Task compilation optimization method, device, equipment and medium
By extracting feature parameters, simplifying data, and optimizing compilation information during the task compilation process, the problems of redundant code and invalid calculations in existing technologies are solved, achieving efficient and accurate task compilation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- PING AN TECH (SHENZHEN) CO LTD
- Filing Date
- 2026-01-14
- Publication Date
- 2026-05-29
AI Technical Summary
Existing technologies do not have dedicated algorithms for specific tasks, resulting in redundant code and invalid calculation steps during compilation, leading to low compilation efficiency and accuracy.
By traversing the task requirement library, extracting feature parameters, using preprocessing algorithms to simplify data, and combining instruction modules and preset mapping models to optimize data compilation information, targeted optimization for different tasks can be achieved.
It improves compilation efficiency and accuracy, reduces data volume, enhances compilation flexibility and speed, and reduces energy consumption through energy management.
Smart Images
Figure CN122111442A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of task compilation technology, and in particular to a task compilation optimization method, apparatus, device, and medium. Background Technology
[0002] Currently, the compilation process lacks optimization for task-specific algorithms, retaining a large amount of redundant code and ineffective computational steps common to general scenarios. Furthermore, the absence of a collaborative mechanism between data processing and instruction compilation leads to low compilation efficiency and accuracy. Therefore, a task compilation optimization method is urgently needed to address these issues. Summary of the Invention
[0003] This invention provides a task compilation optimization method, apparatus, device, and medium to solve the technical problems of low efficiency and accuracy in compilation processing caused by unoptimized algorithms and the lack of collaborative mechanisms in the prior art.
[0004] A task compilation optimization method, comprising: Traverse all preset tasks in the task requirement library and extract the feature parameters corresponding to each preset task; The task collection data corresponding to the same preset task is simplified according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task. Based on the instruction module corresponding to the feature parameters of the preset task, all the simplified data in the simplified dataset are compiled to obtain the data compilation information corresponding to each preset task. Based on the preset mapping model and real-time status data, the data compilation information is optimized to obtain task compilation optimization information corresponding to each preset task.
[0005] A task compilation optimization apparatus, comprising: The parameter extraction module is used to traverse all preset tasks in the task requirement library and extract the feature parameters corresponding to each preset task. The data simplification module is used to simplify the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task. The data compilation module is used to compile all the simplified data in the simplified dataset according to the instruction module corresponding to the feature parameters of the preset task, so as to obtain the data compilation information corresponding to each preset task. The optimization processing module is used to optimize the data compilation information according to the preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task.
[0006] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor being used to perform the above-described task compilation and optimization method.
[0007] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described task compilation and optimization method.
[0008] The aforementioned task compilation optimization method, apparatus, device, and medium, in this invention, extracts parameters from preset tasks by traversing them, thereby determining feature parameters. A preprocessing algorithm simplifies the task-collected data, eliminating redundant information and thus simplifying the data and determining the simplified dataset. Data compilation is performed on all simplified data in the simplified dataset according to the instruction module, achieving coordination between instruction compilation and data processing, thereby reducing the amount of compiled data. Furthermore, targeted optimization is achieved for different tasks, improving the flexibility and efficiency of optimization, and thus improving compilation efficiency and accuracy. Based on a preset mapping model and real-time status data, the data compilation information is optimized, enabling energy consumption control and thus improving the speed of task response. Attached Figure Description
[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention 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.
[0010] Figure 1 This is a flowchart of a task compilation optimization method in one embodiment of the present invention; Figure 2 This is a schematic diagram of a task compilation optimization device in one embodiment of the present invention. Detailed Implementation
[0011] 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, not all, of the embodiments of the present invention. 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.
[0012] In one embodiment, such as Figure 1 As shown, a task compilation optimization method is provided, including the following steps: S10: Traverse all preset tasks in the task requirement library and extract the feature parameters corresponding to each preset task.
[0013] In essence, a task requirement library refers to a collection of all tasks built according to application scenarios. Preset tasks are tasks that are pre-defined, and these tasks differ across different scenarios. For example, in a medical and elderly care scenario, preset tasks might include fall detection, medication reminders, and voice-based medical consultations. Feature parameters refer to key information in each preset task, such as response requirements and computational power requirements.
[0014] Specifically, a task requirement library for the medical and elderly care scenario is obtained, which includes all preset tasks in the medical and elderly care scenario. Then, at least one core indicator corresponding to each preset task is obtained. The compiler is then controlled to traverse all preset tasks in the task requirement library, identify the key information corresponding to each task's core indicator, and extract all key information as feature parameters. In this way, the feature parameters corresponding to each preset task can be extracted.
[0015] S20: The task collection data corresponding to the same preset task is simplified according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task.
[0016] In essence, a preprocessing algorithm refers to a pre-defined algorithm used to streamline task data. Task-acquired data refers to the relevant data collected by various sensors (such as cameras, heart rate monitors, microphones, and accelerometers) for each pre-defined task. A streamlined dataset is the collection of all streamlined data features. One pre-defined task corresponds to one streamlined dataset.
[0017] Specifically, the system receives task data collected for each preset task. Then, based on task priority, it calls the preprocessing algorithm corresponding to each preset task in priority order. Next, the task data collected for the same preset task is simplified according to the preprocessing algorithm, that is, redundant data is removed from the task data collected by the preprocessing algorithm, thus obtaining a simplified dataset corresponding to each preset task. For example, for high-priority tasks (such as fall detection), the preprocessing algorithm for that preset task is called first, filtering out environmental vibration interference signals from the task data collected by the accelerometer and retaining the core feature of the vertical acceleration rate of change.
[0018] S30: Based on the instruction module corresponding to the feature parameters of the preset task, perform data compilation on all the simplified data in the simplified dataset to obtain data compilation information corresponding to each preset task.
[0019] Intuitively, an instruction module refers to a functional module that processes compiled instructions, with different instruction modules corresponding to different task types. For example, fall detection corresponds to an "image feature-motion instruction" module, and medication reminders correspond to a "time text-voice instruction" module. Data compilation information refers to the machine code resulting from compiling all simplified data. Machine code is binary instruction code that can be directly recognized and executed, typically represented by a sequence of 0s and 1s.
[0020] Specifically, the compiler uses the characteristic parameters of each preset task to call the corresponding instruction modules from the instruction template library. Then, based on all instruction modules, it compiles all the simplified data in the simplified dataset. That is, it uses a computing power priority allocation strategy for the compilation of instructions for high-priority tasks, while a time-sharing compilation strategy is used for the compilation of instructions for low-priority tasks, generating instructions during the idle periods of high-priority tasks. In this way, by compiling all the simplified data according to the compilation instructions in the instruction modules, the data compilation information corresponding to each preset task can be obtained. For example, when the simplified dataset for the fall detection task is input, the compiler pauses the compilation of instructions for low-priority tasks and prioritizes the use of 80% of the AI chip's computing power to generate coherent machine code (i.e., data compilation information) containing "emergency braking, alarm, and attitude adjustment".
[0021] S40: Based on the preset mapping model and real-time status data, the data compilation information is optimized to obtain task compilation optimization information corresponding to each preset task.
[0022] Understandably, the preset mapping model refers to a pre-defined energy consumption mapping model. Real-time status data refers to the real-time status data of the AI chip. Task compilation optimization information refers to the low-energy execution code optimized for each task; the low-energy execution code refers to the code required to execute each preset task.
[0023] Specifically, a preset mapping model is obtained, and energy consumption coefficients are mapped to the data compilation information using this model. That is, by using the energy consumption coefficients of different data compilation information labeled with historical data, the energy consumption coefficients of the data compilation information corresponding to each preset task are determined, thus obtaining the energy consumption coefficients corresponding to each data compilation information. For example, the energy consumption coefficient for machine code fragments in image feature extraction is 0.8, and the energy consumption coefficient for machine code fragments in text processing is 0.3. Real-time chip status data is obtained, and energy consumption is trimmed from the data compilation information using this real-time status data. That is, when the chip's real-time status data exceeds a preset threshold, the data compilation information for different tasks is adjusted to reduce energy consumption, thereby obtaining task compilation optimization information corresponding to each preset task.
[0024] In one embodiment, within the financial field, for the task of "analyzing 300 market data entries from Shanghai and Shenzhen stock exchanges," characteristic parameters are extracted: data collection frequency, market data dimensions (transaction price, trading volume, buy and sell orders), calculation threshold, execution timeliness, and computing power / power consumption threshold. Next, invalid market data (such as redundant data with no price fluctuations and abnormal order data) is filtered out, and key dimensions (only transaction price, trading volume, and price change) are retained, reducing the original 1000 market data entries per second to 100 core data entries per second. Then, instruction modules matching the characteristic parameters (such as CPU instruction set optimization modules and GPU parallel computing modules) are called to compile the simplified market data into hierarchical machine code: Level 1 code: basic market data calculation (low energy consumption); Level 2 code: arbitrage signal verification (medium energy consumption); Level 3 code: emergency stop-loss instruction (high priority, low energy consumption) (i.e., data compilation information). Next, the energy consumption status corresponding to each machine code is mapped through a preset mapping model, and then, based on real-time market status (volatility, trading volume) and energy consumption status, low-energy execution code (i.e., task compilation optimization information) is output. Example: When market volatility is less than or equal to the threshold (low activity), redundant verification logic of the secondary code is trimmed to reduce the energy consumption of the executed code; when volatility is greater than the threshold (high activity), all logic is retained but the instruction execution order is optimized.
[0025] In one embodiment, in the medical field, for "real-time ECG analysis," feature parameters are extracted: sampling rate, analysis dimensions (heart rate, ST segment shift, QRS width), alarm threshold, and device power consumption limit. Next, baseline noise (such as electromyography interference and power frequency interference) is filtered, only core QRS complex data is retained (excluding resting periods without waveform fluctuations), and non-abnormal period data is downsampled, reducing the original 250 data points per second to 50 core points per second. Then, an instruction module matching the feature parameters (adapted to low-power MCU chips) is called and compiled into hierarchical machine code: Level 1 code: baseline heart rate calculation (medium power consumption); Level 2 code: abnormal waveform identification (high power consumption); Level 3 code: alarm command execution (low power consumption, triggered only after anomaly confirmation) (i.e., data compilation information). Next, a preset mapping model maps the energy consumption status of the machine code, and then, based on the patient's real-time status (resting / exercise), ECG waveform complexity, and energy consumption status, low-power execution code (i.e., task compilation optimization information) is output. Example: When the patient is in motion (high waveform noise), enhance the noise filtering logic of the first-level code while reducing the trigger frequency of the second-level code (to avoid false positives); when the device's remaining battery power is less than the threshold, trim the non-core waveform analysis of the second-level code, retaining only heart rate and extreme abnormality identification.
[0026] The task compilation optimization method in this embodiment extracts parameters from preset tasks by traversing them, thereby determining feature parameters. A preprocessing algorithm simplifies the task-collected data, eliminating redundant information and defining the simplified dataset. The instruction module compiles all simplified data within the dataset, achieving coordination between instruction compilation and data processing, thus reducing the amount of compiled data. Furthermore, targeted optimization is performed on different tasks, improving the flexibility and efficiency of optimization, and consequently enhancing compilation efficiency and accuracy. Based on a preset mapping model and real-time status data, the data compilation information is optimized, controlling energy consumption and improving task response speed.
[0027] In one embodiment, after step S40, after optimizing the data compilation information according to the preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task, the method further includes: S501, receive the running data of the preset task, and filter out the abnormal data in all the running data that do not conform to the core indicators of the task corresponding to each preset task.
[0028] S502, based on all the abnormal data, determine the abnormal cause corresponding to each of the preset tasks.
[0029] S503, update the compilation rules in the compilation rule base according to all the above-mentioned abnormal reasons to obtain an updated compilation rule base; the compilation rule base includes task priority, energy consumption control strategy, and preprocessing algorithm.
[0030] Understandably, the compilation rule base includes, but is not limited to, task priorities, energy consumption control strategies, and preprocessing algorithms. Runtime data refers to information related to the execution of a preset task, such as actual response latency, task completion accuracy, and chip energy consumption variations. Abnormal data refers to runtime data that does not conform to the core performance indicators of the task. The cause of the abnormality refers to the reason for the abnormal data. Updating the compilation rule base refers to updating the set of compilation rules. The number of abnormal data items can be zero, meaning there is no abnormal data.
[0031] Specifically, the compiler receives all runtime data during the execution of each preset task. It then compares all runtime data with the core metrics of the task, identifying any data that does not conform to these metrics as anomalous data. Next, based on all anomalous data, the cause of the anomalous data for each preset task is determined. This involves first identifying the key fields of each anomalous data point, comprehensively analyzing the processing step nodes corresponding to each key field, and then performing a cause analysis on these processing step nodes to obtain the cause of the anomalous data for each preset task. For example, if the actual response delay of the fall detection task is 0.25 seconds, exceeding the preset requirement of 0.2 seconds, the cause of the delay is analyzed. If it is due to excessive time spent in the data preprocessing stage, the preprocessing algorithm needs to be adjusted; if it is due to excessive energy consumption control leading to insufficient computing power, the preset mapping model needs to be optimized. Finally, the compilation rules in the compilation rule base are updated according to all the reasons for the anomalies. That is, each compilation rule is updated according to the reason for the anomaly in order to eliminate the anomaly and thus obtain an updated compilation rule base. For example, if the data preprocessing stage takes too long, the image feature extraction step is simplified; if the energy consumption regulation is too excessive and the computing power is insufficient, the energy consumption coefficient threshold of high priority tasks is increased.
[0032] In this embodiment, by running data, abnormal data is filtered, thereby enabling the analysis of the causes of abnormal data, the determination of the causes of abnormalities, and the updating of compilation rules, thus improving the accuracy and efficiency of compilation.
[0033] In one embodiment, in step S10, traversing all preset tasks in the task requirement library and extracting feature parameters corresponding to each preset task includes: S101, Obtain the core metrics of the tasks corresponding to each of the preset tasks.
[0034] S102, Traverse the task requirement library, extract parameters for each preset task through the core task indicators, and obtain the feature parameters corresponding to each preset task.
[0035] Understandably, core task metrics refer to the metrics marked in each pre-defined task, such as response latency requirements, data types, and computing power requirements. Each pre-defined task includes at least one core task metric.
[0036] Specifically, all preset tasks corresponding to the current scenario are obtained. Then, the core task metrics are extracted from all metrics corresponding to each preset task. Next, the task requirement library is traversed, and parameters are extracted for each preset task based on the core task metrics. That is, for each preset task in the task requirement library, the key fields corresponding to each task's core metrics are queried, and the feature parameters corresponding to each key field are extracted from the preset task. In this way, the feature parameters corresponding to each task's core metrics can be extracted from all preset tasks. For example, the core task metrics are response latency requirement, data type, and computing power requirement. The feature parameters are: for the fall detection task, the response latency requirement is less than or equal to 0.2 seconds, the data type is image and acceleration data, and the computing power requirement is medium to high load; for the medication reminder task, the response latency requirement is less than or equal to 1 second, the data type is time and text data, and the computing power requirement is low load.
[0037] In this embodiment, key information in the preset task is extracted through the core task indicators, thereby avoiding the retention of redundant information, reducing the amount of data to be compiled in the subsequent compilation, and improving the efficiency of subsequent compilation.
[0038] In one embodiment, in step S20, the step of simplifying the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task includes: S201, according to the task priority corresponding to each preset task, call the preprocessing algorithm corresponding to each preset task.
[0039] S202, receive task collection data corresponding to each of the preset tasks, and perform simplification processing on the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each of the preset tasks.
[0040] Specifically, after obtaining the feature parameters, the task priorities are obtained by sorting them according to their urgency. For example, the priority of fall detection is higher than that of emergency calls, and the priority of emergency calls is higher than that of casual conversation. Then, based on the task priorities corresponding to each preset task, the preprocessing algorithm corresponding to that preset task is called in sequence. Next, the task collection data corresponding to each preset task is received, and the task collection data is simplified using the preprocessing algorithm corresponding to the same preset task. That is, the key information in the task collection data is extracted by the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task. Different preprocessing algorithms handle different task collection data in different ways. One approach is to first receive the task collection data, determine the corresponding preset task, then determine the task priority of that preset task, and then call the preprocessing algorithm corresponding to that preset task.
[0041] In one embodiment, for high-priority tasks (such as fall detection), the preprocessing algorithm for that preset task is invoked first. This involves cropping irrelevant backgrounds from the raw images captured by the camera, retaining only the image data of the elderly person's body area, and extracting key features such as joint movement trajectories and body tilt angles, while removing redundant pixel information. For low-priority tasks (such as daily environmental monitoring), a lightweight preprocessing algorithm is used to extract basic statistical features of the task-collected data (such as average brightness and ambient noise decibels). After preprocessing, a simplified dataset that precisely matches each preset task is generated; the amount of data in this feature set is reduced compared to the task-collected data.
[0042] In this embodiment, task priorities are used to invoke the processing algorithm. The preprocessing algorithm simplifies the task-collected data, removing redundant information and obtaining a simplified dataset. This allows for targeted optimization of tasks with different priorities, improving the flexibility and efficiency of the optimization process.
[0043] In one embodiment, in step S30, the step of compiling all simplified data in the simplified dataset according to the instruction module corresponding to the feature parameters of the preset task to obtain data compilation information corresponding to each preset task includes: S301, based on all the aforementioned feature parameters, match the instruction module corresponding to each of the preset tasks from the task instruction template library.
[0044] S302, based on task priority, the instruction module compiles all the simplified data corresponding to each preset task to obtain data compilation information corresponding to each preset task.
[0045] Specifically, after obtaining the simplified dataset, instruction modules corresponding to each preset task are matched from the task instruction template library based on all feature parameters. That is, dedicated instruction modules are matched from the task instruction template library based on the task type in the feature parameters, thus obtaining all instruction modules corresponding to each preset task. Then, based on task priority, all simplified data corresponding to each preset task is compiled using the instruction modules. Specifically, a computing power priority allocation strategy is used for compiling instructions for high-priority tasks, while a time-sharing compilation strategy is used for compiling instructions for low-priority tasks, generating instructions during the idle periods of high-priority tasks. Thus, according to the compilation instructions in the instruction modules, all simplified data is compiled to obtain the hierarchically optimized machine code corresponding to each preset task, and all hierarchically optimized machine code is identified as data compilation information.
[0046] In this embodiment, feature parameters are used to match instruction modules, thereby determining the compilation instructions. Based on task priority and instruction modules, compilation instructions are executed sequentially, enabling hierarchical compilation of simplified data, determining data compilation information, and thus improving compilation accuracy and efficiency.
[0047] In one embodiment, in step S40, optimizing the data compilation information based on a preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task includes: S401, by using the energy consumption coefficients corresponding to each historical compilation information in the preset mapping model, coefficient mapping is performed on all the data compilation information to obtain the task energy consumption coefficients corresponding to each preset task.
[0048] S402, by using the real-time status data corresponding to each preset task, the energy consumption of the data compilation information corresponding to the same preset task is trimmed to obtain the optimized compilation information corresponding to each preset task.
[0049] S403, Based on the task energy consumption coefficient and the optimization compilation information corresponding to the same preset task, generate task compilation optimization information corresponding to each preset task.
[0050] In essence, the task energy consumption factor is a parameter that quantifies the energy consumption characteristics of executing a single line or group of machine code (instructions that the CPU can directly execute), used to evaluate and calculate the energy consumption during machine code execution. Optimization compilation information refers to machine code after energy consumption has been controlled.
[0051] Specifically, by using the energy consumption coefficients corresponding to each historical compilation information in the preset mapping model, coefficient mapping is performed on all data compilation information. This matches the data compilation information with the machine code segments corresponding to historical data. When a match is successful, the energy consumption coefficient of the machine code segment corresponding to the historical data is determined as the task energy consumption coefficient corresponding to each preset task. Simultaneously, the compiler acquires real-time status data of the AI chip and uses this data to perform energy consumption trimming on the data compilation information corresponding to the same preset task. This involves adjusting the energy consumption of low-priority tasks to reduce energy consumption, thereby obtaining optimized compilation information corresponding to each preset task. For example, when the chip's computing power utilization exceeds a preset computing power threshold, the machine code segments of low-priority tasks are simplified (e.g., reducing the precision of text segmentation), lowering the energy consumption coefficient. When the chip temperature exceeds a preset temperature threshold, machine code compilation for non-urgent tasks is paused. Then, based on the task energy consumption coefficients and optimized compilation information corresponding to the same preset task, optimized task compilation information corresponding to each preset task is generated. In other words, low-energy execution code is generated using the energy consumption coefficients and optimized compilation information and determined as the task compilation optimization information.
[0052] In this embodiment, a preset mapping model is used to map the energy consumption coefficient of the data compilation information. Real-time status data is used to tailor the energy consumption of the data compilation information, thereby determining the optimized compilation information and controlling energy consumption, thus reducing the energy consumption of executing preset tasks.
[0053] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0054] In one embodiment, a task compilation optimization apparatus is provided, which corresponds one-to-one with the task compilation optimization methods described in the above embodiments. For example... Figure 2 As shown, the task compilation and optimization device includes a parameter extraction module 10, a data simplification module 20, a data compilation module 30, and an optimization processing module 40. Detailed descriptions of each functional module are as follows: The parameter extraction module 10 is used to traverse all preset tasks in the task requirement library and extract the feature parameters corresponding to each preset task. The data simplification module 20 is used to simplify the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task. Data compilation module 30 is used to compile all the simplified data in the simplified dataset according to the instruction module corresponding to the feature parameters of the preset task, so as to obtain data compilation information corresponding to each preset task. The optimization processing module 40 is used to optimize the data compilation information according to the preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task.
[0055] In one embodiment, the device further includes: An anomaly filtering unit is used to receive the running data of executing preset tasks and filter out the abnormal data that does not conform to the core indicators of the task corresponding to each preset task from all the running data. The cause analysis unit is used to determine the cause of the abnormality corresponding to each of the preset tasks based on all the abnormal data. The rule update unit is used to update the compilation rules in the compilation rule base according to all the above-mentioned abnormal reasons, so as to obtain an updated compilation rule base; the compilation rule base includes task priority, energy consumption control strategy, and preprocessing algorithm.
[0056] In one embodiment, the parameter extraction module 10 includes: The indicator acquisition unit is used to acquire the core indicators of the task corresponding to each of the preset tasks. The parameter extraction unit is used to traverse the task requirement library and extract parameters for each preset task through the core task indicators to obtain the feature parameters corresponding to each preset task.
[0057] In one embodiment, the data simplification module 20 includes: The algorithm invocation unit is used to invoke the preprocessing algorithm corresponding to each preset task according to the task priority corresponding to each preset task. The data simplification unit is used to receive task collection data corresponding to each of the preset tasks, and to simplify the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each of the preset tasks.
[0058] In one embodiment, the data compilation module 30 includes: The instruction matching unit is used to match the instruction module corresponding to each of the preset tasks from the task instruction template library according to all the feature parameters. The data compilation unit is used to compile all the simplified data corresponding to each preset task based on task priority through the instruction module to obtain data compilation information corresponding to each preset task.
[0059] In one embodiment, the optimization processing module 40 includes: The coefficient mapping unit is used to perform coefficient mapping on all the data compilation information through the energy consumption coefficients corresponding to each historical compilation information in the preset mapping model, so as to obtain the task energy consumption coefficients corresponding to each preset task. The energy consumption trimming unit is used to trim the energy consumption of the data compilation information corresponding to the same preset task by using the real-time status data corresponding to each preset task, so as to obtain the optimized compilation information corresponding to each preset task. The information generation unit is used to generate task compilation optimization information corresponding to each of the preset tasks based on the task energy consumption coefficient and the optimization compilation information corresponding to the same preset task.
[0060] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor being used to perform the above-described task compilation and optimization method.
[0061] Specific limitations regarding the computer equipment, processor, and their various units and modules can be found in the above-described limitations of the task compilation optimization method, and will not be repeated here. Each module in the aforementioned processor can be implemented entirely or partially through software, hardware, or a combination thereof. Understandably, the processor includes a processor, memory, network interface, and database connected via a device bus. Each module of the processor can be embedded in hardware or independent of the processor, or stored in memory in software form, so that the processor can call and execute the operations corresponding to each module. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores operating devices, computer programs, and a database. The internal memory provides an environment for the operation of the operating devices and computer programs in the non-volatile storage media. The database stores the data used by the task compilation optimization method in the above embodiments. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a task compilation optimization method.
[0062] In one embodiment, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described task compilation and optimization method.
[0063] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0064] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0065] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A task compilation optimization method, characterized in that, include: Traverse all preset tasks in the task requirement library and extract the feature parameters corresponding to each preset task; The task collection data corresponding to the same preset task is simplified according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task. Based on the instruction module corresponding to the feature parameters of the preset task, all the simplified data in the simplified dataset are compiled to obtain the data compilation information corresponding to each preset task. Based on the preset mapping model and real-time status data, the data compilation information is optimized to obtain task compilation optimization information corresponding to each preset task.
2. The task compilation optimization method as described in claim 1, characterized in that, After optimizing the data compilation information based on the preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task, the process further includes: Receive the execution data of the preset tasks, and filter out the abnormal data in all the execution data that do not conform to the core indicators of the tasks corresponding to each preset task; Based on all the abnormal data, determine the cause of the abnormality corresponding to each of the preset tasks; The compilation rules in the compilation rule base are updated according to all the aforementioned reasons for the anomalies, resulting in an updated compilation rule base; the compilation rule base includes task priority, energy consumption control strategy, and preprocessing algorithm.
3. The task compilation optimization method as described in claim 1, characterized in that, The process involves traversing all preset tasks in the task requirement database and extracting feature parameters corresponding to each preset task, including: Obtain the core metrics corresponding to each of the preset tasks; Traverse the task requirement library and extract parameters from each preset task using the core task indicators to obtain the feature parameters corresponding to each preset task.
4. The task compilation optimization method as described in claim 1, characterized in that, The step of simplifying the task-collected data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task includes: According to the task priority corresponding to each of the preset tasks, the preprocessing algorithm corresponding to each of the preset tasks is invoked; Receive task collection data corresponding to each of the preset tasks, and perform simplification processing on the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each of the preset tasks.
5. The task compilation optimization method as described in claim 1, characterized in that, The step of compiling all simplified data in the simplified dataset according to the instruction module corresponding to the feature parameters of the preset task, to obtain data compilation information corresponding to each preset task, includes: Based on all the aforementioned feature parameters, the instruction modules corresponding to each of the aforementioned preset tasks are matched from the task instruction template library; Based on task priority, the instruction module compiles all the simplified data corresponding to each preset task to obtain data compilation information corresponding to each preset task.
6. The task compilation optimization method as described in claim 1, characterized in that, The step of optimizing the data compilation information based on a preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task includes: By using the energy consumption coefficients corresponding to each historical compilation information in the preset mapping model, coefficient mapping is performed on all the data compilation information to obtain the task energy consumption coefficients corresponding to each preset task. By using the real-time status data corresponding to each preset task, the energy consumption of the data compilation information corresponding to the same preset task is trimmed to obtain the optimized compilation information corresponding to each preset task. Based on the task energy consumption coefficient and the optimization compilation information corresponding to the same preset task, task compilation optimization information corresponding to each preset task is generated.
7. A task compilation optimization apparatus, characterized in that, include: The parameter extraction module is used to traverse all preset tasks in the task requirement library and extract the feature parameters corresponding to each preset task. The data simplification module is used to simplify the task collection data corresponding to the same preset task according to the preprocessing algorithm to obtain a simplified dataset corresponding to each preset task. The data compilation module is used to compile all the simplified data in the simplified dataset according to the instruction module corresponding to the feature parameters of the preset task, so as to obtain the data compilation information corresponding to each preset task. The optimization processing module is used to optimize the data compilation information according to the preset mapping model and real-time status data to obtain task compilation optimization information corresponding to each preset task.
8. The task compilation optimization apparatus as described in claim 7, characterized in that, The optimization processing module includes: The coefficient mapping unit is used to perform coefficient mapping on all the data compilation information through the energy consumption coefficients corresponding to each historical compilation information in the preset mapping model, so as to obtain the task energy consumption coefficients corresponding to each preset task. The energy consumption trimming unit is used to trim the energy consumption of the data compilation information corresponding to the same preset task by using the real-time status data corresponding to each preset task, so as to obtain the optimized compilation information corresponding to each preset task. The information generation unit is used to generate task compilation optimization information corresponding to each of the preset tasks based on the task energy consumption coefficient and the optimization compilation information corresponding to the same preset task.
9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor being configured to perform the task compilation optimization method as described in any one of claims 1 to 6.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the task compilation optimization method as described in any one of claims 1 to 6.