A task running data determination method, apparatus, medium, device and product

Abstract

The application discloses a task running data determination method, device, medium, equipment and product, wherein the method comprises: in response to the running of a target task, obtaining the benchmark running data of the target task; obtaining the difference between the real-time load data of the task running equipment in response to the running of the target task and the historical load benchmark data of the task running equipment, determining the real-time load influence data of the task running equipment based on the difference and the first load weight; obtaining the average load data of the task running equipment within a preset time period before the running of the target task, determining the lag load influence data of the task running equipment based on the average load data and the second load weight; using a prediction algorithm and determining the predicted running data of the target task based on the benchmark running data, the real-time load influence data and the lag load influence data. The application can improve the accuracy of the prediction of the task running data and ensure the accuracy of the scheduling decision.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to a method, apparatus, medium, device, and product for determining task execution data. Background Technology

[0002] As computer technology continues to develop and improve, service equipment processes tasks corresponding to different service demands in response to the ever-increasing service requirements. During actual task processing, a scheduling system is often used to allocate resources and manage processing progress, thus requiring accurate assessment of the runtime of different tasks. However, in the prediction of actual task runtime data, estimations are typically made only based on historical runtime data, leading to significant prediction errors and impacting the accuracy of the scheduling system's scheduling decisions. Summary of the Invention

[0003] This specification provides a method, apparatus, medium, equipment, and product for determining task operation data, which can improve the accuracy of task operation data prediction and ensure the accuracy of scheduling decisions.

[0004] Firstly, embodiments of this specification provide a method for determining task execution data, including:

[0005] In response to the execution of a target task, baseline execution data of the target task is obtained. The baseline execution data is determined based on the historical execution data of the target task. The baseline execution data includes baseline runtime and baseline resource usage data. The baseline resource usage data is the baseline utilization rate of the processor and the baseline memory usage during the execution of the target task.

[0006] The difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device is obtained. Based on the difference and a first load weight, the real-time load impact data of the task running device is determined. The real-time load data includes the current utilization rate of the processor and the current utilization of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical utilization of memory.

[0007] The average load data of the task running device during a preset time period before running the target task is obtained. Based on the average load data and the second load weight, the lag load impact data of the task running device is determined. The average load data is the average utilization of the processor and the average usage of memory.

[0008] A prediction algorithm is used, and based on the baseline operating data, the real-time load impact data, and the delayed load impact data, the predicted operating data for the target task is determined.

[0009] The prediction algorithm is defined as follows: Predicted operating data = Baseline operating data × (1 + Real-time load impact data + Lagging load impact data).

[0010] Secondly, embodiments of this specification provide a task execution data determination device, comprising:

[0011] A benchmark data acquisition unit is configured to acquire benchmark operation data of the target task in response to the operation of the target task. The benchmark operation data is determined based on the historical operation data of the target task. The benchmark operation data includes benchmark runtime and benchmark resource usage data. The benchmark resource usage data is the benchmark utilization rate of the processor and the benchmark memory usage during the operation of the target task.

[0012] The impact data acquisition unit is used to acquire the difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device, and to determine the real-time load impact data of the task running device based on the difference and a first load weight. The real-time load data includes the current utilization rate of the processor and the current utilization of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical utilization of memory.

[0013] The impact data acquisition unit is further configured to acquire the average load data of the task running device within a preset time period before running the target task, and determine the lag load impact data of the task running device based on the average load data and the second load weight, wherein the average load data is the average utilization rate of the processor and the average usage of memory.

[0014] The data determination unit is used to use a prediction algorithm and, based on the baseline operating data, the real-time load impact data, and the delayed load impact data, determine the predicted operating data of the target task.

[0015] The prediction algorithm is defined as follows: Predicted operating data = Baseline operating data × (1 + Real-time load impact data + Lagging load impact data).

[0016] Thirdly, embodiments of this specification provide an electronic device, including: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to execute the steps of the method described above.

[0017] Fourthly, embodiments of this specification provide a storage medium storing a computer program that, when executed by a processor, implements the steps of the method described above.

[0018] Fifthly, embodiments of this specification provide a computer program product, including: a computer program that, when executed by a processor of an electronic device, causes the processor to perform the steps of the method described above.

[0019] In the embodiments of this specification, the baseline operating data of the target task is obtained during the stage when the target task is about to start running. The baseline operating data is used as the anchor point of the task operating data. The target task operating data is dynamically calibrated by combining the impact of the current load of the task running equipment on the operation of the target task and the impact of the load on the operation of the target task in the period before the target task runs, so as to obtain the final predicted operating data of the target task. This improves the accuracy of the prediction of the task operating data and ensures the accuracy of the scheduling decision. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A system architecture diagram for determining task execution data provided in the embodiments of this specification;

[0022] Figure 2 A flowchart illustrating a method for determining task execution data provided in an embodiment of this specification;

[0023] Figure 3 A flowchart illustrating a method for determining task execution data provided in an embodiment of this specification;

[0024] Figure 4 This is a schematic diagram illustrating an example of a method for determining task execution data provided in an embodiment of this specification.

[0025] Figure 5 This is a schematic diagram of the structure of a task operation data determination device provided in the embodiments of this specification;

[0026] Figure 6 This is a schematic diagram of the structure of a task operation data determination device provided in the embodiments of this specification;

[0027] Figure 7This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. Detailed Implementation

[0028] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0029] Please see Figure 1 This specification provides a system architecture diagram for determining task execution data in its embodiments. Figure 1 As shown, service device 200 is widely used for data processing in various fields, such as big data scheduling platforms, job management systems, and automated operation and maintenance systems. Data processing types can include logical processing of various front-end applications, log management, and data computation. During data processing, tasks are often dispatched to service device 200 via scheduling system 100 for execution. Scheduling system 100 needs to manage the load and resource usage of service device 200, and also control the order of task execution. In this embodiment, service device 200 can be a single task execution device or a cluster of task execution devices, depending on the resources required for data processing in different fields. Scheduling system 100 can be integrated with service device 200 in the same device, or it can be a background device independent of service device 200.

[0030] Ideally, task execution data is often constant. This data can include, but is not limited to, runtime and required resource data, such as processor and memory resources. However, in actual processing, task execution data is affected by the load and resource usage of the service device 200, causing instability in the required data and impacting the service device 200's response time, potentially even leading to its failure. Therefore, by introducing a scheduling system 100, elastic time windows can be calculated and resources reserved based on predictions of different task execution data. In related technologies, execution data prediction is typically done by directly estimating data from several historical executions of the task, resulting in significant prediction errors and affecting the accuracy of the scheduling system 100's scheduling decisions.

[0031] To address the aforementioned issues, this specification provides a method for determining task execution data. When preparing to launch a target task, the method calibrates the target task execution data by acquiring baseline execution data determined from the target task's historical execution data, and by acquiring data on the impact of the current load of the task's running equipment on the target task's execution, as well as the impact of the load on the target task's execution in the period prior to the task's execution. This results in the final predicted execution data for the target task. By combining the task's baseline execution data with the real-time load impact of the task's running equipment and the load impact before task launch, dynamic calibration of the task execution data is achieved, improving the accuracy of task execution data prediction and ensuring the accuracy of scheduling decisions.

[0032] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.), and signals involved in the embodiments of this specification are mainly data obtained from public channels and do not involve the acquisition of privacy data. If acquisition is necessary, it is done with the authorization of the user or with full authorization from all parties, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the target application-related data involved in this specification were all obtained with full authorization.

[0033] based on Figure 1 The system architecture shown below will be combined with... Figures 2-4 The method for determining task execution data provided in the embodiments of this specification will be described in detail.

[0034] Please see Figure 2 This is a flowchart illustrating a method for determining task execution data, provided in an embodiment of this specification. Figure 2 As shown, the method in the embodiments of this specification may include the following steps S102-S106.

[0035] S102, in response to the execution of the target task, acquire the baseline execution data of the target task;

[0036] Specifically, in response to the execution of a target task, the scheduling system can obtain baseline operational data for the target task. The target task can be any type of task that the task execution device needs to process. The baseline operational data can be used to represent the operation of the target task over a historical period, such as the baseline runtime and baseline resource usage data. The baseline resource usage data includes the baseline utilization rate of the processor and the baseline memory usage during the execution of the target task. The specific baseline operational data can be predetermined based on the historical operational data of the target task. The historical time period can be in units such as months, weeks, or days, and can be set according to actual needs.

[0037] S104, obtain the difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device, and determine the real-time load impact data of the task running device based on the difference and the first load weight;

[0038] S106, Obtain the average load data of the task running device during a preset time period before running the target task, and determine the lag load impact data of the task running device based on the average load data and the second load weight.

[0039] Specifically, the scheduling system can acquire real-time load impact data and delayed load impact data of the task-running devices. Real-time load impact data represents the influence of the real-time load data of the task-running devices on the operation of the target task, indicating the load situation in response to the execution of the target task. Delayed load impact data represents the influence of load data over a period of time prior to the execution of the target task, specifically the average load data of the task-running devices over a preset time period before the target task is executed.

[0040] In the embodiments of this specification, regarding real-time load impact data, the scheduling system can first acquire real-time load data. Real-time load data can be the load data of the task running device in response to the target task, such as the current processor utilization rate and the current memory usage. The scheduling system can acquire the current processor utilization rate and the current memory usage of the task running device in response to the target task.

[0041] The system obtains the difference between real-time load data and historical load baseline data of the task running device. Historical load baseline data represents the load status of the task running device over multiple consecutive historical periods. Examples include historical processor utilization and memory usage within each historical period. The specific historical load baseline data can be pre-determined based on the historical load data of the task running device. The historical period can be in units such as month, week, or day, and can be set according to actual needs.

[0042] Furthermore, historical load baseline data can include historical usage baseline data for the processor and historical usage baseline data for memory in the task processing device. Specifically, the historical usage baseline data for the processor can be used to represent representative usage data of the task processing device over multiple historical periods. This is determined by using a set of historical processor usage data from the historical load data. For example, the historical utilization rate of the processor in each historical period can be obtained to form a set of historical processor utilization rates. The median of this set can then be used as the historical utilization rate of the processor corresponding to the median, which is then determined as the historical utilization baseline data for the processor. Alternatively, the average historical utilization rate in the set of historical processor utilization rates can be calculated, and this average historical utilization rate can be used as the historical utilization baseline data for the processor.

[0043] Historical memory usage benchmark data can be used to represent representative usage data of a task processing device over multiple historical periods. Specifically, it is determined by the historical memory usage set in historical load data. For example, the historical memory usage of the task processing device in each historical period can be obtained to form a historical memory usage set. The median of the historical memory usage set can be obtained, and the historical memory usage corresponding to the median can be determined as the historical memory usage benchmark data. Alternatively, the average usage in the historical memory usage set can be calculated, and the average usage can be determined as the historical memory usage benchmark data.

[0044] In the embodiments of this specification, regarding the impact data of delayed load, the scheduling system can obtain the average load data of the task running device within a preset time period before running the target task. The average load data can represent the load status of the task running device before the target task runs, and the preset time period is a selected time period before the target task runs, such as 5 minutes. Furthermore, in response to the running of the target task, target load data for each preset period within the preset time period can be obtained to form a target load data set. The average load data can then be used to determine the average load data of the task running device. The average load data may include the average processor utilization and the average memory usage, and the target load data set may include a target processor utilization set and a target memory usage set.

[0045] After obtaining the average load data, the lag load impact data of the task running equipment can be determined based on the average load data and the second load weight. The second load weight can be obtained by fitting the target load data set of the running processing equipment.

[0046] S108, using a prediction algorithm, and based on the baseline operating data, the real-time load impact data, and the delayed load impact data, determine the predicted operating data for the target task;

[0047] Specifically, the prediction algorithm integrates baseline operating data, real-time load impact data, and delayed load impact data to obtain the predicted operating data for the target task.

[0048] The prediction algorithm can be: Predicted running data = Baseline running data × (1 + Real-time load impact data + Lagging load impact data).

[0049] Steps S102, S104, and S106 in the embodiments of this specification can be executed simultaneously or in a specific order; the embodiments of this specification do not impose any restrictions on this.

[0050] In the embodiments of this specification, the baseline operating data of the target task is obtained during the stage when the target task is about to start running. The baseline operating data is used as the anchor point of the task operating data. The target task operating data is dynamically calibrated by combining the impact of the current load of the task running equipment on the operation of the target task and the impact of the load on the operation of the target task in the period before the target task runs, so as to obtain the final predicted operating data of the target task. This improves the accuracy of the prediction of the task operating data and ensures the accuracy of the scheduling decision.

[0051] Please see Figure 3 This is a flowchart illustrating a method for determining task execution data, provided in an embodiment of this specification. Figure 3 As shown, the method in the embodiments of this specification may include the following steps S202-S218.

[0052] S202, Obtain the historical operation data of the target task within the historical time period;

[0053] S204, determine the baseline runtime of the target task based on the historical runtime set;

[0054] S206, Determine the baseline resource usage data for the target task based on the historical resource usage data set;

[0055] Specifically, historical operational data can include a set of historical runtimes and a set of historical resource usage data. Historical time periods can be in units such as months, weeks, or days, and can be set according to actual needs.

[0056] The baseline runtime can be used to represent the representative runtime of a target task within a historical time period. Specifically, it is determined by the set of historical runtimes in the historical running data. For example, the historical runtime of each run of the target task within a historical time period can be obtained to form a set of historical runtimes. The median can be obtained from the set of historical runtimes, and the historical runtime corresponding to the median can be determined as the baseline runtime. Alternatively, the average historical runtime in the set of historical runtimes can be calculated, and the average historical runtime can be determined as the baseline runtime, etc.

[0057] Baseline resource usage data can be used to represent representative resource usage data of a target task within a historical time period. Specifically, it is determined by the historical resource usage data set in the historical operation data. For example, for the baseline runtime, the historical runtime required for each run of the target task within the historical time period can be obtained to form a historical runtime set. The median can be obtained from the historical runtime set, and the historical runtime corresponding to the median can be determined as the baseline runtime. Alternatively, the average historical runtime in the historical runtime set can be calculated, and the average historical runtime can be determined as the baseline runtime, and so on.

[0058] Furthermore, the historical resource usage data set can be the set of historical processor usage and memory usage for each run of the target task within a historical time period. The baseline resource usage data can be the baseline processor usage and memory usage during the runtime of the target task. The calculation methods for the baseline processor usage and memory usage can refer to the calculation methods for the baseline runtime (e.g., taking the median or the average, etc.), which will not be elaborated here.

[0059] Optionally, when acquiring historical running data, or when determining baseline running data based on historical running data, abnormal data in the historical running data can be removed, such as abnormal data recorded when the task execution times out or fails.

[0060] S208, in response to the execution of the target task, acquire the baseline execution data of the target task;

[0061] Specifically, in response to the operation of a target task, the scheduling system can obtain the baseline operation data of the target task. The target task can be any type of task that the task running device needs to process. The baseline operation data can be used to represent the operation status of the target task in the past historical period, such as the baseline runtime of the target task, baseline resource usage data, etc. The baseline operation data can be determined in advance based on the historical operation data of the target task. In response to the operation of the target task, it can indicate that the target task is in a state of waiting to be started.

[0062] S210, obtain the running priority of the target task, and determine the sensitivity adjustment factor corresponding to the target task based on the running priority and priority weight;

[0063] Specifically, for different tasks or task types corresponding to different tasks, running priorities can be set. When responding to the running of a target task, the running priority of the target task can be further obtained, and the sensitivity adjustment factor corresponding to the target task can be determined based on the running priority and priority weight.

[0064] It should be noted that the running priority can be used to indicate the importance of the task or the task type corresponding to the task. For example, the higher the running priority, the higher the importance of the task, and the more resources the task can be allocated during the running process. If the running priority is the priority of the task type corresponding to the task, then the running priority of the task type can also be expressed as the running priority of the task.

[0065] Priority weights are weight values ​​obtained by fitting the execution priorities of different tasks or different task types using a nonlinear least squares method.

[0066] The sensitivity adjustment factor can be used to represent the degree to which the runtime of a task is affected by the load on the service device. The higher the running priority, the less the task is affected by the load on the task running device, and the smaller the sensitivity adjustment factor of the task.

[0067] Of course, the order of importance represented by the above-mentioned running priorities, as well as the correspondence between the size of the sensitivity adjustment factor and the running priority, can be adjusted according to actual needs, and will not be elaborated here.

[0068] Optionally, the sensitivity adjustment factor can be calculated using the following formula:

[0069] W(p) = 2 / [1 + e c×(1-p) ]

[0070] Where W(p) represents the sensitivity adjustment factor, p is the task's running priority, c is the priority weight, and e is the natural constant.

[0071] It is understandable that when p approaches 1, W(p) approaches 1; when p approaches 0, W(p) approaches infinity.

[0072] S212, obtain the difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device, and determine the real-time load impact data of the task running device based on the difference and the first load weight;

[0073] Specifically, the scheduling system can first acquire real-time load data. This real-time load data can be the load data of the task running device in response to the target task, such as the current processor utilization and current memory usage. The scheduling system can acquire the current processor utilization and current memory usage of the task running device in response to the target task.

[0074] The system obtains the difference between real-time load data and historical load baseline data of the task running device. Historical load baseline data represents the load status of the task running device over multiple consecutive historical periods. Examples include historical processor utilization and memory usage within each historical period. The specific historical load baseline data can be pre-determined based on the historical load data of the task running device. The historical period can be in units such as month, week, or day, and can be set according to actual needs.

[0075] Furthermore, historical load baseline data can include historical usage baseline data for the processor and historical usage baseline data for memory in the task processing device. Specifically, the historical usage baseline data for the processor can be used to represent representative usage data of the task processing device over multiple historical periods. This is determined by using a set of historical processor usage data from the historical load data. For example, the historical utilization rate of the processor in each historical period can be obtained to form a set of historical processor utilization rates. The median of this set can then be used as the historical utilization rate of the processor corresponding to the median, which is then determined as the historical utilization baseline data for the processor. Alternatively, the average historical utilization rate in the set of historical processor utilization rates can be calculated, and this average historical utilization rate can be used as the historical utilization baseline data for the processor.

[0076] Historical memory usage benchmark data can be used to represent representative usage data of a task processing device over multiple historical periods. Specifically, it is determined by the historical memory usage set in historical load data. For example, the historical memory usage of the task processing device in each historical period can be obtained to form a historical memory usage set. The median of the historical memory usage set can be obtained, and the historical memory usage corresponding to the median can be determined as the historical memory usage benchmark data. Alternatively, the average usage in the historical memory usage set can be calculated, and the average usage can be determined as the historical memory usage benchmark data.

[0077] The first difference between the current utilization of the processor and the historical utilization baseline data of the processor is obtained, and the second difference between the current utilization of the memory and the historical utilization baseline data of the memory is obtained.

[0078] The real-time load impact data of the task running equipment is determined based on the difference and the first load weight, wherein the first load weight can be obtained by fitting the historical load data of the running processing equipment.

[0079] Furthermore, the real-time load impact data of the task-running device can be determined based on the first difference and the first weight corresponding to the processor, and the second difference and the second weight corresponding to the memory. The first and second weights can be obtained by fitting historical load data of the running processing device, respectively. For example, the first weight can be obtained by fitting the historical usage data set of the processor using a nonlinear least squares method, and the second weight can be obtained by fitting the historical usage data set of the memory using a nonlinear least squares method, etc.

[0080] S214, obtain the average load data of the task running device during a preset time period before running the target task, and determine the lag load impact data of the task running device based on the average load data and the second load weight.

[0081] Specifically, the scheduling system can acquire the average load data of the task-running devices within a preset time period before running the target task. This average load data represents the load status of the task-running devices before the target task runs, and the preset time period is a selected time interval, such as 5 minutes. Furthermore, in response to the running of the target task, the system can acquire target load data for each preset period within the preset time period, forming a target load data set. The average load data of the task-running devices is then determined using this target load data set. The average load data may include the average processor utilization and the average memory usage, and the target load data set may include a set of target processor utilization and a set of target memory usage.

[0082] After obtaining the average load data, the lag load impact data of the task running equipment can be determined based on the average load data and the second load weight. The second load weight can be obtained by fitting the target load data set of the running processing equipment.

[0083] Furthermore, the average processor usage data and average memory usage data of the task running device during a preset time period before running the target task can be obtained. The average processor usage data can be used to represent the processor usage in the task running device before the target task runs, and the average memory usage data can be used to represent the memory usage in the task running device before the target task runs. The specific calculation method can refer to the aforementioned calculation process for average load data, which will not be elaborated here.

[0084] Based on the average processor usage data and its corresponding third weight, and the average memory usage data and its corresponding fourth weight, the lag load impact data of the task running device is determined. The third and fourth weights can be obtained by fitting them to the target load data set of the running processing device. For example, the third weight can be obtained by fitting the target processor usage set using a nonlinear least squares method, and the fourth weight can be obtained by fitting the target memory usage set using a nonlinear least squares method, and so on.

[0085] S216, Using a prediction algorithm, and based on the baseline operating data, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data, the predicted operating data of the target task is determined;

[0086] Specifically, the prediction algorithm integrates baseline operating data, sensitivity adjustment factor, real-time load impact data, and lag load impact data to obtain the predicted operating data for the target task.

[0087] The prediction algorithm can be: Predicted running data = Baseline running data × (1 + Sensitivity adjustment factor × Real-time load impact data + Lagging load impact data).

[0088] In the embodiments of this specification, to determine the predicted runtime of a target task, a prediction algorithm can be used, and the predicted runtime of the target task can be determined based on the baseline runtime, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data; to determine the predicted resource usage data of a target task, the prediction algorithm can be used, and the predicted resource usage data of the target task can be determined based on the baseline resource usage data, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data.

[0089] S218, if the data difference between the actual running data and the predicted running data of the target task meets a preset range, and the duration of the preset range is greater than or equal to a preset duration, then the first load weight, the second load weight and the priority weight are updated based on the nonlinear least squares method.

[0090] Specifically, if the difference between the actual running data and the predicted running data of the target task meets a preset range, and the duration of the difference is greater than or equal to a preset duration (e.g., the difference between the actual running time and the predicted running time is greater than 15%, and the duration of the difference exceeds 15% for more than 3 days), then the retraining process for the first load weight, the second load weight, and the priority weight can be triggered. Specifically, the data during the duration, such as the actual running data of the target task and the actual load impact data of the task running equipment (real-time load impact data and lagging load impact data), can be re-input into the prediction algorithm and refitted using the nonlinear least squares method to obtain the updated first load weight, second load weight, and priority weight, which can then be used to determine the predicted running data of the target task.

[0091] The specific process of fitting can be as follows:

[0092] Collect a set of training samples from the historical time period prior to this training. Each training sample may include the actual running data of the target task, the baseline running data, the running priority, as well as the first difference, the second difference, the average processor usage data, and the average memory usage data of the task running device.

[0093] The prediction algorithm during the refitting process can be:

[0094] t pred（i） =t base（i） ×[1+W(p （i） )×(a×r cpu（i） +b×r mem（i） ) + (x × r pre_cpu（i） +y×r pre_mem（i） )];

[0095] Among them, t pred t represents the predicted running data. base This represents the baseline operating data, W(p) represents the sensitivity adjustment factor, a and b represent the first and second weights in the first load weighting, respectively, x and y represent the third and fourth weights in the second load weighting, respectively, and r cpu Represents the first difference, r mem Represents the second difference, r pre_cpu r represents the average utilization of the processor.pre_mem Let represent the average memory usage, and (i) represent the i-th training sample. The parameter vector is θ = [c, a, b, x, y]. T Let be the weight parameters to be fitted, and c be the priority weight.

[0096] Calculate the relative residual for each training sample:

[0097] r i (θ)=[t pred（i） (θ)-t real（i） ] / t real（i） ;

[0098] Among them, t real Indicates the actual runtime.

[0099] Minimize the sum of squared residuals of all training samples as the objective function:

[0100] ;

[0101] Where M is the total number of training samples.

[0102] Furthermore, using the most recently determined weight parameters as initial parameters, the Levenberg-Marquardt algorithm is used for iterative solution, including: calculating the relative residuals under the initial parameters, calculating the objective function value, and solving the incremental equation to update the initial parameters: θ. (k+1) =θ (k) +△θ. Where K represents the number of iterations, and △θ is a custom variable set for the equation.

[0103] The convergence conditions for the iteration include the change in the objective function value obtained in this calculation compared to the objective function value obtained in the previous calculation being less than a preset threshold, or K reaching the preset maximum number of iterations, etc. When the convergence conditions are met, the latest weight parameters c', a', b', x', y' are obtained, thereby completing the process of weight refitting.

[0104] It should be noted that, regarding the aforementioned historical time period, historical cycle, and preset time period, the duration of the historical time period and multiple consecutive historical cycles can be the same or different. The preset time period is specifically the time period before the target task runs, and its duration can be shorter than the duration of the historical time period and multiple consecutive historical cycles. The specific settings can be made according to actual needs.

[0105] Regarding the first weight, second weight, third weight, and fourth weight mentioned above, the first weight can be the same as the second weight, and the third weight and fourth weight can be the same. For example, in the process of calculating the weight, the fitting can be performed simultaneously based on the load data of the processor and the load data of the memory in the task running device.

[0106] In the embodiments of this specification, by acquiring baseline operating data of the target task during the stage before the target task is started and running, and using the baseline operating data as the anchor point for the task operating data, and combining the impact of the current load of the task running equipment on the target task operation and the impact of the load on the target task operation in the period before the target task operation, the target task operating data is dynamically calibrated to obtain the final predicted operating data of the target task, thereby improving the accuracy of the prediction of task operating data and ensuring the accuracy of scheduling decisions. At the same time, by combining the task sensitivity adjustment factor, the priority of the task can be systematically correlated with the task's sensitivity to the load, further improving the accuracy of the prediction of task operating data. By updating the weight parameters involved in the prediction algorithm when the actual operating data of the task continuously deviates from the predicted operating data within a certain period of time, the prediction algorithm is dynamically adjusted, ensuring the effectiveness of the prediction algorithm.

[0107] Please see Figure 4 This is a schematic diagram illustrating an example of a method for determining task execution data provided in an embodiment of this specification. Figure 4 As shown, in the process of determining the predictive operation data for the target task, the historical operation data and operation priority of the target task can be obtained separately. The baseline operation data of the target task can be further determined through the historical operation data, and the sensitivity adjustment factor of the target task can be further determined through the operation priority.

[0108] For the task-running equipment, real-time load data and average load data before the task can be obtained separately. The real-time load data can be used to further determine the real-time load impact data of the task-running equipment, and the average load data can be used to further determine the lag load impact data of the task-running equipment.

[0109] Then, the prediction algorithm set in the prediction algorithm engine is used to fuse and calculate the baseline running data, sensitivity adjustment factor, real-time load impact data and lag load impact data to obtain the predicted running data of the target task.

[0110] Optionally, the calculation process of the sensitivity adjustment factor, real-time load impact data, and lagged load impact data can be incorporated into the prediction algorithm, then the prediction algorithm formula is:

[0111] Predicted running data = t base ×[1+W(p)×(a×r cpu +b×r mem ) + (x × r pre_cpu +y×r pre_mem )];

[0112] Among them, t base This represents the baseline operating data, W(p) represents the sensitivity adjustment factor, a and b represent the first and second weights in the first load weighting, respectively, x and y represent the third and fourth weights in the second load weighting, respectively, and r cpu Represents the first difference, r mem Represents the second difference, r pre_cpu r represents the average utilization of the processor. pre_mem This indicates the average amount of memory used.

[0113] It should be noted that the predicted runtime data can be at least one of the predicted runtime of the target task and the predicted resource usage data. If the predicted runtime data is the predicted runtime, then t base The calculation is performed using the baseline runtime; if the predicted runtime data is the predicted resource usage data, then t base Using baseline resource usage data for calculations, it's understandable that if the predicted resource usage data is the predicted utilization rate of the processor, then t base Using the processor's baseline utilization, if the predicted resource usage data is the predicted memory usage, then t base Baseline memory usage.

[0114] After determining the predicted operational data for the target task, the scheduling system can further calculate the elastic time window and predict resource requirements based on the predicted operational data, and finally output the scheduling decision for the target task to ensure the normal operation of the target task.

[0115] Furthermore, the embodiments in this specification also incorporate the actual operational data of the target task to update the aforementioned weights. For the specific update process, please refer to the detailed description of the above embodiments, which will not be repeated here.

[0116] In the embodiments of this specification, by acquiring baseline operating data of the target task during the stage before the target task is started and running, and using the baseline operating data as the anchor point for the task operating data, and combining the impact of the current load of the task running equipment on the target task operation and the impact of the load on the target task operation in the period before the target task operation, the target task operating data is dynamically calibrated to obtain the final predicted operating data of the target task, thereby improving the accuracy of the prediction of task operating data and ensuring the accuracy of scheduling decisions. At the same time, by combining the task sensitivity adjustment factor, the priority of the task can be systematically correlated with the task's sensitivity to the load, further improving the accuracy of the prediction of task operating data. By updating the weight parameters involved in the prediction algorithm when the actual operating data of the task continuously deviates from the predicted operating data within a certain period of time, the prediction algorithm is dynamically adjusted, ensuring the effectiveness of the prediction algorithm.

[0117] based on Figure 1 The system architecture shown below will be combined with... Figures 5-6 This specification provides a detailed description of the task execution data determination device provided in the embodiments. It should be noted that... Figures 5-6 The task execution data determination device in the above embodiment can be the scheduling system, or it can be a module in the scheduling system used to predict the runtime of the target task. Specifically, it is used to execute the functions described in this specification. Figures 2-4 The methods shown in the embodiments are illustrated for ease of explanation, showing only the parts related to the embodiments of this specification. For specific technical details not disclosed, please refer to this specification. Figures 2-4 The example shown.

[0118] Please see Figure 5 This is a schematic diagram of a task execution data determination device provided in an embodiment of this specification. Figure 5 As shown, the device 1 may include a reference data acquisition unit 11, an influence data acquisition unit 12, and a data determination unit 13.

[0119] The benchmark data acquisition unit 11 is used to acquire benchmark operation data of the target task in response to the operation of the target task. The benchmark operation data is determined based on the historical operation data of the target task. The benchmark operation data includes benchmark runtime and benchmark resource usage data. The benchmark resource usage data is the benchmark utilization rate of the processor and the benchmark memory usage during the operation of the target task.

[0120] The impact data acquisition unit 12 is used to acquire the difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device, and to determine the real-time load impact data of the task running device based on the difference and the first load weight. The real-time load data includes the current utilization rate of the processor and the current utilization of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical utilization of memory.

[0121] The impact data acquisition unit 12 is further configured to acquire the average load data of the task running device within a preset time period before running the target task, and determine the lag load impact data of the task running device based on the average load data and the second load weight, wherein the average load data is the average utilization rate of the processor and the average usage of memory.

[0122] The data determination unit 13 is used to determine the predicted operating data of the target task based on the baseline operating data, the real-time load impact data, and the lag load impact data.

[0123] In the embodiments of this specification, the baseline operating data of the target task is obtained during the stage when the target task is about to start running. The baseline operating data is used as the anchor point of the task operating data. The target task operating data is dynamically calibrated by combining the impact of the current load of the task running equipment on the operation of the target task and the impact of the load on the operation of the target task in the period before the target task runs, so as to obtain the final predicted operating data of the target task. This improves the accuracy of the prediction of the task operating data and ensures the accuracy of the scheduling decision.

[0124] Please see Figure 6 This is a schematic diagram of a task execution data determination device provided in an embodiment of this specification. Figure 6 As shown, the device 1 may include a baseline data acquisition unit 11, an impact data acquisition unit 12, a data determination unit 13, a historical data acquisition unit 14, a baseline duration determination unit 15, a baseline resource data determination unit 16, an adjustment factor determination unit 17, and a weight update unit 18.

[0125] The historical data acquisition unit 14 is used to acquire the historical operation data of the target task within a historical time period. The historical operation data includes a historical runtime set and a historical resource usage data set.

[0126] The baseline duration determination unit 15 is used to determine the baseline duration of the target task based on the historical runtime set;

[0127] The baseline resource data determination unit 16 is used to determine the baseline resource usage data of the target task based on the historical resource usage data set;

[0128] The benchmark data acquisition unit 11 is used to acquire benchmark operation data of the target task in response to the operation of the target task. The benchmark operation data is determined based on the historical operation data of the target task. The benchmark operation data includes benchmark runtime and benchmark resource usage data. The benchmark resource usage data is the benchmark utilization rate of the processor and the benchmark memory usage during the operation of the target task.

[0129] The adjustment factor determination unit 17 is used to obtain the running priority of the target task and determine the sensitivity adjustment factor corresponding to the target task based on the running priority and the priority weight.

[0130] The impact data acquisition unit 12 is used to acquire the difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device, and to determine the real-time load impact data of the task running device based on the difference and the first load weight. The real-time load data includes the current utilization rate of the processor and the current utilization of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical utilization of memory.

[0131] In specific implementation, the data acquisition unit 12 is used for:

[0132] Obtain the current processor usage data and current memory usage data of the task execution device in response to the target task during runtime;

[0133] Obtain a first difference between the current usage data of the processor and the historical usage baseline data of the processor, and obtain a second difference between the current usage data of the memory and the historical usage baseline data of the memory;

[0134] Based on the first difference and the first weight corresponding to the processor, and the second difference and the second weight corresponding to the memory, the real-time load impact data of the task running device is determined.

[0135] The impact data acquisition unit 12 is further configured to acquire the average load data of the task running device within a preset time period before running the target task, and determine the lag load impact data of the task running device based on the average load data and the second load weight, wherein the average load data is the average utilization rate of the processor and the average usage of memory.

[0136] In a specific implementation, the influence data acquisition unit 12 is specifically used for:

[0137] Obtain the average processor usage data and average memory usage data of the task running device during a preset time period before running the target task;

[0138] Based on the average usage data of the processor and the third weight corresponding to the processor, and the average usage data of the memory and the fourth weight corresponding to the memory, the lag load impact data of the task running device is determined.

[0139] The data determination unit 13 is used to use a prediction algorithm and, based on the baseline operating data, the real-time load impact data, and the lag load impact data, determine the predicted operating data of the target task.

[0140] In a specific implementation, the data determination unit 13 is specifically used for:

[0141] A prediction algorithm is used, and based on the baseline operating data, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data, the predicted operating data for the target task is determined.

[0142] Specifically, the data determination unit 13 is used for:

[0143] A prediction algorithm is used, and the predicted runtime of the target task is determined based on the baseline runtime, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data.

[0144] Using the aforementioned prediction algorithm, and based on the baseline resource usage data, the sensitivity adjustment factor, the real-time load impact data, and the lagged load impact data, the predicted resource usage data for the target task is determined.

[0145] The weight update unit 18 is used to update the first load weight, the second load weight, and the priority weight based on the nonlinear least squares method if the data difference between the actual running data and the predicted running data of the target task meets a preset range and the duration of the preset range is greater than or equal to a preset duration.

[0146] In the embodiments of this specification, by acquiring baseline operating data of the target task during the stage before the target task is started and running, and using the baseline operating data as the anchor point for the task operating data, and combining the impact of the current load of the task running equipment on the target task operation and the impact of the load on the target task operation in the period before the target task operation, the target task operating data is dynamically calibrated to obtain the final predicted operating data of the target task, thereby improving the accuracy of the prediction of task operating data and ensuring the accuracy of scheduling decisions. At the same time, by combining the task sensitivity adjustment factor, the priority of the task can be systematically correlated with the task's sensitivity to the load, further improving the accuracy of the prediction of task operating data. By updating the weight parameters involved in the prediction algorithm when the actual operating data of the task continuously deviates from the predicted operating data within a certain period of time, the prediction algorithm is dynamically adjusted, ensuring the effectiveness of the prediction algorithm.

[0147] It should be noted that the task operation data determination device provided in the above embodiments is only illustrated by the division of the above functional modules when executing the task operation data determination method. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the task operation data determination device and the task operation data determination method embodiments provided in the above embodiments belong to the same concept, and the implementation process is detailed in the method embodiments, which will not be repeated here.

[0148] The embodiment numbers in this specification are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0149] This specification also provides a storage medium storing a computer program, which, when executed by a processor, implements the above-described functionality. Figures 2-4 The task execution data determination method described in the illustrated embodiment can be found in the following document for a detailed execution process. Figures 2-4 The specific details of the illustrated embodiments will not be elaborated here.

[0150] This specification also provides a computer program product, which includes a computer program that, when executed by a processor of an electronic device, enables the processor to at least perform the functions described above. Figures 2-4 The method provided in the illustrated embodiment.

[0151] Please see Figure 7 This diagram illustrates the structure of an electronic device provided in an exemplary embodiment of this specification. The electronic device in this specification may include one or more components such as a processor 110, a memory 120, an input device 130, an output device 140, and a bus 150. The processor 110, memory 120, input device 130, and output device 140 may be connected via the bus 150.

[0152] Processor 110 may include one or more processing cores. Processor 110 connects to various parts of the electronic device using various interfaces and lines, and performs various functions of the terminal and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 120, and by calling data stored in memory 120. Optionally, processor 110 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). Processor 110 may integrate one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), and a modem. The CPU primarily handles the operating system, user page, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into processor 110 and may be implemented separately using a communication chip.

[0153] The memory 120 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 120 may include non-transitory computer-readable storage medium. The memory 120 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 120 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the various method embodiments described above, etc. The operating system may be the Android system, including systems deeply developed based on the Android system, the iOS system developed by Apple Inc., including systems deeply developed based on the iOS system, or other systems.

[0154] The memory 120 can be divided into operating system space and user space. The operating system runs in the operating system space, while native and third-party applications run in user space. To ensure that different third-party applications can achieve good running performance, the operating system allocates corresponding system resources for each application. However, different application scenarios within the same third-party application have different requirements for system resources. For example, in local resource loading scenarios, third-party applications have high requirements for disk read speed; in animation rendering scenarios, third-party applications have high requirements for GPU performance. Since the operating system and third-party applications are independent of each other, the operating system often cannot promptly perceive the current application scenario of a third-party application, resulting in the operating system's inability to adapt system resources accordingly.

[0155] In order for the operating system to distinguish the specific application scenarios of third-party applications, it is necessary to establish data communication between the third-party applications and the operating system. This would allow the operating system to obtain the current scenario information of the third-party applications at any time, and then perform targeted system resource adaptation based on the current scenario.

[0156] The input device 130 is used to receive input instructions or data, and includes, but is not limited to, a keyboard, mouse, camera, microphone, or touch device. The output device 140 is used to output instructions or data, and includes, but is not limited to, a display device and a speaker. In one example, the input device 130 and the output device 140 can be combined, and the input device 130 and the output device 140 can be a touch display screen.

[0157] The touch display screen can be designed as a full-screen, curved screen, or irregularly shaped screen. It can also be designed as a combination of a full-screen and a curved screen, or a combination of an irregularly shaped screen and a curved screen; however, this specification does not limit the specific design of the embodiments.

[0158] In addition, those skilled in the art will understand that the structure of the electronic device shown in the above figures does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, the electronic device may also include radio frequency circuits, input units, sensors, audio circuits, WiFi modules, power supplies, Bluetooth modules, etc., which will not be described in detail here.

[0159] exist Figure 7 In the illustrated electronic device, the processor 110 can be used to call computer applications stored in the memory 120 and specifically perform the following operations:

[0160] In response to the execution of a target task, baseline execution data of the target task is obtained. The baseline execution data is determined based on the historical execution data of the target task. The baseline execution data includes baseline runtime and baseline resource usage data. The baseline resource usage data is the baseline utilization rate of the processor and the baseline memory usage during the execution of the target task.

[0161] The difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device is obtained. Based on the difference and a first load weight, the real-time load impact data of the task running device is determined. The real-time load data includes the current utilization rate of the processor and the current utilization of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical utilization of memory.

[0162] The average load data of the task running device during a preset time period before running the target task is obtained. Based on the average load data and the second load weight, the lag load impact data of the task running device is determined. The average load data is the average utilization of the processor and the average usage of memory.

[0163] A prediction algorithm is used, and based on the baseline operating data, the real-time load impact data, and the delayed load impact data, the predicted operating data for the target task is determined.

[0164] The prediction algorithm is defined as follows: Predicted operating data = Baseline operating data × (1 + Real-time load impact data + Lagging load impact data).

[0165] In one embodiment, before executing the operation in response to the target task and acquiring baseline running data for the target task, the processor 110 further performs the following operations:

[0166] Obtain historical operation data of the target task within a historical time period, the historical operation data including a set of historical runtime and a set of historical resource usage data;

[0167] The baseline runtime of the target task is determined based on the historical runtime set;

[0168] The baseline resource usage data for the target task is determined based on the historical resource usage data set.

[0169] In one embodiment, when the processor 110 performs the following operations, it acquires the difference between the real-time load data of the task running device in response to the target task and the historical load baseline data of the task running device, and determines the real-time load impact data of the task running device based on the difference and a first load weight:

[0170] Obtain the current processor usage data and current memory usage data of the task execution device in response to the target task during runtime;

[0171] Obtain a first difference between the current usage data of the processor and the historical usage baseline data of the processor, and obtain a second difference between the current usage data of the memory and the historical usage baseline data of the memory;

[0172] Based on the first difference and the first weight corresponding to the processor, and the second difference and the second weight corresponding to the memory, the real-time load impact data of the task running device is determined.

[0173] In one embodiment, when the processor 110 performs the following operations, it acquires the average load data of the task running device during a preset time period before running the target task, and determines the lag load impact data of the task running device based on the average load data and a second load weight:

[0174] Obtain the average processor usage data and average memory usage data of the task running device during a preset time period before running the target task;

[0175] Based on the average usage data of the processor and the third weight corresponding to the processor, and the average usage data of the memory and the fourth weight corresponding to the memory, the lag load impact data of the task running device is determined.

[0176] In one embodiment, before executing a prediction algorithm and determining the predicted runtime data for the target task based on the baseline runtime data, the real-time load impact data, and the lag load impact data, the processor 110 further performs the following operations:

[0177] Obtain the running priority of the target task, and determine the sensitivity adjustment factor corresponding to the target task based on the running priority and priority weight;

[0178] The step of employing a prediction algorithm and determining the predicted operational data for the target task based on the baseline operational data, the real-time load impact data, and the lagged load impact data includes:

[0179] Based on the baseline operating data, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data, the predicted operating data for the target task is determined.

[0180] In one embodiment, when the processor 110 executes a prediction algorithm and determines the predicted running data of the target task based on the baseline running data, the sensitivity adjustment factor, the real-time load impact data, and the hysteresis load impact data, it specifically performs the following operations:

[0181] A prediction algorithm is used, and the predicted runtime of the target task is determined based on the baseline runtime, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data.

[0182] Using the aforementioned prediction algorithm, and based on the baseline resource usage data, the sensitivity adjustment factor, the real-time load impact data, and the lagged load impact data, the predicted resource usage data for the target task is determined.

[0183] In one embodiment, the processor 110 also performs the following operations:

[0184] If the difference between the actual running data and the predicted running data of the target task meets a preset range, and the duration of the preset range is greater than or equal to a preset duration, then the first load weight, the second load weight, and the priority weight are updated based on the nonlinear least squares method.

[0185] In the embodiments of this specification, by acquiring baseline operating data of the target task during the stage before the target task is started and running, and using the baseline operating data as the anchor point for the task operating data, and combining the impact of the current load of the task running equipment on the target task operation and the impact of the load on the target task operation in the period before the target task operation, the target task operating data is dynamically calibrated to obtain the final predicted operating data of the target task, thereby improving the accuracy of the prediction of task operating data and ensuring the accuracy of scheduling decisions. At the same time, by combining the task sensitivity adjustment factor, the priority of the task can be systematically correlated with the task's sensitivity to the load, further improving the accuracy of the prediction of task operating data. By updating the weight parameters involved in the prediction algorithm when the actual operating data of the task continuously deviates from the predicted operating data within a certain period of time, the prediction algorithm is dynamically adjusted, ensuring the effectiveness of the prediction algorithm.

[0186] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0187] The above-disclosed embodiments are merely preferred embodiments of this specification and should not be construed as limiting the scope of this specification. Therefore, any equivalent variations made in accordance with the claims of this specification shall still fall within the scope of this specification.

Claims

1. A method for determining task execution data, comprising: In response to the execution of a target task, baseline execution data of the target task is obtained. The baseline execution data is determined based on the historical execution data of the target task. The baseline execution data includes baseline runtime and baseline resource usage data. The baseline resource usage data is the baseline utilization rate of the processor and the baseline memory usage during the execution of the target task. The difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device is obtained. Based on the difference and a first load weight, the real-time load impact data of the task running device is determined. The real-time load data includes the current utilization rate of the processor and the current usage of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical usage of memory. The average load data of the task running device during a preset time period before running the target task is obtained. Based on the average load data and the second load weight, the lag load impact data of the task running device is determined. The average load data is the average utilization of the processor and the average usage of memory. A prediction algorithm is used, and based on the baseline operating data, the real-time load impact data, and the delayed load impact data, the predicted operating data for the target task is determined. The prediction algorithm is defined as follows: Predicted operating data = Baseline operating data × (1 + Real-time load impact data + Lagging load impact data).

2. The method according to claim 1, further comprising, before acquiring the baseline operating data of the target task in response to the operation of the target task: Obtain historical operation data of the target task within a historical time period, the historical operation data including a set of historical runtime and a set of historical resource usage data; The baseline runtime of the target task is determined based on the historical runtime set; The baseline resource usage data for the target task is determined based on the historical resource usage data set.

3. The method according to claim 1, wherein obtaining the difference between the real-time load data of the task running device in response to the target task and the historical load baseline data of the task running device, and determining the real-time load impact data of the task running device based on the difference and a first load weight, comprises: Obtain the current processor usage data and current memory usage data of the task execution device in response to the target task during runtime; Obtain a first difference between the current usage data of the processor and the historical usage baseline data of the processor, and obtain a second difference between the current usage data of the memory and the historical usage baseline data of the memory; Based on the first difference and the first weight corresponding to the processor, and the second difference and the second weight corresponding to the memory, the real-time load impact data of the task running device is determined.

4. The method according to claim 1, wherein obtaining the average load data of the task running device within a preset time period before running the target task, and determining the lag load impact data of the task running device based on the average load data and the second load weight, comprises: Obtain the average processor usage data and average memory usage data of the task running device during a preset time period before running the target task; Based on the average usage data of the processor and the third weight corresponding to the processor, and the average usage data of the memory and the fourth weight corresponding to the memory, the lag load impact data of the task running device is determined.

5. The method according to claim 1, further comprising, before employing a prediction algorithm and determining the predicted operating data of the target task based on the baseline operating data, the real-time load impact data, and the lagged load impact data: Obtain the running priority of the target task, and determine the sensitivity adjustment factor corresponding to the target task based on the running priority and priority weight; The step of employing a prediction algorithm and determining the predicted operational data for the target task based on the baseline operational data, the real-time load impact data, and the lagged load impact data includes: A prediction algorithm is used, and based on the baseline operating data, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data, the predicted operating data for the target task is determined.

6. The method according to claim 5, wherein the step of employing a prediction algorithm and determining the predicted operating data of the target task based on the baseline operating data, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data includes: A prediction algorithm is used, and the predicted runtime of the target task is determined based on the baseline runtime, the sensitivity adjustment factor, the real-time load impact data, and the lag load impact data. Using the aforementioned prediction algorithm, and based on the baseline resource usage data, the sensitivity adjustment factor, the real-time load impact data, and the lagged load impact data, the predicted resource usage data for the target task is determined.

7. The method according to claim 5, further comprising: If the difference between the actual running data and the predicted running data of the target task meets a preset range, and the duration of the preset range is greater than or equal to a preset duration, then the first load weight, the second load weight, and the priority weight are updated based on the nonlinear least squares method.

8. A task execution data determination device, comprising: A benchmark data acquisition unit is used to acquire benchmark operation data of the target task in response to the operation of the target task. The benchmark operation data is determined based on the historical operation data of the target task. The benchmark operation data includes benchmark runtime and benchmark resource usage data. The benchmark resource usage data is the benchmark utilization rate of the processor and the benchmark memory usage during the operation of the target task. The impact data acquisition unit is used to acquire the difference between the real-time load data of the task running device in response to the target task and the historical load benchmark data of the task running device, and to determine the real-time load impact data of the task running device based on the difference and a first load weight. The real-time load data includes the current utilization rate of the processor and the current utilization of memory. The historical load benchmark data is determined based on the historical load data of the task running device in multiple historical periods. The historical load data includes the historical utilization rate of the processor and the historical utilization of memory. The impact data acquisition unit is further configured to acquire the average load data of the task running device within a preset time period before running the target task, and determine the lag load impact data of the task running device based on the average load data and the second load weight, wherein the average load data is the average utilization rate of the processor and the average usage of memory. The data determination unit is used to use a prediction algorithm and, based on the baseline operating data, the real-time load impact data, and the lag load impact data, determine the predicted operating data of the target task. The prediction algorithm is defined as follows: Predicted operating data = Baseline operating data × (1 + Real-time load impact data + Lagging load impact data).

9. An electronic device, comprising: Processor and memory; The memory stores a computer program adapted to be loaded by the processor and to execute the steps of the method as described in any one of claims 1 to 7.

10. A storage medium storing a computer program that, when executed by a processor, implements the steps of the method as claimed in any one of claims 1 to 7.

11. A computer program product, comprising: A computer program, when executed by a processor of an electronic device, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

Images