Low-power-consumption management method of cloud computer equipment and related equipment

Abstract

The invention provides a low-power-consumption management method for cloud computer equipment. The method comprises the following steps: monitoring a user operation behavior and a system running state in real time through a client; generating a hierarchical dormancy decision based on the user operation behavior and the system operation state; when the hierarchical dormancy decision meets a preset first dormancy threshold value, the system is controlled to enter a first dormancy state, and the first dormancy state comprises closing non-core peripherals and keeping internal memory power supply and network connection; in the first dormancy state, when the hierarchical dormancy decision meets a preset second dormancy threshold value, the system is controlled to enter a second dormancy state, and the second dormancy state comprises the steps that all the application sub-processes and the system service process store data and exit, all the peripheral modules enter the dormancy state, and main power supply of all the peripheral modules is powered off. The problems that an existing low-power-consumption management method for the cloud computer equipment is single in dormancy condition, cannot adapt to complex use scenes and is prone to causing task interruption or mistaken dormancy are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud computers, and particularly relates to a low-power consumption management method of a cloud computer device and related equipment. BACKGROUND

[0002] With the popularization of cloud computing technology, cloud computers, as a kind of virtual desktop service based on the cloud, are widely used in enterprise office, remote collaboration and other fields. At present, the cloud computer device triggers hibernation when the user is not operating, and only supports a fixed time threshold, which cannot adapt to complex use scenarios (such as background task running, network transmission, etc.), and is prone to cause task interruption or false hibernation. Therefore, the low-power consumption management method of the existing cloud computer device has the problems of single hibernation condition, inability to adapt to complex use scenarios, and easy task interruption or false hibernation. SUMMARY

[0003] The embodiment of the present application provides a low-power consumption management method of a cloud computer device, which aims to solve the problems of single hibernation condition, inability to adapt to complex use scenarios, and easy task interruption or false hibernation of the low-power consumption management method of the existing cloud computer device. By monitoring the user operation behavior and the system running state in real time through the client, a hierarchical hibernation decision is generated according to the user operation behavior and the system running state, and when the hierarchical hibernation decision meets a preset first hibernation threshold, the system is controlled to enter a first hibernation state, and when the hierarchical hibernation decision meets a preset second hibernation threshold in the first hibernation state, the system is controlled to enter a second hibernation state, thereby solving the problems of single hibernation condition, inability to adapt to complex use scenarios, and easy task interruption or false hibernation of the low-power consumption management method of the existing cloud computer device.

[0004] In a first aspect, the embodiment of the present application provides a low-power consumption management method of a cloud computer device, which comprises the following steps: monitoring the user operation behavior and the system running state in real time through the client; generating a hierarchical hibernation decision based on the user operation behavior and the system running state; controlling the system to enter a first hibernation state when the hierarchical hibernation decision meets a preset first hibernation threshold, wherein the first hibernation state comprises turning off non-core peripherals, keeping memory power supply and network connection; controlling the system to enter a second hibernation state when the hierarchical hibernation decision meets a preset second hibernation threshold in the first hibernation state, wherein the second hibernation state comprises saving data and exiting of each application sub-process and system service process, entering a hibernation state of each peripheral module, and power-off of the main power supply of each peripheral module.

[0005] Optionally, the controlling the system to enter the first hibernation state comprises: issuing a first hibernation signal to the system through the client, the system shutting down power supply of non-core peripheral modules; maintaining power supply of the system memory and network connection.

[0006] Optionally, the hierarchical hibernation decision is generated based on the user operation behavior and the system running state, and includes: The hierarchical hibernation decision is generated by inputting the user operation behavior and the system running state into a trained hibernation decision model, the trained hibernation decision model being obtained by training a pre-trained hibernation decision model based on a training data set, the training data set including sample user operation behavior, no-operation behavior duration annotation data corresponding to the sample user operation behavior, sample system running state, and application program running parameter annotation data and network load annotation data corresponding to the sample system running state.

[0007] Optionally, when the hierarchical hibernation decision meets a preset first hibernation threshold, after the system enters a first hibernation state, the method further includes: When the duration of the first hibernation state exceeds a preset time threshold, the system is controlled to enter a second hibernation state.

[0008] Optionally, the control of the system to enter the second hibernation state includes: issuing a second hibernation signal to the system through the client, the second hibernation signal being transmitted to each application sub-process and system service process, so that each process performs a data saving operation and safely exits; After each process exits, a completion second hibernation signal is fed back to the system kernel, and the kernel drives each peripheral module to enter a low-power state; The kernel sends the completion second hibernation signal to a micro control unit through a communication interface, and after the micro control unit receives the completion second hibernation signal, performs a hierarchical hardware power-off operation.

[0009] Optionally, after the micro control unit receives the completion second hibernation signal, the hierarchical hardware power-off operation includes: After the micro control unit receives the completion second hibernation signal, all input device power supplies are cut off, and each peripheral module main power supply is sequentially turned off according to a preset time sequence.

[0010] Optionally, after the micro control unit receives the completion second hibernation signal and performs the hierarchical hardware power-off operation, the method further includes: A low-power monitoring module of the micro control unit monitors a wake-up instruction, and when the wake-up instruction is received, the system power supply is restored.

[0011] In a second aspect, the embodiments of the present application further provide a low-power consumption management apparatus of a cloud computer device, which comprises: a monitoring module, configured to monitor user operation behavior and system running state in real time through a client; a generating module, configured to generate a hierarchical sleep decision based on the user operation behavior and the system running state; a first control module, configured to control the system to enter a first sleep state when the hierarchical sleep decision meets a preset first sleep threshold, the first sleep state comprising turning off non-core peripherals, keeping memory power supply and network connection; a second control module, configured to control the system to enter a second sleep state when the hierarchical sleep decision meets a preset second sleep threshold in the first sleep state, the second sleep state comprising saving data and exiting by each application sub-process and system service process, entering a sleep state by each peripheral module, and powering off the main power supply of each peripheral module.

[0012] In a third aspect, the embodiments of the present application provide an electronic device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the low-power consumption management method of the cloud computer device provided by the embodiments of the present application when executing the computer program. In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores a computer program, wherein the computer program is executable on a processor to implement the steps in the low-power consumption management method of the cloud computer device provided by the embodiments of the present application.

[0013] In this embodiment of the invention, user operation behavior and system operating status are monitored in real time by a client. Based on the user operation behavior and system operating status, a hierarchical hibernation decision is generated. When the hierarchical hibernation decision meets a preset first hibernation threshold, the control system enters a first hibernation state, which includes shutting down non-core peripherals and maintaining memory power supply and network connection. In the first hibernation state, when the hierarchical hibernation decision meets a preset second hibernation threshold, the control system enters a second hibernation state, which includes each application subprocess and system service process saving data and exiting, each peripheral module entering hibernation state, and the main power supply of each peripheral module being cut off. This invention, by monitoring user operation behavior and system operating status in real time by a client and generating hierarchical hibernation decisions based on these factors, and by controlling the system to enter a first hibernation state when the hierarchical hibernation decision meets a preset first hibernation threshold, and vice versa, solves the problem that existing low-power management methods for cloud computing devices have limited hibernation conditions, cannot adapt to complex usage scenarios, and are prone to task interruption or false hibernation. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart of a low-power management method for a cloud computer device provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a low-power management device for a cloud computer device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] like Figure 1 As shown, Figure 1is a flowchart of a low-power consumption management method of a cloud computer device provided by an embodiment of the present application, and the low-power consumption management method of the cloud computer device comprises the following steps: 101, monitoring user operation behavior and system running state in real time through a client.

[0018] In the embodiment of the present application, the low-power consumption management method of the cloud computer device can be applied to a cloud computer, and the cloud computer is a virtual desktop service based on cloud computing technology, which migrates the computing, storage and running environment of a traditional personal computer to a cloud server, and a user can access a complete cloud computer through a lightweight terminal device (such as a notebook, a tablet or a mobile phone) connected to a network.

[0019] The client can be understood as a user interface on the cloud computer device, which is used for interaction with the user and receiving the operation behavior of the user.

[0020] The user operation behavior includes operation behavior and non-operation behavior of the user on the cloud computer device, and the duration of the non-operation behavior.

[0021] The system running state can be understood as the working state of the system, which can be the running parameters of the current application program, network load conditions and the like when the user has no operation behavior.

[0022] It should be noted that the operation behavior of the user and the running state of the system can be monitored and collected in real time and uninterruptedly through the client program on the cloud computer device.

[0023] 102, generating a hierarchical sleep decision based on the user operation behavior and the system running state.

[0024] In the embodiment of the present application, the user operation behavior and the system running state can be input into a trained sleep decision model to generate a hierarchical sleep decision. The trained sleep decision model can be a sleep decision model constructed based on deep learning or machine learning, such as GRU (Gated Recurrent Unit), SVM (Support Vector Machine) and the like. The GRU (Gated Recurrent Unit) is a recurrent neural network variant proposed by Cho et al. in 2014, which simplifies the gating mechanism of LSTM, reduces the number of parameters and computational complexity while maintaining long-term memory capability. The core idea of the SVM (Support Vector Machine) is to maximize the distance between different class sample points, so as to ensure the accuracy and generalization ability of classification.

[0025] The trained sleep decision model can generate a hierarchical sleep decision according to the user operation behavior and the system running state.

[0026] The hierarchical sleep decision can be used to determine the cloud computer device to enter different low-power states according to the user operation behavior and the system running state.

[0027] 103、When the hierarchical sleep decision meets a preset first sleep threshold, the system is controlled to enter a first sleep state.

[0028] In the embodiments of the present application, the first sleep can be understood as a light sleep, and the light sleep can be understood as a sleep mode in which non-core peripheral modules are turned off, memory power supply is maintained, and network connection is maintained. The non-core peripheral modules can be understood as hardware devices that are not essential to the core functions of the system, such as display screens, audio speakers, etc. When the first sleep state is entered, the non-core peripheral modules are turned off to save energy.

[0029] The preset first sleep threshold can be a light sleep threshold set by the system in advance. Specifically, when no user operation behavior is detected for a preset duration, that is, when no user operation time is detected for a preset duration, it is determined that the hierarchical sleep decision meets the preset first sleep threshold. When the preset light sleep threshold is met, the system can be controlled to enter a light sleep state. The first sleep threshold can be dynamically set. Historical operation data of the user can be collected, and the entering time t0 of the cloud computer device of the user into the first sleep state and the wake-up time t1 corresponding to the next operation behavior after entering the first sleep state can be extracted from the historical operation data. If the average value of the difference t1-t0=△t1 between the wake-up time t1 and the entering time t0 is less than the first sleep threshold, the first sleep threshold is increased by a preset difference and an adjustment ratio. If the average value of the difference t1-t0=△t1 between the wake-up time t1 and the entering time t0 is greater than the first sleep threshold, the first sleep threshold is shortened by a preset difference and an adjustment ratio. In this way, the wake-up demand of the user for the light sleep state can be adapted.

[0030] The first sleep state includes turning off non-core peripherals, maintaining memory power supply, and maintaining network connection.

[0031] The first sleep state can be a light sleep state.

[0032] In a possible embodiment, for example, the preset light sleep threshold is that the duration of no user operation behavior of the user exceeds a preset no-operation behavior time threshold, and the network load state is greater than a preset network load threshold. When the hierarchical sleep decision meets the preset light sleep threshold, the system is controlled to enter a light sleep state. The preset no-operation behavior time threshold can be a user no-operation behavior duration threshold set by the system in advance. The preset network load can be a network load threshold set by the system in advance.

[0033] 104、in the first sleep state, when the hierarchical sleep decision satisfies the preset second sleep threshold, the control system enters the second sleep state.

[0034] In the embodiment of the present application, the second sleep state includes that each application sub-process and system service process saves data and exits, each peripheral module enters the sleep state, and the main power supply of each peripheral module is powered off.

[0035] The second sleep can be deep sleep, which can be understood as the system stopping running.

[0036] The preset second sleep threshold can be a deep sleep threshold preset by the system, and when the preset deep sleep threshold is satisfied in the first sleep state, the system can be controlled to enter the deep sleep state. The second sleep threshold can be dynamically set, historical operation data of the user can be collected, the entering time t2 of the cloud computer device of the user into the second sleep state and the next operation behavior corresponding to the wake-up time t3 after entering the second sleep state can be extracted according to the historical operation data, if the difference t3-t2=△t2 between the wake-up time t3 and the entering time t2 is less than the average value of the second sleep threshold, the second sleep threshold is adjusted to increase by a preset difference and an adjustment ratio, if the difference t3-t2=△t2 between the wake-up time t3 and the entering time t2 is greater than the average value of the second sleep threshold, the second sleep threshold is adjusted to shorten by a preset difference and an adjustment ratio. In this way, the wake-up demand of the user for the deep sleep state can be adapted.

[0037] The second sleep state can be a deep sleep state.

[0038] In a possible embodiment, for example, the preset deep sleep threshold is that the duration of no operation behavior of the user is greater than a preset no operation behavior time threshold, no background program is running, and the network load state is lower than a preset network load threshold, in the light sleep state, when the hierarchical sleep decision satisfies the preset deep sleep threshold, the control system enters the deep sleep mode, in the deep sleep mode, the application sub-process and the system service process, and the main power supply of the peripheral module are closed. The preset no operation behavior time threshold can be a user no operation behavior duration threshold preset by the system. The preset network load can be a network load threshold preset by the system.

[0039] In the embodiment of the present application, the user operation behavior and the system running state are monitored in real time by the client; the hierarchical sleep decision is generated based on the user operation behavior and the system running state; when the hierarchical sleep decision meets the preset first sleep threshold, the system is controlled to enter the first sleep state, and the first sleep state includes turning off the non-core peripherals, keeping the memory power supply and network connection; in the first sleep state, when the hierarchical sleep decision meets the preset second sleep threshold, the system is controlled to enter the second sleep state, and the second sleep state includes saving data and exiting by each application sub-process and system service process, entering the sleep state by each peripheral module, and powering off the main power supply of each peripheral module. The present application monitors the user operation behavior and the system running state in real time by the client, generates the hierarchical sleep decision according to the user operation behavior and the system running state, controls the system to enter the first sleep state when the hierarchical sleep decision meets the preset first sleep threshold, controls the system to enter the second sleep state when the hierarchical sleep decision meets the preset second sleep threshold in the first sleep state, and solves the problem that the existing low-power management method of the cloud computer device has a single sleep condition and cannot adapt to complex use scenarios, which easily leads to task interruption or false sleep.

[0040] It can be understood that in the specific embodiments of the present application, operation behavior data, running data, sleep data, load data, user data and other related data are involved, and when the embodiments in the present application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data, as well as the training, deployment and calling of algorithm models, need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0041] Optionally, in the step of controlling the system to enter the first sleep state, the client can issue a first sleep signal to the system, and the system turns off the power supply of the non-core peripheral module; the memory power supply and network connection of the system are kept.

[0042] In the embodiment of the present application, the first sleep signal described above can be a light sleep signal, which is used to instruct the system to enter a light sleep state.

[0043] The non-core peripheral module described above can be understood as a hardware device that is not essential to the core function of the system, such as a display screen, an audio speaker, etc.

[0044] It can be understood that the client issues a first sleep signal to the system, and the system turns off the power supply of the non-core peripheral module after receiving the first sleep signal, and enters the first sleep state, and the memory power supply and network connection of the system are kept when entering the first sleep state, so as to quickly resume operation when needed.

[0045] Optionally, in the step of generating the hierarchical hibernation decision based on the user operation behavior and the system running state, the hierarchical hibernation decision can be generated by inputting the user operation behavior and the system running state into the trained hibernation decision model.

[0046] In the embodiment of the present application, the user operation behavior includes the operation behavior and the non-operation behavior of the user on the cloud computer device, and the duration of the non-operation behavior.

[0047] The system running state can be understood as the working state of the system, which can be the running parameters of the current application program, network load conditions, etc. during the non-operation behavior of the user.

[0048] The trained hibernation decision model is obtained by training the pre-trained hibernation decision model based on the training data set.

[0049] The training data set includes sample user operation behavior, non-operation behavior duration annotation data corresponding to the sample user operation behavior, sample system running state, application program running parameter annotation data corresponding to the sample system running state, and network load annotation data. The annotation data is a process of converting original data into a form understandable by a machine learning model, which enables the machine to learn and perform classification, detection, etc. by adding semantic labels or structured information to the data.

[0050] The pre-trained hibernation decision model can be a hibernation decision model constructed based on deep learning or machine learning, such as GRU (‌Gated Recurrent Unit), SVM (Support Vector Machine), etc.

[0051] The training can be supervised training, which uses a set of data with known labels to train the model, and optimizes the model parameters to enable the model to predict the labels of new data or make decisions based on the characteristics of existing data. During the training process, the parameters of the model can be adjusted using a minimum loss function to minimize the difference between the output labels of the model and the input data. The loss function is used to measure the difference between the predicted results of the model and the true results, and the purpose is to minimize the loss function value by adjusting the model parameters, thereby improving the prediction accuracy. The loss function can be a mean square error loss function, a cross-entropy loss function, etc. The parameter adjustment refers to the process of adjusting the weights and biases of the model to optimize the performance of the model.

[0052] It should be noted that, in the training process, the minimum loss function can be taken as the optimization objective, the model parameters are adjusted through the back propagation algorithm, the adjustment process of the model parameters is iterated, until the error loss is less than a preset value, or the iteration number reaches a preset number, the training process is ended, and the trained hibernation decision model is obtained. The back propagation algorithm is a supervised learning algorithm that updates the weights by calculating the gradient of the loss function to minimize the error between the predicted output and the true value.

[0053] The trained hibernation decision model can generate a hierarchical hibernation decision according to the user operation behavior and the system running state.

[0054] The hierarchical hibernation decision is used to determine the low-power state of the cloud computer device entering different hibernation levels.

[0055] In a possible embodiment, for example, the duration of the user inactivity behavior exceeds a preset inactivity time threshold, and the system running state shows that the application running parameter and the network load are lower than a preset threshold, the trained hibernation decision model generates a hibernation decision for the system to enter a deep hibernation state.

[0056] In another possible embodiment, for example, the duration of the user inactivity behavior exceeds a preset inactivity time threshold, and the system running state shows that the application running parameter and the network load are greater than or equal to a preset threshold, the trained hibernation decision model generates a hibernation decision for the system to enter a light hibernation state; in the light hibernation state, when the system running state shows that the application running parameter and the network load are lower than a preset threshold, the trained hibernation decision model generates a hibernation decision for the system to enter a deep hibernation state.

[0057] Optionally, after the step of controlling the system to enter the first hibernation state when the hierarchical hibernation decision meets a preset first hibernation threshold, the system can be controlled to enter a second hibernation state when the duration of the first hibernation state exceeds a preset time threshold.

[0058] In the embodiment of the present application, the preset time threshold can be a duration threshold of the first hibernation state preset by the system.

[0059] The second hibernation state can be a deep hibernation state, and the second hibernation state includes saving data and exiting of each application sub-process and system service process, entering a hibernation state of each peripheral module, and power-off of the main power supply of each peripheral module.

[0060] It can be understood that when the duration of the first hibernation state exceeds the preset time threshold, the system is controlled to enter the second hibernation state.

[0061] Optionally, in the step of the control system entering the second sleep state, a second sleep signal can be sent to the system by the client, and the second sleep signal is transmitted to each application sub-process and system service process, so that each process performs a data saving operation and safely exits; after each process exits, a second sleep signal completion feedback is fed back to the system kernel, the kernel drives each peripheral module to enter a low-power state; the kernel sends a second sleep signal completion to the micro control unit through the communication interface, and after the micro control unit receives the second sleep signal completion, a hierarchical hardware power-off operation is performed.

[0062] In the embodiment of the application, the second sleep signal can be a deep sleep signal, which is used to indicate that the system enters a deep sleep state.

[0063] The application sub-process can be understood as a process for executing a specific task. It can be understood that an application program can have multiple sub-processes, and multiple sub-processes can run in parallel to improve the execution efficiency of the program.

[0064] The system service process can be understood as a long-term task running in the background, providing system-level functions such as network management, print service, etc.

[0065] The peripheral module includes a WIFI module, a Bluetooth module, an Ethernet module, a SPEAK module, a MIC module, etc.

[0066] The hardware includes a processor, a memory, a peripheral module, etc.

[0067] The low-power state can be a sleep state.

[0068] Specifically, a second sleep signal is sent to the system by the client, and the second sleep signal is transmitted to each application sub-process and system service process, so that each process performs a data saving operation and safely exits. After each process safely exits, the system kernel feeds back a second sleep signal completion, so that each peripheral module enters a low-power state, the kernel sends a second sleep signal completion to the micro control unit through the communication interface, and after the micro control unit receives it, a hierarchical hardware power-off operation is performed, thereby realizing low-power management of the entire system.

[0069] Optionally, in the step of the micro control unit receiving the second sleep signal completion and performing the hierarchical hardware power-off operation, after the micro control unit receives the second sleep signal completion, it cuts off the power supply of all input devices, and according to a preset time sequence, sequentially turns off the main power supply of each peripheral module.

[0070] In the embodiment of the application, the preset time sequence can be a sequence and time interval for turning off the main power supply of each peripheral module, which is set by the system in advance. For example, the power supply of the display, hard disk and other peripheral modules is turned off first, and finally the power supply of the host computer is turned off.

[0071] The input device is a device for inputting information to the system, such as a keyboard, a mouse, a touch screen, etc.

[0072] The peripheral module includes a WIFI module, a Bluetooth module, an Ethernet module, a SPEAK module, a MIC module, etc.

[0073] In a possible embodiment, after receiving the second sleep completion signal, the micro control unit first cuts off the power supply of all input devices, then closes the main power supply of each peripheral module according to the preset timing, and when all peripheral modules have completed the power-off operation, the micro control unit cuts off its own power supply, so that the whole system enters the second sleep state.

[0074] Optionally, after the micro control unit receives the second sleep completion signal and performs the step of hierarchical hardware power-off operation, the low-power monitoring module of the micro control unit can monitor the wake-up instruction, and when the wake-up instruction is received, the system power supply is restored.

[0075] In the embodiment of the application, the low-power monitoring module can be composed of a low-power processor or chip, which is used to monitor the wake-up instruction after the micro control unit enters the second sleep state.

[0076] The wake-up instruction can be understood as an instruction for waking up the system.

[0077] It should be noted that the low-power monitoring module can maintain a certain working state when the micro control unit enters the second sleep state, and monitor the wake-up instruction. When the low-power monitoring module receives the wake-up instruction, an interrupt signal is sent to the micro control unit, so that the micro control unit exits the second sleep state and restores the system power supply.

[0078] As shown in Figure 2 The low-power management device of the cloud computer device provided by the embodiment of the application comprises: A monitoring module 201 is configured to monitor user operation behavior and system running state in real time through a client; A generating module 202 is configured to generate a hierarchical sleep decision based on the user operation behavior and the system running state; A first control module 203 is configured to control the system to enter a first sleep state when the hierarchical sleep decision meets a preset first sleep threshold, and the first sleep state comprises closing non-core peripherals, maintaining memory power supply and network connection; The second control module 204 is configured to control the system to enter a second sleep state when the hierarchical sleep decision meets a preset second sleep threshold in the first sleep state, and the second sleep state comprises that each application sub-process and system service process saves data and exits, each peripheral module enters a sleep state, and main power supply of each peripheral module is powered off.

[0079] Optionally, the first control module 203 is further configured to send a first sleep signal to the system through the client, and the system powers off non-core peripheral modules; and the system memory power supply and network connection are maintained.

[0080] Optionally, the generation module 202 is further configured to input the user operation behavior and the system running state into a trained sleep decision model to generate a hierarchical sleep decision, the trained sleep decision model is obtained by training a pre-trained sleep decision model by using a training data set, and the training data set comprises sample user operation behaviors, no-operation behavior duration label data corresponding to the sample user operation behaviors, sample system running states, application program running parameter label data corresponding to the sample system running states, and network load label data.

[0081] Optionally, the device is further configured to control the system to enter a second sleep state when the duration of the first sleep state exceeds a preset time threshold.

[0082] Optionally, the second control module 204 is further configured to send a second sleep signal to the system through the client, and the second sleep signal is transmitted to each application sub-process and system service process, so that each process performs a data saving operation and safely exits; after each process exits, the system kernel is fed back a complete second sleep signal, the kernel drives each peripheral module to enter a low-power state; the kernel sends the complete second sleep signal to a micro control unit through a communication interface, and the micro control unit receives the complete second sleep signal and performs a hierarchical hardware power-off operation.

[0083] Optionally, the second control module 204 is further configured to, after the micro control unit receives the complete second sleep signal, cut off power supply of all input devices, and sequentially turn off main power supply of each peripheral module according to a preset time sequence.

[0084] Optionally, the device is further configured to monitor a wake-up instruction through a low-power monitoring module of the micro control unit, and restore the system power supply when the wake-up instruction is received.

[0085] As Figure 3As shown, the embodiment of the present application also provides an electronic device, comprising a processor, wherein the processor can execute the low-power consumption management method of any one of the cloud computer devices.

[0086] Specifically, the electronic device comprises a processor 301 and a memory 302, and a computer program for executing the low-power consumption management method of the cloud computer device stored in the memory 302 and capable of running on the processor 301, wherein: The processor 301 runs the computer program of the low-power consumption management method of the cloud computer device stored in the memory 302, and executes the following steps: Real-time monitoring of user operation behavior and system running state through the client; Generating a hierarchical sleep decision based on the user operation behavior and the system running state; When the hierarchical sleep decision meets a preset first sleep threshold, controlling the system to enter a first sleep state, wherein the first sleep state comprises turning off non-core peripherals, keeping memory power supply and network connection; In the first sleep state, when the hierarchical sleep decision meets a preset second sleep threshold, controlling the system to enter a second sleep state, wherein the second sleep state comprises saving data and exiting of each application sub-process and system service process, entering a sleep state of each peripheral module, and power-off of main power supply of each peripheral module.

[0087] Optionally, the processor 301 executes the control of the system to enter the first sleep state, comprising: Issuing a first sleep signal to the system through the client, and the system turning off the power supply of the non-core peripheral module; Keeping the memory power supply and network connection of the system.

[0088] Optionally, the processor 301 executes the generation of the hierarchical sleep decision based on the user operation behavior and the system running state, comprising: Inputting the user operation behavior and the system running state into a trained sleep decision model to generate a hierarchical sleep decision, wherein the trained sleep decision model is obtained by training a pre-trained sleep decision model with a training data set, and the training data set comprises sample user operation behavior, corresponding no-operation behavior duration annotation data of the sample user operation behavior, sample system running state, and application program running parameter annotation data and network load annotation data corresponding to the sample system running state.

[0089] Optionally, after the hierarchical sleep decision meets the preset first sleep threshold and the system enters the first sleep state, the processor 301 executes the method further comprising: When the duration of the first sleep state exceeds a preset time threshold, the system is controlled to enter a second sleep state.

[0090] Optionally, the control of the system to enter the second sleep state performed by the processor 301 comprises: A second sleep signal is issued to the system through the client, and the second sleep signal is transmitted to each application sub-process and system service process, so that each process performs a data saving operation and safely exits; After each process exits, a completion second sleep signal is fed back to the system kernel, and the kernel drives each peripheral module to enter a low-power state; The kernel sends the completion second sleep signal to the micro control unit through the communication interface, and after the micro control unit receives the completion second sleep signal, a hierarchical hardware power-off operation is performed.

[0091] Optionally, after the micro control unit receives the completion second sleep signal, the hierarchical hardware power-off operation performed by the processor 301 comprises: After the micro control unit receives the completion second sleep signal, the power supply of all input devices is cut off, and each peripheral module main power supply is sequentially turned off according to a preset time sequence.

[0092] Optionally, after the micro control unit receives the completion second sleep signal and performs the hierarchical hardware power-off operation, the method performed by the processor 301 further comprises: A low-power monitoring module of the micro control unit monitors a wake-up instruction, and when the wake-up instruction is received, the system power supply is restored.

[0093] The embodiment of the application also provides a computer readable storage medium, and the computer readable storage medium stores a computer program, the computer program is executed by a processor to realize each process of the low-power management method of the cloud computer device provided by the embodiment of the application, and the same technical effects can be achieved. To avoid repetition, it will not be repeated here.

[0094] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by a computer program instructing related hardware, and the program can be stored in a computer readable storage medium. The program can include the processes of the above-mentioned embodiment when executed. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0095] The above merely provides the preferred embodiment of the application, and cannot allude the protection scope of the application, therefore any equivalent changes made according to the claims of the application shall be within the scope of the application.

Claims

1. A low power management method of a cloud computer device, the method comprising: The method comprises: ​ Monitoring user operation behavior and system running state in real time through a client; Generating a hierarchical hibernation decision based on the user operation behavior and the system running state; When the hierarchical hibernation decision meets a preset first hibernation threshold, controlling the system to enter a first hibernation state, the first hibernation state comprising turning off non-core peripherals, keeping memory power supply and network connection; In the first hibernation state, when the hierarchical hibernation decision meets a preset second hibernation threshold, controlling the system to enter a second hibernation state, the second hibernation state comprising saving data and exiting by each application sub-process and system service process, each peripheral module entering a hibernation state, and main power supply of each peripheral module being powered off.

2. The method of claim 1, wherein, The control system enters the first hibernation state, comprising: The client sends a first hibernation signal to the system, and the system turns off power supply of non-core peripheral modules; The system memory power supply and network connection are kept.

3. The method of claim 1, wherein, The hierarchical hibernation decision is generated based on the user operation behavior and the system running state, comprising: The user operation behavior and the system running state are input into a trained hibernation decision model to generate a hierarchical hibernation decision, the trained hibernation decision model being obtained by training a pre-trained hibernation decision model using a training data set, the training data set comprising sample user operation behavior, no-operation behavior duration annotation data corresponding to the sample user operation behavior, sample system running state, and application program running parameter annotation data and network load annotation data corresponding to the sample system running state.

4. The method of claim 1, wherein, After the control system enters the first hibernation state when the hierarchical hibernation decision meets the preset first hibernation threshold, the method further comprises: When the duration of the first hibernation state exceeds a preset time threshold, controlling the system to enter the second hibernation state.

5. The method of claim 1, wherein, The control system enters the second hibernation state, comprising: The client sends a second hibernation signal to the system, and the second hibernation signal is transmitted to each application sub-process and system service process, so that each process performs a data saving operation and safely exits; After each process exits, the system kernel is fed back a complete second hibernation signal, and the kernel drives each peripheral module to enter a low-power state; The kernel sends the complete second hibernation signal to a micro control unit through a communication interface, and after the micro control unit receives the complete second hibernation signal, performs a hierarchical hardware power-off operation.

6. The method of claim 5, wherein, After the micro control unit receives the complete second hibernation signal, the hierarchical hardware power-off operation comprises: After the micro control unit receives the complete second hibernation signal, the micro control unit cuts off power supply of all input devices and sequentially turns off main power supply of each peripheral module according to a preset time sequence.

7. The method of claim 6, wherein, After the micro control unit receives the complete second hibernation signal and performs the hierarchical hardware power-off operation, the method further comprises: A low-power monitoring module of the micro control unit monitors a wake-up instruction, and when the wake-up instruction is received, the system power supply is restored.

8. A low-power management device for a cloud computing device, characterized in that, The low-power management device of the cloud computer device comprises: A monitoring module is configured to monitor user operation behavior and system running state in real time through a client; A generating module is configured to generate a hierarchical sleep decision based on the user operation behavior and the system running state; A first control module is configured to control the system to enter a first sleep state when the hierarchical sleep decision meets a preset first sleep threshold, the first sleep state including turning off non-core peripherals, keeping memory power supply and network connection; A second control module is configured to control the system to enter a second sleep state when the hierarchical sleep decision meets a preset second sleep threshold in the first sleep state, the second sleep state including saving data and exiting by each application sub-process and system service process, entering a sleep state by each peripheral module, and powering off the main power supply of each peripheral module.

9. An electronic device, comprising: The cloud computer device comprises: A memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps in the low-power consumption management method of the cloud computer device according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is stored in the computer readable storage medium, and the computer program is executed by the processor to implement the steps in the low-power consumption management method of the cloud computer device according to any one of claims 1 to 7.

Images