0055] The management device 130 is used to execute the computing device management method provided in the embodiment of the present application. The specific functions of the management device 130 will be introduced below and will not be repeated here. In some embodiments, the management device 130 can be implemented by software. Exemplarily, the code of the management device 130 can be executed by any one or more processing units in the processing unit cluster 120. Exemplarily, the code of the management device 130 can be executed by a processor outside the processing unit cluster 120.
[0056] The above describes the system architecture provided by the embodiment of the present application. Next, in conjunction with the system architecture, the computing device management method provided by the embodiment of the present application is introduced.
[0057] In step 201, the management device 130 receives a computing power display indication, which is issued by a user when the computing device 100 runs a service. Exemplarily, the service may be a user service, such as a live broadcast service, an image rendering service, etc. The embodiments of the present application do not specifically limit the service. While the computing device 100 is running a service, the user may issue a computing power display indication to the computing device 100. The management device 130 may receive the computing power display indication in step 201.
[0058] In step 202, the management device 130 displays the computing power information of the processing unit cluster 120 on the display screen 110, wherein the computing power information of the processing unit cluster 120 includes the computing power information A1 of the processing unit 121, and the processing unit 121 is used to provide part or all of the computing power required for the computing device 100 to run the business. That is, all or part of the computing power required for the computing device 100 to run the business can be provided by the processing unit 121. In some embodiments, all or part of the computing power currently required for the computing device 100 to run the business is provided by the processing unit 121. In some embodiments, the computing power currently required for the computing device 100 to run the business is provided by the processing unit other than the processing unit 121 in the processing unit cluster 120, but the computing device 100 can use the computing power of the processing unit 121 to run the business in the future. For example, when the computing power of the processing unit that currently provides computing power for the computing device 100 to run the business cannot meet the needs of the business operation, the processing unit 121 can provide computing power for the computing device 100 to run the business to meet the needs of the business operation.
[0059] In some embodiments, the computing power information of the processing unit cluster 120 displayed in step 201 may only include computing power information A1. In some embodiments, the computing power information of the processing unit cluster 120 displayed in step 201 may include computing power information of other processing units in addition to computing power information A1, such as computing power information of processing unit 122.
[0060] Next, taking the display computing power information A1 as an example, the computing power information of the display processing unit is introduced.
[0061] See also Figure 3 The management device 130 includes an acquisition module 131 and a data module 132. The acquisition module 131 is used to acquire the computing power calculation data of the processing unit, and the computing power calculation data refers to the data used to calculate the computing power information of the processing unit. The data module 132 is used to save the data acquired by the acquisition module 131. The details are as follows.
[0062] The computing power calculation data of the processing unit 121 includes the computing power configuration information of the processing unit 121. The computing power configuration information is also called capacity configuration information, which is static information of the configuration. Exemplarily, the manufacturer or design manufacturer of the computing device 100 can configure the computing power configuration information of the processing unit 121, and preset the computing power configuration information of the processing unit 121 to the computing device 100, for example, to the baseboard management controller (BMC) of the computing device 100. The computing power configuration information includes the maximum operating frequency and maximum computing power of the processing unit 111. Among them, the maximum computing power of the processing unit is the computing power when the processing unit runs at the maximum operating frequency. The acquisition module 131 can obtain the computing power configuration information of the processor 121 from the BMC or other components storing the computing power configuration information. The acquisition module 131 can save the acquired computing power configuration information in the data module 132.
[0063] The computing power calculation data of the processing unit 121 includes the current usage information and the current operating frequency of the processing unit 121. The acquisition module 131 can acquire the current usage information and the current operating frequency of the processing unit 121 according to the time interval. Among them, the current usage information is also called actual usage information, which is used to reflect the current actual usage of the processing unit. The current operating frequency is also called the actual operating frequency, which refers to the current actual operating frequency of the processing unit. The length of the time interval is preset, such as 100ms, 50ms. The length of the time interval can be set freely. The current usage information includes the calculation time and idle time within a time interval. The time interval is equal to the sum of the calculation time and idle time within the time interval. Among them, the calculation time includes the time consumed by the user-state computing task, the time consumed by the kernel-state computing task, and the time waiting for input output (IO). The acquisition module 131 can obtain the current usage information and the current operating frequency of the processing unit 121 from the operating system of the computing device 100. The acquisition module 131 can save the acquired current usage information and current operating frequency in the data module 132.
[0064] In some embodiments, the computing power calculation data of the processing unit 121 includes the power consumption of the processing unit 121. The acquisition module 131 may acquire the power consumption of the processing unit 121 at the above time intervals. The power consumption of the processing unit refers to the power of the processing unit, and the acquisition module 131 may acquire the power consumption of the processing unit 121 from the operating system (e.g., the kernel of the operating system). The acquisition module 131 may save the acquired power consumption of the processing unit in the data module 132.
[0065] Continue reading Figure 3 The management device 130 includes a calculation module 133. The calculation module 133 can obtain the computing power calculation data of the processing unit 121 from the data module 132, and calculate the computing power information of the processing unit 121 based on the computing power calculation data. The details are as follows.
[0066] In some embodiments, the computing power information of the processing unit 121 includes the computing power of the processing unit 121. Unless otherwise specified, the computing power of the processing unit 121 refers to the current computing power of the processing unit 121, that is, the actual computing power of the processing unit 121. Figure 4A , the calculation module 133 can obtain the computing power configuration information of the processing unit 121 from the data module 132, and obtain the maximum computing power and the maximum operating frequency of the processing unit 121 from the computing power configuration information. The calculation module 133 obtains the current operating frequency of the processing unit 121 from the data module 132. Then, the calculation module 133 calculates the computing power of the processing unit 121 based on the maximum computing power, the maximum operating frequency and the current operating frequency of the processing unit 121 through formula (1).
[0067] Hashrate = (current operating frequency ÷ maximum operating frequency) × maximum hashrate.
[0068] In some embodiments, the computing power information of the processing unit 121 includes the computing power usage of the processing unit 121. Figure 4B , the calculation module 133 can obtain the current usage information of the processing unit 121 from the data module 132, and calculate the current computing power utilization of the processing unit 121 based on the current usage information. Exemplarily, the current usage information includes the calculation duration and the idle duration in the current time interval. The current computing power utilization is obtained by dividing the calculation duration by the duration of the current time interval (that is, the sum of the calculation duration and the idle duration). Then, the current computing power utilization of the processing unit 121 is multiplied by the current computing power of the processing unit 121 to obtain the computing power usage of the processing unit 121. Among them, unless otherwise specified, the computing power usage of the processing unit 121 refers to the current computing power usage of the processing unit 121, that is, the actual computing power usage of the processing unit 121.
[0069] In some embodiments, the computing power information of the processing unit 121 includes the computing power margin of the processing unit 121. As specifically described above, the computing power margin of the processing unit 121 refers to the current computing power margin of the processing unit 121, that is, the current actual computing power margin of the processing unit 121. The calculation module 133 can calculate the computing power and computing power usage of the processing unit 121, and then subtract the computing power usage of the processing unit 121 from the computing power of the processing unit 121 to obtain the computing power margin of the processing unit 121.
[0070] In some embodiments, the computing power information of the processing unit 121 includes the energy efficiency ratio of the processing unit. As specifically described above, the energy efficiency ratio of the processing unit 121 refers to the current energy efficiency ratio of the processing unit 121, that is, the current actual energy efficiency ratio of the processing unit 121. Exemplarily, the calculation module 133 can calculate the computing power usage of the processing unit 121 and obtain the power consumption of the processing unit 121 from the data module 132. The calculation module 133 calculates the computing power usage of the processing unit 121 by the power consumption of the processing unit 121 to obtain the energy efficiency ratio of the processing unit 121.
[0071] In this way, the computing power information of the processing unit 121 can be calculated. With reference to the calculation method of the computing power information of the processing unit 121, the computing power information of other processing units in the processing unit cluster 120 can be calculated. In addition, when the processing unit 121 includes multiple computing cores, for example, when the processing unit 121 is a multi-core processor or a computing core group, the computing power information of each computing core can be calculated separately. It will not be repeated here one by one.
[0072] After obtaining the computing power information of the processing unit, the management device 130 may display the computing power information of the processing unit through the display screen 110. Next, the processing unit 121 is still taken as an example for introduction.
[0073] In some embodiments, see Figure 3 The management device 130 includes a management module 134. The management module 134 can display the computing power information of the processing unit 121 through the display screen 110. For example, the management module 134 and the display screen 110 share an area in the memory. Whenever the calculation module 132 calculates the computing power information of the processing unit 121, the management module 134 transmits the computing power information to the area. The display screen 110 obtains the computing power information from the area and displays it.
[0074] In some embodiments, see Figure 5The computing power information of the processing unit displayed on the display screen 110 may include any one of the computing power, computing power margin, computing power usage, and energy efficiency ratio of the processing unit, or any combination of multiple ones. The management device 130 may simultaneously display the computing power information of one or more processing units on the display screen 110, for example, the computing power information of the processing unit 121 and / or the computing power information of the processing unit 122 may be displayed.
[0075] In some embodiments, the management device 130 may display the trend of the computing power information of the processing unit over time on the display screen 110. That is, the computing power information of the processing unit displayed on the display screen 110 includes the computing power information of the processing unit at multiple time points in history. Figure 5 As shown, the management device 130 can display the trend of the energy efficiency ratio of the processing unit over time on the display screen 110.
[0076] Back to Figure 2 After step 201, the management device 130 may execute step 203, and display the computing power information A2 of the processing unit 121 when receiving a computing power adjustment instruction issued by the user.
[0077] The computing power adjustment instruction is used to instruct the computing device 100 to adjust the computing power of the processing unit 121. Exemplarily, the computing power adjustment instruction can instruct the computing device 100 to perform any one of: changing the operating frequency of the processing unit 121, shutting down the processing unit 121, and turning on the processing unit 121. Among them, the computing power adjustment of the processing unit 121 can be specifically performed by the management device 130. That is, while the management device 130 displays the computing power information of the processing unit through the display screen 110, it also provides the user with the function of adjusting the computing power of the processing unit. Among them, the management module 134 in the management device 130 can receive the computing power adjustment instruction issued by the user for the processing unit, and adjust the computing power of the processing unit according to the computing power adjustment instruction.
[0078] The computing power information A2 is the computing power information of the processing unit 121 after the computing power is adjusted. Specifically, the computing power information A2 is the computing power information of the processing unit 121 after the computing device 100 (specifically the management device 130) adjusts the computing power of the processing unit 121 according to the computing power adjustment instruction. That is to say, after the computing power of the processing unit 121 is adjusted, the management device 130 can calculate the computing power information A2 based on the computing power calculation data of the processing unit 121 after the computing power adjustment. The computing power information A2 includes any one or a combination of any multiple of the computing power, computing power margin, computing power usage, and energy efficiency ratio of the processing unit after the computing power adjustment. Among them, the calculation method of the computing power information A2 can be specifically implemented by referring to the above introduction to the calculation method of the computing power information A1, which will not be repeated here.
[0079] In some embodiments, Figure 5 As shown, the display screen 110 may be a touch screen, and the management module 130 may display a computing power management area on the display screen 110. The user may operate in the management area to issue a computing power adjustment instruction. Figure 5 As shown, in step 201, the display screen 110 displays the computing power margin, computing power usage, and energy efficiency ratio of the processing unit 121. Among them, the energy efficiency ratio of the processing unit 121 is relatively low, and the computing power margin is relatively large. In this case, the user can initiate a computing power adjustment instruction through the computing power management area. The computing power adjustment instruction is used to instruct to reduce the operating frequency of the processing unit 121, so as to reduce the power consumption of the processing unit 121 by reducing the operating frequency of the processing unit 121, thereby improving the energy efficiency ratio of the processing unit 121. After the management module 130 responds to the computing power adjustment instruction and reduces the operating frequency of the processing unit 121, the management module 130 can calculate the computing power information A2 and display the computing power information A2 through the display screen 110. As shown in FIG. Figure 5 As shown, after reducing the operating frequency of the processing unit 121, the computing power margin of the processing unit 121 is reduced and the energy efficiency ratio is improved.
[0080] In some embodiments, in step 201, the processing unit 121 may be in an unactivated state, at which point the computing power information A1 shows that the computing power of the processing unit 121 is zero. If the computing power usage of other processing units in the processing unit cluster 120 is too large (for example, the computing power usage is greater than a preset threshold value Y1) or the computing power margin is too small (for example, the computing power margin is less than a preset threshold value Y2), and the operating frequency of other processing units has reached the maximum operating frequency. This indicates that the business load of other processing units in the processing unit cluster 120 or the computing core group where the processing unit 121 is located is relatively heavy. In this case, the user can initiate a computing power adjustment instruction, which is used to instruct the processing unit 121 to be turned on to share the business load of other processing units.
[0081] In some embodiments, the computing power information A1 shows that the computing power margin of the processing unit 121 is too large (for example, the computing power margin is greater than the preset threshold value Y3) or the computing power usage is too small (for example, the computing power usage is less than the preset threshold value Y4). This means that the business load of the processing unit 121 is light, or even zero. In this case, the user can initiate a computing power adjustment instruction, which is used to instruct to shut down the processing unit 121 to save the power consumption of the processing unit 121, thereby improving the overall energy efficiency ratio of the processing unit cluster 120.
[0082] In some embodiments, the management module 130 can display the computing power information of multiple processing units on the display screen 110, and the multiple processing units can provide computing power for the computing device 100 to run the business. In this way, the user can set the business scheduling strategy between the multiple processing units based on the computing power information of the multiple processing units. Among them, the business scheduling strategy is divided into a polling scheduling strategy and a strategy of giving priority to the processing unit with the largest computing power margin. The polling scheduling strategy refers to polling between multiple processing units so that the processing units in the multiple processing units have equal opportunities to execute the business. The strategy of giving priority to the processing unit with the largest computing power margin refers to giving priority to using the processing unit with the largest computing power margin among multiple processing units to execute the business.
[0083] Next, the computing power information of the processing unit 121 and the computing power information of the processing unit 122 are displayed as an example for introduction.
[0084] like Figure 5 As shown, display screen 110 displays computing power information of processing unit 121 and computing power information of processing unit 122, wherein processing unit 121 is used to provide a portion of computing power required for computing device 100 to run the business, and processing unit 122 is used to provide another portion of computing power required for computing device 100 to run the business.
[0085] The computing power information displayed on the display screen 110 includes computing power margin. Thus, the user can know the computing power margin of the processing unit 121 and the computing power margin of the processing unit 122, and issue a policy indication according to the computing power margin of the processing unit 121 and the computing power margin of the processing unit 122. The policy indication is used to indicate a service scheduling policy, such as a polling scheduling policy or a policy of giving priority to scheduling the processing unit with the largest computing power margin. The management device 130 can receive the policy indication and set the service scheduling policy between the processing unit 121 and the processing unit 122.
[0086] Specifically, if the difference between the computing power margin of processing unit 121 and the computing power margin of processing unit 122 is small, the policy indication is used to indicate the polling scheduling policy. Based on the policy indication, management module 130 can set the service scheduling policy between processing unit 121 and processing unit 122 to the polling scheduling policy.
[0087] If the difference between the computing power margin of processing unit 121 and the computing power margin of processing unit 122 is large, the policy indication is used to indicate a policy of giving priority to the processing unit with the largest computing power margin. Based on the policy indication, the management module 130 can set the service scheduling policy between processing unit 121 and processing unit 122 to a policy of giving priority to the processing unit with the largest computing power margin.
[0088] Exemplarily, the size of the difference between the computing power margin of processing unit 121 and the computing power margin of processing unit 122 can be measured by a preset threshold value Y5. Specifically, the computing power margin of processing unit 121 and the computing power margin of processing unit 122 are subtracted to obtain a difference. If the absolute value of the difference is greater than the threshold value Y5, it means that the difference between the computing power margin of processing unit 121 and the computing power margin of processing unit 122 is large. If the absolute value of the difference is less than or equal to the threshold value Y5, it means that the difference between the computing power margin of processing unit 121 and the computing power margin of processing unit 122 is small.
[0089] In some embodiments, Figure 3 As shown, the management module 134 in the management device 130 also has an alarm function, which can issue an alarm when the computing power information of the processing unit is abnormal. Among them, the abnormal computing power information can be determined based on the alarm condition set by the user. Taking the processing unit 121 as an example, the user can issue an alarm condition for the processing unit 121. Exemplarily, the alarm condition may include a preset computing power margin threshold, a preset computing power usage threshold, a preset energy efficiency ratio threshold, etc. The management device 130 can record the alarm condition and determine whether the computing power information of the processing unit 121 meets the alarm condition. When the computing power information of the processing unit 121 meets the alarm condition, an alarm can be issued. For example, the alarm condition is the computing power margin threshold, and when the computing power margin of the processing unit 121 is greater than the computing power margin threshold, the management module 130 can issue an alarm. For another example, the alarm condition is a computing power usage threshold. When the computing power usage of the processing unit 121 is greater than the computing power usage threshold, the management module 130 may issue an alarm to prompt the user to adjust the computing power of the processing unit 121. For another example, the alarm condition is an energy efficiency ratio threshold. When the energy efficiency ratio of the processing unit 121 is less than or greater than the energy efficiency ratio threshold, the management module 130 may issue an alarm. The alarm issued by the management module 130 may prompt the user to adjust the computing power of the processing unit 121 to optimize the usage strategy of the processing unit 121.
[0090] In some embodiments, the user can also set the maximum or minimum operating frequency of the processing unit. Still taking the processing unit 121 as an example, the user can issue a limit frequency for the processing unit 121. The management device 130 can receive the limit frequency and limit the operating frequency of the processing unit 121 based on the limit frequency. When the limit frequency is the maximum operating frequency set by the user, the management module 130 controls the operating frequency of the processing unit 121 to be less than the limit frequency. When the limit frequency is the minimum operating frequency set by the user, the management module 130 controls the operating frequency of the processing unit 121 to be greater than the limit frequency.
[0091] In some embodiments, the processing unit 121 may be a computing core group. Under the operation of the user, the management device 130 may adjust the computing cores in the computing core group. For example, when the computing power information A1 or computing power information A2 displayed on the display screen indicates that the computing power margin of the computing core group is too small (for example, the computing power margin is less than the preset threshold value Y2), and the operating frequency of each computing core in the computing core group has reached the maximum frequency, the user may issue a computing core addition instruction. The management module 130 may respond to the instruction and add one or more computing cores to the computing core group to increase the computing power of the computing core group. For another example, when the computing power information A1 or computing power information A2 displayed on the display screen indicates that the computing power margin of the computing core group is too large (for example, the computing power margin is greater than the preset threshold value Y3), the user may issue a computing core shutdown instruction. The management module 130 may respond to the instruction and shut down one or more computing cores in the computing core group. And so on, which are not listed one by one here.
[0092] In summary, through the computing device management method provided in the embodiment of the present application, the user can understand the computing power information of the processing unit and adjust the computing power of the processing unit, so that the user can continuously optimize the computing power usage strategy of the processing unit, thereby achieving the goal of maximizing the energy efficiency ratio of the processing unit while meeting the computing power requirements of the business load.
[0093] See also Figure 6 The embodiment of the present application further provides a computing device management apparatus 600. The computing device managed by the apparatus 600 includes a display screen and a processing unit cluster. The processing units in the processing unit cluster have computing power, and the computing power of the processing units is adjustable.
[0094] like Figure 6 As shown, the device 600 includes: a receiving unit 610 and a display unit 620. The receiving unit 610 is used to receive a computing power display indication, which is issued by a user when the computing device runs a business; the display unit 620 is used to: in response to the computing power display indication, display the computing power information of the processing unit cluster on the display screen, the computing power information of the processing unit cluster includes the first computing power information of the first processing unit in the processing unit cluster, and the first processing unit is used to provide part or all of the computing power required for the computing device to run the business; the display unit 620 is also used to: when receiving a computing power adjustment indication issued by the user, display the second computing power information of the first processing unit on the display screen, wherein the computing power adjustment indication is used to instruct the computing device to adjust the computing power of the first processing unit, and the second computing power information is the computing power information of the first processing unit after the computing power adjustment.
[0095] In some embodiments, the computing power information of the processing unit cluster also includes computing power information of a second processing unit in the processing unit cluster, wherein the first processing unit is used to provide a portion of the computing power required for the computing device to run the service, and the second processing unit is used to provide another portion of the computing power required for the computing device to run the service. The apparatus 600 also includes: a setting unit 630. The receiving unit 610 is also used to: receive a policy indication issued by the user for the first processing unit and the second processing unit; the setting unit 630 is used to: set a service scheduling policy between the first processing unit and the second processing unit based on the policy indication.
[0096] In some embodiments, the device further includes 600: an alarm unit 640. The receiving unit 610 is further configured to: receive an alarm condition issued by the user for the first processing unit; and the alarm unit 640 is configured to: issue an alarm when the first computing power information meets the alarm condition.
[0097] In some embodiments, the device 600 further includes a control unit 650. The receiving unit 610 is further configured to receive a limit frequency for the first processing unit issued by the user; and the control unit 650 is configured to control the operating frequency of the first processing unit to be greater than or less than the limit frequency.
[0098] The functions of the functional units of the device 600 can also refer to the above description of Figure 2 The introduction and implementation of the various method embodiments shown will not be repeated here.
[0099] This embodiment of the application provides a computing device 700. Figure 7 As shown, the computing device 700 includes a processor 710, a memory 720, a display screen 730, and a processing unit cluster 740. The memory 720 is used to store executable programs. The processor 710 is used to execute the executable programs stored in the memory 720, so that the computing device 700 can execute the above Figure 2 The operations performed by the management device 130 in the illustrated embodiment.
[0100] In some embodiments, Figure 7 As shown, processor 710 is a computing device outside of processing unit cluster 740. In some embodiments, processor 710 is one or more processing units in processing unit cluster 740.
[0101] The present application also provides a computer program product including instructions. The computer program product may be a software or program product including instructions that can be run on a computing device or stored in any available medium. When the computer program product is run on a computing device, the computing device is caused to execute Figure 2The method shown.
[0102] The present application also provides a computer-readable storage medium. The computer-readable storage medium may be any available medium that can be stored by a computing device or a data storage device such as a data center that contains one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk). The computer-readable storage medium includes instructions that instruct the computing device to execute Figure 2 The method shown.
[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.
Claims
1. A computing device management method, characterized in that: The computing device includes a display screen and a processing unit cluster, wherein the processing units in the processing unit cluster have computing power, and the computing power of the processing units is adjustable; the method includes: receiving a computing power display indication, wherein the computing power display indication is issued by a user when the computing device runs a service; In response to the computing power display indication, displaying computing power information of the processing unit cluster on the display screen, the computing power information of the processing unit cluster including first computing power information of a first processing unit in the processing unit cluster, the first processing unit being used to provide part or all of the computing power required for the computing device to run the service; When a computing power adjustment instruction issued by the user is received, the second computing power information of the first processing unit is displayed on the display screen, wherein the computing power adjustment instruction is used to instruct the computing device to adjust the computing power of the first processing unit, and the second computing power information is the computing power information of the first processing unit after the computing power adjustment.
2. The method according to claim 1, characterized in that: The computing power information of the processing unit cluster also includes computing power information of a second processing unit in the processing unit cluster, wherein the first processing unit is used to provide a part of the computing power required for the computing device to run the service, and the second processing unit is used to provide another part of the computing power required for the computing device to run the service; The method further comprises: receiving a policy instruction issued by the user for the first processing unit and the second processing unit; Based on the policy indication, a service scheduling policy between the first processing unit and the second processing unit is set.
3. The method according to claim 1 or 2, characterized in that: The method further comprises: receiving an alarm condition issued by the user for the first processing unit; When the first computing power information meets the alarm condition, an alarm is issued.
4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: receiving a frequency restriction request from the user for the first processing unit; The operating frequency of the first processing unit is controlled to be greater than or less than the limiting frequency.
5. The method according to any one of claims 1 to 4, characterized in that The first computing power information includes at least one of the computing power, computing power usage, computing power margin, and energy efficiency ratio of the first processing unit.
6. The method according to any one of claims 1 to 5, characterized in that The computing power adjustment instruction is used to instruct the computing device to execute any one of: changing the operating frequency of the first processing unit, shutting down the first processing unit, and turning on the first processing unit.
7. The method according to any one of claims 1 to 6, characterized in that The first processing unit is any one of the following: A processor, a computing core in a multi-core processor, and a computing core group, wherein the computing core group includes at least two computing cores.
8. The method according to any one of claims 1 to 7, characterized in that The first computing power information includes computing power information of the first processing unit at multiple historical time points.
9. A computing device management apparatus, characterized in that: The computing device includes a display screen and a processing unit cluster, wherein the processing units in the processing unit cluster have computing power, and the computing power of the processing units is adjustable, and the apparatus includes: A receiving unit, configured to receive a computing power display indication, wherein the computing power display indication is issued by a user when the computing device runs a service; a display unit, configured to display computing power information of a processing unit cluster on the display screen in response to the computing power display indication, wherein the computing power information of the processing unit cluster includes first computing power information of a first processing unit in the processing unit cluster, and the first processing unit is configured to provide part or all of the computing power required for the computing device to run the service; The display unit is also used to: when receiving a computing power adjustment instruction issued by a user, display the second computing power information of the first processing unit on the display screen, wherein the computing power adjustment instruction is used to instruct the computing device to adjust the computing power of the first processing unit, and the second computing power information is the computing power information of the first processing unit after the computing power adjustment.
10. The device according to claim 9, characterized in that The computing power information of the processing unit cluster also includes computing power information of a second processing unit in the processing unit cluster, wherein the first processing unit is used to provide a part of the computing power required for the computing device to run the service, and the second processing unit is used to provide another part of the computing power required for the computing device to run the service; the device also includes: a setting unit; wherein, The receiving unit is used to: receive a policy instruction issued by the user for the first processing unit and the second processing unit; The setting unit is used to set a service scheduling policy between the first processing unit and the second processing unit based on the policy indication.
11. The device according to claim 9 or 10, characterized in that The device further comprises: an alarm unit; wherein, The receiving unit is used to: receive an alarm condition issued by the user for the first processing unit; The alarm unit is used to issue an alarm when the first computing power information meets the alarm condition.
12. The device according to any one of claims 9 to 11, characterized in that The device further comprises: a control unit; wherein, The receiving unit is used to: receive a restricted frequency for the first processing unit issued by the user; The control unit is used to control the operating frequency of the first processing unit to be greater than or less than the limiting frequency.
13. A computing device, characterized in that: include: A memory for storing executable programs; A processor, configured to execute the method according to any one of claims 1 to 8 by running the executable program.
14. A computer-readable storage medium, characterized in that: The method comprises computer program instructions, and when the computer program instructions are executed by a computing device, the computing device performs the method according to any one of claims 1 to 8.
15. A computer program product comprising instructions, characterized in that When the instructions are executed by a computing device, the computing device is caused to perform the method according to any one of claims 1 to 8.