Power consumption optimization method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of computers, in particular to a power consumption optimization method and device, electronic equipment and a storage medium. According to the method, the real-time operation data and the network state data of the middle-end device, the source-end device and the edge device in the collaborative network are acquired, so that the overall resource and network conditions can be comprehensively grasped, the limitation of single device resource management is broken through, and the cloud edge-end collaborative advantage is fully utilized. Based on the determined target task execution device, the optimal allocation of resources can be realized, and the resource utilization rate is improved. And meanwhile, the execution condition is monitored and the power consumption optimization strategy is adjusted, so that the energy consumption can be effectively reduced. In addition, the dynamic decision under the collaborative network enables task execution to be more efficient, and delay is reduced. Besides, according to the scheme, the strategy can be flexibly adjusted according to real-time data, and the dynamic adaptability is enhanced, so that the problems of low resource utilization rate, high energy consumption, high delay and insufficient dynamic adaptability in an existing power consumption management method are solved, and the overall efficiency of power consumption management of the intelligent equipment is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a power consumption optimization method, device, electronic device and storage medium. Background Art

[0002] In the application scenarios of smart devices (such as smart glasses, cloud phones, drones, and IoT devices), optimizing the power consumption of computing tasks has become a key technical challenge that the industry urgently needs to overcome. Current mainstream power management methods are mostly limited to the resource management model of a single device, which makes it difficult to fully realize the synergy potential of cloud computing and edge computing.

[0003] Specifically, it manifests as: low resource utilization and inability to dynamically integrate cloud, edge, and end resources, resulting in an unbalanced distribution of computing tasks and the coexistence of overload and idle equipment; high energy consumption, and under high load conditions, complex tasks cannot be efficiently offloaded to the cloud or edge nodes, exacerbating the power consumption of end devices; high task processing delays, especially for tasks with low latency requirements, which fail to fully utilize the computing advantages of edge nodes, affecting user experience; insufficient dynamic adaptability, and traditional methods lack real-time perception and adjustment mechanisms for device load, network conditions, and user behavior, making it difficult to meet complex and changing application requirements.

[0004] In addition, existing technologies mostly focus on power consumption optimization at a single level, such as the cloud or the end side, making it difficult to achieve a balance between energy consumption and performance at the system level. Summary of the Invention

[0005] In view of this, the embodiments of the present invention provide a power consumption optimization method, device, electronic device and storage medium to solve the problems that the current smart device power consumption management method is limited to the resource management of a single device itself, it is difficult to utilize the advantages of cloud-edge collaboration, and there are problems such as low resource utilization, high energy consumption, high latency and insufficient dynamic adaptability.

[0006] In a first aspect, an embodiment of the present invention provides a power consumption optimization method, the method comprising:

[0007] Acquire current network status data of the collaborative network and real-time operation data of each end in the collaborative network, wherein the collaborative network includes the end device, the source end device, and the edge device;

[0008] Determining an execution device corresponding to a target task currently to be distributed based on the real-time operation data and the network status data;

[0009] Sending the target task to the execution device, and monitoring the execution status of the target task by the execution device;

[0010] A power consumption optimization strategy of the execution device is adjusted based on the execution situation.

[0011] Furthermore, the obtaining of the current network status data of the collaborative network and the real-time operation data of each terminal in the collaborative network includes:

[0012] Collecting network load data and user behavior data of the terminal devices in the collaborative network, and using the network load data and the user behavior data as real-time operation data of the terminal devices;

[0013] Collecting real-time operating data of edge devices and cloud devices in the collaborative network;

[0014] The current network status data of the collaborative network is determined through the network data fed back by the edge device and the cloud device.

[0015] Furthermore, determining the execution device corresponding to the target task to be distributed based on the real-time operation data and the network status data includes:

[0016] Obtaining task requirement information of the target task;

[0017] Taking a device in the collaborative network that meets the task requirement information as a candidate execution device;

[0018] evaluating execution capability data of the candidate execution device according to the real-time operation data and the network status data;

[0019] Verify availability data of the candidate execution device having the highest execution capability data;

[0020] If the availability data meets the preset execution condition, the candidate execution device with the highest execution capability data is used as the execution data.

[0021] Furthermore, the evaluating the execution capability data of the candidate execution device according to the real-time operation data and the network status data includes:

[0022] Obtaining a pre-built execution capability evaluation model, wherein the execution capability evaluation model summarizes weights corresponding to multiple evaluation indicators;

[0023] Obtaining weights corresponding to the real-time operation data and the network status data from the execution capability evaluation model;

[0024] The execution capability data of the candidate device is calculated using the real-time operation data, the weight corresponding to the real-time operation data, the network status data, and the weight corresponding to the network status data.

[0025] Furthermore, monitoring the execution status of the target task by the execution device includes:

[0026] Obtain the execution node corresponding to the target task and the expected data corresponding to each execution node;

[0027] Monitoring actual data of the target task being executed by the execution device according to the execution node;

[0028] The actual data is compared with the expected data to obtain the execution status of the target task at each execution node.

[0029] Furthermore, the adjusting the power consumption optimization strategy of the execution device based on the execution situation includes:

[0030] Analyzing the execution status to obtain key factors causing high power consumption;

[0031] generating an initial optimization strategy based on the device hardware configuration of the execution device and the key factors;

[0032] Testing the initial optimization strategy in a simulation environment to evaluate improvements made to the initial optimization strategy;

[0033] If the improvement does not meet the preset requirements, the initial optimization strategy is revised and the revised initial optimization strategy is evaluated until the improvement meets the preset requirements. The final optimization strategy is then used as the power consumption optimization strategy.

[0034] Furthermore, generating an initial power consumption optimization strategy based on the device hardware configuration of the execution device and the key factors includes:

[0035] Analyze the correlation between the key factors and the hardware configuration of the device to determine the key cause of high power consumption;

[0036] Develop an optimization strategy framework for hardware limitations based on key reasons;

[0037] The optimization strategy framework is filled in according to the optimization direction carried by the optimization strategy framework using the hardware configuration of the device to obtain an initial power consumption optimization strategy.

[0038] In a second aspect, an embodiment of the present invention provides a power consumption optimization device, the device comprising:

[0039] an acquisition module, configured to acquire current network status data of a collaborative network and real-time operation data of each end in the collaborative network, wherein the collaborative network includes the end device, the source end device, and the edge device;

[0040] A determination module, configured to determine an execution device corresponding to a target task to be currently distributed based on the real-time operation data and the network status data;

[0041] a sending module, configured to send the target task to the execution device and monitor the execution status of the target task by the execution device;

[0042] An execution module is used to adjust the power consumption optimization strategy of the execution device based on the execution status.

[0043] In a third aspect, an embodiment of the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions being stored in the memory, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0044] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method of the first aspect or any corresponding embodiment thereof.

[0045] By obtaining real-time operating data and network status data of mid-end devices, source devices, and edge devices in the collaborative network, this application can fully grasp the overall resource and network status, break through the limitations of single device resource management, and make full use of the advantages of cloud-edge collaboration. Based on this, the target task execution device is determined to achieve optimal resource allocation and improve resource utilization. At the same time, monitoring the execution status and adjusting the power consumption optimization strategy can effectively reduce energy consumption. Moreover, dynamic decision-making under the collaborative network makes task execution more efficient and reduces delays. In addition, the solution can flexibly adjust the strategy according to real-time data and enhance dynamic adaptability, thereby solving the problems of low resource utilization, high energy consumption, high latency, and insufficient dynamic adaptability in existing power management methods, and improving the overall efficiency of smart device power management. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 is a flowchart of a power consumption optimization method according to some embodiments of the present invention;

[0048] Figure 2 is a schematic diagram of the structure of a collaborative network according to some embodiments of the present invention;

[0049] Figure 3is a structural block diagram of a power consumption optimization method and apparatus according to an embodiment of the present invention;

[0050] Figure 4 Schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0052] According to an embodiment of the present invention, a power consumption optimization method, apparatus, electronic device, and storage medium are provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0053] In this embodiment, a power consumption optimization method is provided. Figure 1 FIG. 1 is a flow chart of a method for optimizing power consumption according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0054] Step S101 : obtaining the current network status data of the collaborative network to obtain the real-time operation data of each end in the collaborative network, wherein the collaborative network includes an end device, a source end device, and an edge device.

[0055] The following specific implementation process is used to obtain the current network status data of the collaborative network and the real-time operation data of each end. First, at the end device level, the device's built-in sensors and monitoring programs are used to collect real-time operation data such as CPU usage, battery power, and memory usage. At the same time, network status data such as network bandwidth, latency, and packet loss rate are obtained through the network interface. The source device also uses the internal monitoring module to monitor its own task processing progress and resource usage in real time, and collect network connection parameters when communicating with other devices.

[0056] For edge devices, a dedicated data collection service is deployed to periodically collect operational information such as computing resource load and storage usage, as well as network status data when communicating with end devices, source devices, and the cloud. Each device then encrypts and packages the collected data using a pre-defined communication protocol. At regular intervals or when significant data changes occur, the data is sent to a designated data aggregation node. This node pre-processes the received data, including cleaning and formatting, and ultimately integrates it into a complete collaborative network status data and real-time operational data from each end, providing data support for subsequent collaborative decision-making and task scheduling.

[0057] Specifically, the current network status data of the collaborative network is obtained to obtain the real-time operation data of each end in the collaborative network, including: collecting network load data and user behavior data of the end devices in the collaborative network, and using the network load data and user behavior data as the real-time operation data of the end devices; collecting the real-time operation data of the edge devices and cloud devices in the collaborative network; and determining the current network status data of the collaborative network through the network data fed back by the edge devices and cloud devices.

[0058] The architecture of the collaborative network is as follows Figure 2 As shown, a dedicated monitoring program and sensors are embedded in the end device. The monitoring program collects real-time network load data, including actual network bandwidth usage, data transmission delay, and packet loss rate, to assess network communication pressure. Simultaneously, user actions on the device, such as tapping the screen, scrolling through pages, entering text, and switching applications, are recorded to generate user behavior data. This network load data and user behavior data are integrated and packaged to form real-time operational data for the end device, providing foundational information for subsequent system analysis and decision-making.

[0059] A data collection module is deployed on edge devices and cloud devices. This module regularly scans key operational indicators of these devices. For edge devices, this module primarily collects information such as CPU usage, memory usage, storage resource usage, and the queue length and progress of currently processing tasks. For cloud devices, in addition to the aforementioned indicators, it also collects information such as cluster resource allocation, task scheduling status, and data storage read / write performance. The collection module standardizes this data to ensure a consistent format, facilitating subsequent data transmission and analysis.

[0060] During operation, edge and cloud devices continuously monitor network parameters when communicating with other devices, such as link bandwidth, round-trip latency, network jitter, and connection stability between end devices, other edge devices, and the cloud. These parameters are fed back to the data processing center in real time. After receiving this feedback, the data processing center conducts a comprehensive analysis of the network data and, through statistical calculations and trend prediction, assesses the overall state of the collaborative network, including network traffic levels, data transmission efficiency, and potential network failure risks. Ultimately, this data is generated to represent the current state of the collaborative network.

[0061] Step S102: determining the execution device corresponding to the target task to be distributed based on the real-time operation data and the network status data.

[0062] In an embodiment of the present application, determining the execution device corresponding to the target task to be distributed based on real-time operation data and network status data includes the following steps A1-A5:

[0063] Step A1: Obtain task requirement information of the target task.

[0064] By parsing the metadata in the task code, reading the configuration file attached when the task is submitted, or receiving task parameters entered by the user, the specific requirements of the task on computing resources (such as the number of CPU cores, memory capacity requirements), storage resources (required storage space size), network conditions (bandwidth requirements, delay tolerance), processing time limit (task deadline), etc. are obtained, forming complete task requirement information and providing a basis for subsequent equipment screening.

[0065] In step A2, the devices in the collaborative network that meet the task requirement information are selected as candidate execution devices.

[0066] Based on the acquired task requirement information, the collaborative network's end devices, edge devices, and cloud devices are traversed. For each device, its resource configuration (such as CPU performance, memory size, storage capacity), network access capabilities (available bandwidth, latency range), and other information are compared with the task requirements. Only when a device's various resource and capability indicators meet the task requirements is it included in the candidate execution device list, thus selecting a set of devices that can theoretically perform the target task.

[0067] Step A3: Evaluate the execution capability data of the candidate execution devices based on the real-time operation data and the network status data.

[0068] Specifically, based on the real-time operation data and the network status data, the execution capability data of the candidate execution device is evaluated, including: obtaining a pre-built execution capability evaluation model, wherein the execution capability evaluation model summarizes the weights corresponding to multiple evaluation indicators; obtaining the real-time operation data and the weights corresponding to the network status data from the execution capability evaluation model; and calculating the execution capability data of the candidate device using the real-time operation data, the weights corresponding to the real-time operation data, the network status data, and the weights corresponding to the network status data.

[0069] First, the built execution capability assessment model needs to be loaded from the storage system or database. This model is typically a pre-trained machine learning model or an evaluation framework designed using an expert system. It contains multiple evaluation metrics (such as real-time operation data and network status data) and their corresponding weights. These weights reflect the importance of each metric in the evaluation process.

[0070] After loading the execution capability assessment model, the weights corresponding to the real-time operational data and network status data need to be extracted from the model. These weights are usually determined during the model construction phase through data analysis, expert experience, or machine learning algorithms and are used for weighted processing in subsequent calculations.

[0071] After obtaining real-time operational data, network status data, and their corresponding weights, these data and weights need to be combined and summed or subjected to other mathematical operations. Specifically, the real-time operational data is multiplied by its weight, the network status data is multiplied by its weight, and the two products are then added together to obtain the candidate device's comprehensive execution capability data.

[0072] Step A4: Verify the availability data of the candidate execution device with the highest execution capability data.

[0073] After obtaining the execution capability data for each candidate execution device, the device with the highest execution capability data is selected for availability verification. By sending test commands or heartbeat packets to the device, the system verifies that it is operating normally and can respond to external requests. The system also queries the device's current task scheduling to determine if there are any task conflicts or resource occupation. Furthermore, the system checks the stability of the device's network connection with other related devices in the collaborative network. This verification information is collected and organized to form availability data for the device, determining whether it is ready to immediately execute the target task.

[0074] Step A5: If the availability data meets the preset execution condition, the candidate execution device with the highest execution capability data is used as the execution data.

[0075] The obtained availability data is compared with the pre-set execution conditions. The pre-set execution conditions include device online status, no task conflicts, normal network connection, etc. If the availability data of the device fully meets the pre-set conditions, indicating that it can reliably execute the target task, the device is determined as the final execution device, and its related information (such as device identification, execution capability data, resource allocation, etc.) is output as execution data for subsequent task allocation and execution process; if it is not met, the candidate execution device with the next highest execution capability data is re-evaluated until a device that meets the conditions is found.

[0076] Step S103: Send the target task to the execution device, and monitor the execution status of the target task by the execution device.

[0077] In an embodiment of the present application, monitoring the execution status of the target task executed by the execution device includes: obtaining the execution node corresponding to the target task and the expected data corresponding to each execution node; monitoring the actual data of the execution device executing the target task according to the execution node; comparing the actual data with the expected data to obtain the execution status of the target task at each execution node.

[0078] After receiving the target task, the system first extracts the execution node information related to the task from the task configuration file, task scheduling policy library, or pre-set task execution rules. These execution nodes can be one or more end devices, edge devices, or cloud devices. At the same time, based on the task requirements and the performance parameters of each execution node, combined with historical task execution data or theoretical calculation models, expected data is set for each execution node. This includes expected processing time, resource usage (such as CPU utilization and memory usage), data output format and accuracy, and other indicators. This forms a complete set of expected task execution data, providing a reference standard for subsequent monitoring and evaluation.

[0079] After the target tasks are assigned to each execution node, a monitoring program deployed on the execution device or through the device's own management interface collects various data about the device during task execution in real time. The collected data focuses on different types of execution nodes: end devices primarily collect information about resource usage such as CPU, memory, and battery, as well as task execution progress; edge and cloud devices, in addition to resource usage data, also monitor information such as task scheduling queues and network data transmission volume. The monitoring program formats the collected data at regular intervals or when key events occur, and transmits it via the network to a data monitoring center to ensure dynamic tracking of task execution.

[0080] After receiving the actual data uploaded by each execution node, the data monitoring center compares and analyzes it against the corresponding expected data obtained in step one. Using methods such as difference calculation and percentage deviation calculation, the center quantifies the discrepancies between the actual and expected data. For example, it calculates the difference between actual and expected processing time, or the deviation rate between actual and expected resource usage. Combining this difference data with pre-set evaluation rules, the center conducts a comprehensive evaluation of the task execution status of each execution node, determining whether the task is meeting performance standards and whether there are any inefficiencies or resource waste. Ultimately, the center generates a detailed performance report for the target task at each execution node, providing a basis for task optimization and subsequent decision-making.

[0081] Step S104: adjusting the power consumption optimization strategy of the execution device based on the execution status.

[0082] In an embodiment of the present application, adjusting the power consumption optimization strategy of the execution device based on the execution status includes the following steps B1-B4:

[0083] Step B1: Analyze the execution status to obtain key factors leading to high power consumption.

[0084] First, we conduct an in-depth analysis of the execution status reports of the target tasks at each execution node, and identify the execution nodes with abnormal power consumption by combining the comparison results of real-time operation data with expected data. Through statistical analysis (such as correlation analysis and trend analysis), we locate the key influencing factors in high power consumption scenarios. For example, the end device does not offload high-load tasks to the edge node, causing the CPU to continue to run at full load, the network latency between the edge node and the cloud is too high, causing repeated data transmission, and the device power management strategy is not adapted to the task priority. At the same time, combined with user behavior data (such as real-time computing needs triggered by high-frequency operations) and network status data (such as task retries caused by bandwidth fluctuations), we further confirm the root cause of high power consumption and form a list of key factors.

[0085] Step B2: generating an initial optimization strategy based on the hardware configuration of the execution device and key factors.

[0086] Specifically, based on the device hardware configuration and key factors of the execution device, an initial power consumption optimization strategy is generated, including: analyzing the correlation between the key factors and the device hardware configuration to determine the key reasons for high power consumption; formulating an optimization strategy framework for hardware limitations based on the key reasons; using the device hardware configuration to fill the optimization strategy framework according to the optimization direction carried by the optimization strategy framework to obtain the initial power consumption optimization strategy.

[0087] Step B3: Test the initial optimization strategy in a simulation environment and evaluate the improvement of the initial optimization strategy.

[0088] Reproduce the execution scenario of the target task in a simulated environment, and configure the device parameters (such as CPU performance and network latency), task load (such as data input volume and processing complexity), and user behavior (such as operation frequency) in the collaborative network to be consistent with the actual scenario. Deploy the initial optimization strategy and run the task, collecting power consumption data (such as the battery consumption rate of the end device and the peak power consumption of the edge node) and task processing efficiency (such as latency and throughput) in real time. By comparing key indicators before and after optimization (such as the power consumption reduction ratio and the reduction rate of task completion time), evaluate the improvement effect of the initial strategy on the high power consumption problem and determine whether the preset optimization goals (such as a 20% reduction in power consumption and a latency of less than 50ms) have been achieved.

[0089] In step B4, if the improvement does not meet the preset requirements, the initial optimization strategy is revised and the revised initial optimization strategy is evaluated until the improvement meets the preset requirements. The final optimization strategy is then used as the power consumption optimization strategy.

[0090] If the improvement effect of the initial strategy in the simulation test does not meet the expectations, analyze the problems exposed during the strategy execution process (such as unreasonable task offloading threshold settings leading to edge node overload and insufficient DVFS adjustment range), and modify the strategy parameters based on hardware characteristics and task requirements (such as lowering the offloading threshold and refining the voltage and frequency adjustment gears). After the correction, conduct the simulation test again and repeat the evaluation process until the optimized power consumption indicators and task performance indicators meet the preset requirements. Finally, the strategy formed through multiple rounds of iterative optimization is used as the official power consumption optimization strategy and deployed to the execution devices in the collaborative network to achieve continuous optimization of high power consumption issues.

[0091] As an example, when a user uses smart glasses for real-time object recognition, the system first obtains real-time data from the collaborative network: it detects that the battery level of the smart glasses (end device) is only 20% and the CPU utilization is 85%. The edge node (such as a nearby edge server) has a latency of 40ms (<50ms) and a CPU idle rate of 60%. The cloud network bandwidth is stable but the round-trip latency is 120ms. Based on this data, it determines that the current target task (object recognition) must prioritize reducing end device power consumption and meeting real-time requirements: image preprocessing tasks (lightweight computing) are retained on the end device, using the glasses' CPU to complete preliminary feature extraction, while inference tasks (high computing power requirements) are offloaded to the edge node to reduce the end-side load. After task distribution, the system continuously monitors execution status and finds that after disabling the GPU rendering module on the end device, the battery consumption rate drops from 1.5% to 0.8% per minute, and the edge node inference latency stabilizes at 35ms, reducing the overall task processing latency by 40% compared to the original solution. Based on this execution situation, the system further optimizes the power supply strategy: it sets a dynamic voltage regulation mechanism for the end device (automatically reduces the CPU main frequency to 1.2GHz when the battery power is less than 25%), and configures a task priority queue for the edge node (prioritizing low-latency tasks). Ultimately, while ensuring recognition accuracy, the battery life of the smart glasses is extended by 30 minutes, and the task delay is controlled within 80ms.

[0092] As another example, in a scenario where a drone performs video analysis, the system first obtains real-time data from the collaborative network: it detects that the drone's (end device) battery is only 15% charged, the CPU load is 70%, the edge node (such as a ground base station) is currently idle at 80% and has sufficient network bandwidth (uplink rate of 20Mbps), the cloud server's computing resource utilization is 45%, and the round-trip latency is 180ms. Based on this data, the system performs layered processing on the current target task (video analysis): lightweight tasks such as initial decoding and noise reduction of the video stream are retained on the drone's end-side, using its local computing resources to complete preprocessing; complex target recognition tasks (such as vehicle detection) are offloaded to the cloud, leveraging the cloud's high computing power to accelerate processing; and non-real-time tasks such as video metadata storage are assigned to edge nodes to reduce the pressure on end-to-cloud data transmission.

[0093] After the task is assigned, the system continuously monitors its execution: When the drone performs preliminary processing, it reduces its battery consumption rate from 2.1% to 1.3% per minute by disabling non-essential sensors (such as lidar) and reducing the screen refresh rate. The latency for edge nodes to complete metadata storage is stabilized at 20ms, and the latency for returning target recognition results in the cloud is 150ms, reducing the overall task processing latency by 420ms compared to a purely end-to-end solution. Based on this performance, the system further optimizes the power strategy: setting a low-power mode for the drone (automatically switching to an energy-saving processor core when the battery level falls below 20%) and configuring a dynamic sleep mechanism for the edge node (entering a low-power state when no tasks are being performed). Ultimately, this extends the drone's flight time by 25 minutes and reduces the overall power consumption of the end device by 37% compared to pure end-to-end computing, achieving a balance between energy consumption and performance while ensuring real-time video analysis.

[0094] By obtaining real-time operating data and network status data of mid-end devices, source devices, and edge devices in the collaborative network, this application can fully grasp the overall resource and network status, break through the limitations of single device resource management, and make full use of the advantages of cloud-edge collaboration. Based on this, the target task execution device is determined to achieve optimal resource allocation and improve resource utilization. At the same time, monitoring the execution status and adjusting the power consumption optimization strategy can effectively reduce energy consumption. Moreover, dynamic decision-making under the collaborative network makes task execution more efficient and reduces delays. In addition, the solution can flexibly adjust the strategy according to real-time data and enhance dynamic adaptability, thereby solving the problems of low resource utilization, high energy consumption, high latency, and insufficient dynamic adaptability in existing power management methods, and improving the overall efficiency of smart device power management.

[0095] In this embodiment, a power consumption optimization device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation methods. The details that have been described will not be repeated here. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

[0096] This embodiment provides a power consumption optimization device, such as Figure 3 Shown, including:

[0097] An acquisition module 301 is configured to acquire current network status data of the collaborative network to obtain real-time operation data of each end in the collaborative network, wherein the collaborative network includes end devices, source end devices, and edge devices;

[0098] A determination module 302 is configured to determine an execution device corresponding to a target task to be distributed based on real-time operation data and network status data;

[0099] The sending module 304 is used to send the target task to the execution device and monitor the execution status of the target task by the execution device;

[0100] The execution module 304 is configured to adjust the power consumption optimization strategy of the execution device based on the execution status.

[0101] In an embodiment of the present application, the acquisition module 301 is used to collect network load data and user behavior data of the terminal devices in the collaborative network, and use the network load data and user behavior data as the real-time operation data of the terminal devices; collect the real-time operation data of the edge devices and cloud devices in the collaborative network; and determine the current network status data of the collaborative network through the network data fed back by the edge devices and cloud devices.

[0102] In an embodiment of the present application, the determination module 302 is used to obtain task requirement information of the target task; select the device that meets the task requirement information in the collaborative network as a candidate execution device; evaluate the execution capability data of the candidate execution device based on real-time operation data and network status data; verify the availability data of the candidate execution device with the highest execution capability data; if the availability data meets the preset execution conditions, the candidate execution device with the highest execution capability data is selected as the execution data.

[0103] In an embodiment of the present application, the determination module 302 is used to obtain a pre-built execution capability evaluation model, wherein the execution capability evaluation model summarizes the weights corresponding to multiple evaluation indicators; obtains the real-time operation data and the weights corresponding to the network status data from the execution capability evaluation model; and calculates the execution capability data of the candidate device using the real-time operation data, the weights corresponding to the real-time operation data, the network status data, and the weights corresponding to the network status data.

[0104] In an embodiment of the present application, the execution module 304 is used to obtain the execution node corresponding to the target task and the expected data corresponding to each execution node; monitor the actual data of the execution device executing the target task according to the execution node; compare the actual data with the expected data to obtain the execution status of the target task at each execution node.

[0105] In an embodiment of the present application, the execution module 304 is used to analyze the execution status to obtain the key factors leading to high power consumption; generate an initial optimization strategy based on the device hardware configuration of the execution device and the key factors; test the initial optimization strategy in a simulation environment and evaluate the improvement of the initial optimization strategy; if the improvement does not meet the preset requirements, the initial optimization strategy is corrected and the corrected initial optimization strategy is evaluated until the improvement meets the preset requirements, and the final optimization strategy is used as the power consumption optimization strategy.

[0106] In an embodiment of the present application, the execution module 304 is used to analyze the correlation between key factors and the device hardware configuration to determine the key reasons leading to high power consumption; formulate an optimization strategy framework for hardware limitations based on the key reasons; and use the device hardware configuration to fill the optimization strategy framework according to the optimization direction carried by the optimization strategy framework to obtain an initial power consumption optimization strategy.

[0107] See also Figure 4 , Figure 4 is a structural diagram of a computer device provided by an optional embodiment of the present invention, such as Figure 4 As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0108] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0109] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0110] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created based on the use of a computer device for displaying a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0111] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0112] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0113] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0114] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A power consumption optimization method, characterized in that: The method comprises: Acquire current network status data of the collaborative network and real-time operation data of each end in the collaborative network, wherein the collaborative network includes the end device, the source end device, and the edge device; Determining an execution device corresponding to a target task currently to be distributed based on the real-time operation data and the network status data; Sending the target task to the execution device, and monitoring the execution status of the target task by the execution device; A power consumption optimization strategy of the execution device is adjusted based on the execution situation.

2. The method according to claim 1, characterized in that The acquiring of the current network status data of the collaborative network and the real-time operation data of each terminal in the collaborative network includes: Collecting network load data and user behavior data of the terminal devices in the collaborative network, and using the network load data and the user behavior data as real-time operation data of the terminal devices; Collecting real-time operating data of edge devices and cloud devices in the collaborative network; The current network status data of the collaborative network is determined through the network data fed back by the edge device and the cloud device.

3. The method according to claim 1, characterized in that The determining, based on the real-time operation data and the network status data, an execution device corresponding to the target task to be currently distributed includes: Obtaining task requirement information of the target task; Taking a device in the collaborative network that meets the task requirement information as a candidate execution device; evaluating execution capability data of the candidate execution device according to the real-time operation data and the network status data; Verify availability data of the candidate execution device having the highest execution capability data; If the availability data meets the preset execution condition, the candidate execution device with the highest execution capability data is used as the execution data.

4. The method according to claim 3, characterized in that The evaluating the execution capability data of the candidate execution device according to the real-time operation data and the network status data includes: Obtaining a pre-built execution capability evaluation model, wherein the execution capability evaluation model summarizes weights corresponding to multiple evaluation indicators; Obtaining weights corresponding to the real-time operation data and the network status data from the execution capability evaluation model; The execution capability data of the candidate device is calculated using the real-time operation data, the weight corresponding to the real-time operation data, the network status data, and the weight corresponding to the network status data.

5. The method according to claim 4, characterized in that The monitoring of the execution status of the target task by the execution device includes: Obtain the execution node corresponding to the target task and the expected data corresponding to each execution node; Monitoring actual data of the target task being executed by the execution device according to the execution node; The actual data is compared with the expected data to obtain the execution status of the target task at each execution node.

6. The method according to claim 1, characterized in that The adjusting the power consumption optimization strategy of the execution device based on the execution situation includes: Analyzing the execution status to obtain key factors causing high power consumption; generating an initial optimization strategy based on the device hardware configuration of the execution device and the key factors; Testing the initial optimization strategy in a simulation environment to evaluate improvements made to the initial optimization strategy; If the improvement does not meet the preset requirements, the initial optimization strategy is revised and the revised initial optimization strategy is evaluated until the improvement meets the preset requirements. The final optimization strategy is then used as the power consumption optimization strategy.

7. The method according to claim 6, characterized in that The generating of an initial power consumption optimization strategy based on the device hardware configuration of the execution device and the key factors includes: Analyze the correlation between the key factors and the hardware configuration of the device to determine the key cause of high power consumption; Develop an optimization strategy framework for hardware limitations based on key reasons; The optimization strategy framework is filled in according to the optimization direction carried by the optimization strategy framework using the hardware configuration of the device to obtain an initial power consumption optimization strategy.

8. A power consumption optimization device, characterized in that: The device comprises: an acquisition module, configured to acquire current network status data of a collaborative network and real-time operation data of each end in the collaborative network, wherein the collaborative network includes the end device, the source end device, and the edge device; A determination module, configured to determine an execution device corresponding to a target task to be currently distributed based on the real-time operation data and the network status data; a sending module, configured to send the target task to the execution device and monitor the execution status of the target task by the execution device; An execution module is used to adjust the power consumption optimization strategy of the execution device based on the execution status.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images