A resource scheduling method, a terminal device, and a storage medium

Abstract

Embodiments of the present application disclose a resource scheduling method, a terminal device and a storage medium. The resource scheduling method comprises: obtaining current application scenario information; selecting a corresponding user perception model according to the current application scenario information; the user perception model represents a corresponding relationship between a user perception score and a time delay; determining a pre-configured reference point corresponding to the user perception model; and performing model conversion on the user perception model based on the pre-configured reference point to obtain a performance factor and time delay corresponding relationship; determining a real-time performance factor corresponding to a real-time time delay based on the performance factor and time delay corresponding relationship, so as to adjust the performance capability of a frequency adjustment scheduling mechanism by using the real-time performance factor when the real-time time delay arrives.

Description

Technical Field

[0001] This invention relates to the field of resource scheduling, and more particularly to a resource scheduling method, terminal device, and storage medium. Background Technology

[0002] Currently, the technical solution for resource scheduling involves collecting historical resource scheduling data for application scenarios and allocating the necessary resources to the corresponding application scenarios based on this historical data. However, this existing resource scheduling method is prone to problems. When resource demand increases, the frequency modulation scheduling mechanism cannot promptly allocate the corresponding resources, leading to a deterioration in the performance of the current application scenario. Conversely, when resource demand decreases, the frequency modulation scheduling mechanism cannot promptly release resources, resulting in increased power consumption of terminal devices. Summary of the Invention

[0003] In view of this, the embodiments of this application aim to provide a resource scheduling method, a terminal device, and a storage medium that can meet the different performance and power consumption requirements of current application scenarios.

[0004] To achieve the above objectives, the technical solution of this application is implemented as follows:

[0005] In a first aspect, embodiments of this application provide a resource scheduling method, the method comprising:

[0006] Obtain current application scenario information; and select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency;

[0007] Determine the pre-configured baseline point corresponding to the user perception model; and based on the pre-configured baseline point, perform model transformation on the user perception model to obtain the correspondence between performance factors and latency.

[0008] Based on the correspondence between performance factors and latency, the real-time performance factors corresponding to the real-time latency are determined so that when the real-time latency arrives, the performance capabilities of the frequency modulation scheduling mechanism can be adjusted using the real-time performance factors.

[0009] Secondly, embodiments of this application provide a terminal device, the terminal device comprising:

[0010] The application scenario recognition module is used to obtain information about the current application scenario;

[0011] The user perception model selection module is used to select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency.

[0012] A parameter pre-definition or update module is used to determine the pre-configured benchmark point corresponding to the user perception model;

[0013] The model conversion module is used to perform downsampling and quantization processing on the user perception model based on the pre-configured benchmark points to obtain a set of reference points corresponding to the user perception model, and to perform calculations on the set of reference points to obtain the relationship between performance factors and latency.

[0014] The real-time delay counting module is used to determine the real-time delay.

[0015] The performance factor determination module is used to perform model transformation on the user perception model to obtain the correspondence between performance factors and latency; and to determine the real-time performance factors corresponding to the real-time latency based on the correspondence between performance factors and latency.

[0016] The kernel scheduler module is used to adjust the performance capability of the frequency modulation scheduling mechanism based on the real-time performance factors when the real-time delay arrives.

[0017] Thirdly, embodiments of this application provide a terminal device, the terminal device including: a processor, a memory, and a communication bus; the processor implements the above-mentioned resource scheduling method when executing a running program stored in the memory.

[0018] Fourthly, embodiments of this application provide a storage medium on which a computer program is stored, which, when executed by a processor, implements the above-described resource scheduling method.

[0019] This application provides a resource scheduling method, terminal device, and storage medium. The method includes: acquiring current application scenario information; selecting a corresponding user perception model based on the current application scenario information; the user perception model representing the correspondence between user perception score and latency; determining a pre-configured benchmark point corresponding to the user perception model; and performing model transformation on the user perception model based on the pre-configured benchmark point to obtain a correspondence between performance factors and latency; and determining a real-time performance factor corresponding to the real-time latency based on the correspondence between performance factors and latency, so as to adjust the performance capability of the frequency modulation scheduling mechanism using the real-time performance factor when the real-time latency arrives. By adopting the above implementation scheme, when the frequency modulation scheduling mechanism schedules resources, a user perception model is added. Based on the user perception model, the performance factors corresponding to each latency are determined for the user perception score in the user perception model, and the correspondence between performance factors and latency is obtained. In latency-related scenarios, the performance factors corresponding to user perception can be considered, and the speed of the frequency modulation scheduling mechanism can be adjusted in a timely manner, thereby changing the resource scheduling scheme. As a result, when the resources required by the current application scenario increase, the frequency modulation scheduling mechanism can schedule the corresponding resources in a timely manner to meet the performance requirements of the current application scenario; when the resources required by the current application scenario decrease, resources can be released in a timely manner to reduce unnecessary power consumption. Attached Figure Description

[0020] Figure 1 This is an example illustrating the correspondence between user-perceived ratings and latency. Figure 1 ;

[0021] Figure 2 This is an example of the relationship between performance factors and latency. Figure 1 ;

[0022] Figure 3 A flowchart of a resource scheduling method provided in this application embodiment Figure 1 ;

[0023] Figure 4 This application provides an exemplary illustration of the correspondence between user perception rating and latency. Figure 2 ;

[0024] Figure 5 A schematic diagram of a set of reference points obtained after downsampling an exemplary user perception model provided in this application embodiment;

[0025] Figure 6 This application provides an exemplary embodiment of the result showing the relationship between performance factor and latency after performing calculations on a set of reference points obtained through downsampling. Figure 2 ;

[0026] Figure 7 A flowchart illustrating an exemplary adjustment of the relationship between initial performance factors and latency, provided as an embodiment of this application;

[0027] Figure 8 This application provides an example of the relationship between a performance target point and a pre-configured reference point. Figure 1 ;

[0028] Figure 9 A schematic diagram illustrating an exemplary adjusted relationship between performance factor and latency, provided as an embodiment of this application. Figure 1 ;

[0029] Figure 10 This application provides an example of the relationship between a performance target point and a pre-configured reference point. Figure 2 ;

[0030] Figure 11 This application provides an exemplary illustration of the adjusted performance factor and latency relationship in an embodiment. Figure 2 ;

[0031] Figure 12 Flowchart of another resource scheduling method provided in the embodiments of this application Figure 2 ;

[0032] Figure 13 A schematic diagram of an exemplary terminal device 1 provided in this application embodiment. Figure 1 ;

[0033] Figure 14 A schematic diagram of the structure of a terminal device 1 provided in this application embodiment. Figure 2 ;

[0034] Figure 15 A schematic diagram of the structure of a terminal device 1 provided in this application embodiment. Figure 3 . Detailed Implementation

[0035] In order to gain a more detailed understanding of the features and technical content of the embodiments of this application, the technical solution of this application will be further described in detail below with reference to the accompanying drawings and specific embodiments. The accompanying drawings are for reference only and are not intended to limit the embodiments of this application.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to be limiting of this application.

[0037] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0038] Currently, in existing technologies, the resources allocated to application scenarios by frequency modulation scheduling mechanisms are fixed. In some application scenarios, if the resource requirements of the current application scenario increase, the frequency modulation scheduling mechanism cannot provide additional performance capabilities, resulting in a deterioration in the performance of the current application scenario. In other application scenarios, if the resource requirements of the current application scenario decrease, the frequency modulation scheduling mechanism cannot reduce the original performance capabilities, resulting in higher power consumption of the terminal device.

[0039] Regarding the aforementioned existing technologies, in latency-related scenarios, the existing solutions do not consider the user's subjective perception factors. There is no direct correlation between the user's perceived latency rating and the latency itself. Therefore, in the relationship between perceived latency and latency, the perceived latency rating does not change with increasing latency. Figure 1The relationship between user perception score and latency can be represented by a horizontal line. Under different latency levels, there is no correlation between latency and the corresponding user perception score; the user perception score does not change with latency. Correspondingly, since the user perception score does not change with latency, the relationship between performance factors and latency in latency-related scenarios, where there is no correlation between user perception score and latency, also means that performance factors will not change with increasing latency. Figure 2 The relationship between performance factors and latency is represented by a horizontal line. Therefore, when performing resource scheduling, the frequency modulation scheduling mechanism cannot provide weights or guidance for increasing or decreasing performance capabilities. In this case, if the resource requirements of the current application scenario increase, the frequency modulation scheduling mechanism cannot provide additional performance capabilities and cannot meet the resource requirements of the current application scenario. Conversely, if the resource requirements of the current application scenario are small, the frequency modulation scheduling mechanism cannot reduce the original performance capabilities, resulting in excessive resource utilization and increased power consumption of the terminal device.

[0040] To address the aforementioned problems, embodiments of this application provide a resource scheduling method applied to terminal devices, such as... Figure 3 As shown in the embodiments of this application, the resource scheduling method mainly includes the following steps:

[0041] S101. Obtain the current application scenario information; and select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency.

[0042] This application provides a resource scheduling method applied to a terminal device. The specific type of terminal device is not limited in this application; it can be any user device, such as a smartphone, personal computer (PC), laptop, tablet, or portable wearable device—any device with communication and storage functions.

[0043] In this embodiment of the application, when the terminal device executes an application, the terminal device will automatically detect the current application scenario information corresponding to the application.

[0044] In this embodiment, the current application scenario information on the terminal device may include latency-related scenario information such as application cold start scenario, application warm start scenario, and screen projection scenario. Specifically, it can be selected according to the actual situation, and no specific limitation is made in this embodiment.

[0045] For example, when a terminal device executes a calculator application, if the terminal device detects that the calculator application process does not exist in the current process, it indicates that the calculator application is being launched for the first time. This application launch method can be understood as a cold start, meaning that the current application scenario information corresponding to the calculator application is a cold start scenario. If the terminal device detects that the calculator application process exists in the current process, the terminal device can directly load the calculator application when it executes the calculator application again. This application launch method can be understood as a warm start, meaning that the current application scenario information corresponding to the current calculator application is a warm start scenario.

[0046] In this embodiment of the application, the terminal device stores multiple continuous curves representing the user perception model. The continuous curves in the terminal device can identify the correspondence between user perception score and latency in the user perception model.

[0047] It should be noted that the user perception model stored in the terminal device can be the user perception model preset when the terminal device leaves the factory, or it can be an updated or newly added user perception model pushed by the system of the terminal device.

[0048] In the embodiments of this application, such as Figure 4 As shown, the user perception model is the correspondence between user perception score and latency in the current application scenario. To make it easier to explain this application, in this embodiment, the user perception model can be divided into three stages: the first stage, the second stage, and the third stage.

[0049] Phase 1: During this phase, user perception scores remain relatively stable as latency increases.

[0050] Phase Two: In this phase, the intersection of the average user perception score and the user perception model is used as the dividing line. In the first half of the second phase, the change in user perception score will gradually intensify as the time delay increases. In the second half of the second phase, the change in user perception score will gradually slow down as the time delay increases.

[0051] Phase 3: During this phase, user perception scores remain relatively stable as latency increases.

[0052] It should be noted that the above is only an illustrative representation of the changes in user perception ratings. In actual terminal devices, the user perception model does not have these three stages.

[0053] In this embodiment of the application, after the terminal device obtains the current application scenario information, it searches among multiple user perception models in the terminal device to find a user perception model that matches the current application scenario information.

[0054] S102. Determine the pre-configured baseline point corresponding to the user perception model; and based on the pre-configured baseline point, perform model transformation on the user perception model to obtain the correspondence between performance factors and latency.

[0055] In this embodiment of the application, after obtaining the user perception model corresponding to the current application scenario information, a pre-configured benchmark point is first determined from the obtained user perception model. The pre-configured benchmark point is used to perform model transformation on the user perception model to obtain the correspondence between performance factors and latency.

[0056] In this embodiment, since the user perception model represents the correspondence between user perception scores and latency, different latency levels correspond to different user perception scores. During the determination of the pre-configured benchmark point, the average value of all user perception scores in the user perception model is first calculated. Then, the point corresponding to the average user perception score is found on the curve corresponding to the user perception model. This point is the pre-configured benchmark point of the user perception model corresponding to the current application scenario information. Figure 4 The pre-configured baseline is shown in the figure.

[0057] It should be noted that, in the embodiments of this application, the method of determining the pre-configuration benchmark point is not limited to the method of determining the pre-configuration benchmark point by the average value of the user perception score. Other methods of determining the pre-configuration benchmark point from the user perception model are all within the protection scope of this application. The specific method can be selected according to the actual situation, and no specific limitation is made in this application.

[0058] In this embodiment of the application, after obtaining the pre-configured benchmark point corresponding to the user perception model, it is necessary to perform model transformation on the user perception model based on the pre-configured benchmark point to obtain the correspondence between performance factors and latency.

[0059] Specifically, the model conversion process first obtains the screen refresh rate or user demand parameters of the terminal device, and then downsamples and quantizes the user perception model based on the screen refresh rate or user demand parameters to obtain a set of reference points corresponding to the user perception model.

[0060] In the embodiments of this application, downsampling quantization is the process of reducing the sampling rate of a specific signal. It is usually used to reduce the data transmission rate or data size. The downsampling factor (commonly represented by the symbol M) is generally an integer or rational number greater than 1. The downsampling factor expresses how many times the original sampling period has become, or equivalently, how many times the original sampling rate has become.

[0061] In this embodiment, since the screen refresh rate is quantizable, when downsampling and quantizing the user perception model, the screen refresh rate of the terminal device can be obtained, and the user perception model can be downsampled and quantized using the screen refresh rate of the terminal device to convert the user perception model into a corresponding set of reference points. Alternatively, the user perception model can be downsampled and quantized into a corresponding set of reference points using the obtained user demand parameters. The display format of the set of reference points can be referenced. Figure 5 As shown, Figure 5 The set of reference points shown includes pre-configured baseline reference points.

[0062] It should be noted that downsampling and quantization of the user perception model is not limited to the downsampling and quantization method based on screen refresh rate or user demand parameters as described in this application. Other methods of downsampling and quantization to obtain a set of reference points are also within the scope of protection of this application. No specific limitation is made in this application, and the method can be selected according to the actual situation.

[0063] After obtaining a set of reference points through the above downsampling and quantization processing, the pre-configured reference points are used as a benchmark to perform calculations on this set of reference points to obtain the correspondence between performance factors and latency.

[0064] It should be noted that, in the embodiments of this application, the operation on a set of reference points may include: performing differential mirroring operation, table lookup operation, and / or quantization operation on a set of reference points. Specifically, it may be performing differential mirroring followed by quantization operation, or performing differential mirroring operation after quantization, or performing only differential mirroring operation. The method of performing the operation on a set of reference points obtained after downsampling and quantization can be selected according to the actual situation, and the embodiments of this application do not impose specific limitations.

[0065] In the embodiments of this application, reference is made to Figure 6 , for the basis Figure 5 The mapping between user perception scores and latency is used to obtain the mapping between performance factors and latency. The value corresponding to the performance factor is a coefficient multiplied by the performance capability when the frequency modulation scheduling mechanism schedules resources. When the performance factor value is 1, the frequency modulation speed is multiplied by the coefficient 1, indicating that the performance capability of the current frequency modulation scheduling mechanism does not need to be adjusted when scheduling resources. When the value corresponding to the performance factor is not 1, the frequency modulation speed is multiplied by the coefficient that is not 1. When the current frequency modulation scheduling mechanism schedules resources, its performance capability changes according to the change of the current coefficient.

[0066] Figure 6 The performance factor and latency relationship shown is based on... Figure 5 The average user perception score in the data is the baseline. Figure 5The calculations are performed using a set of reference points to obtain the relationship between performance factors and latency. Figure 5 Since the reference point changes gradually in the first and third stages, it can be concluded that no additional performance capability of adjusting the frequency modulation scheduling mechanism is required in the first and third stages. Therefore, in Figure 5 The reference points for the first and third stages in the calculation correspond to Figure 6 The performance factor is 1; due to the drastic changes in the reference point in the second stage, it can be concluded that the frequency modulation speed of the frequency modulation scheduling mechanism needs to be adjusted in a timely manner in the second stage. Therefore, in Figure 5 The reference point in the second stage corresponds to the calculation. Figure 6 If the performance factor is greater than 1, it indicates that the frequency modulation scheduling mechanism needs to accelerate the adjustment of the frequency modulation scheduling speed according to the corresponding performance factor.

[0067] Furthermore, in another embodiment of this application, after obtaining the pre-configured reference point, the user perception model can be adjusted based on the obtained pre-configured reference point to obtain the adjusted user perception model.

[0068] In one optional embodiment, the user perception model can be adjusted based on the acquired pre-configured baseline, according to the latency sensitivity parameters, product update parameters, software update parameters and / or device performance update parameters of the target user identified by the terminal device, to obtain an adjusted user perception model.

[0069] Specifically, in the embodiments of this application, adjusting the user perception model based on a pre-configured benchmark can be achieved by increasing or decreasing the user perception score corresponding to each delay in the obtained user perception model, based on the product update parameters or received device performance requirement parameters and using the pre-configured benchmark as a reference, thereby obtaining a user perception model adjusted based on the benchmark.

[0070] For example, with Figure 4 The user perception model shown is used as a baseline. When the pre-configured baseline is moved upward based on product update parameters or received device performance requirement parameters, the user perception score corresponding to each latency in the user perception model is increased, thus obtaining a result based on... Figure 4 The user perception model moves upwards; when the pre-configured baseline is adjusted downwards based on product update parameters or received device performance requirement parameters, the user perception score corresponding to each latency in the user perception model is reduced, thus obtaining a user perception score based on... Figure 4 A user perception model that moves downwards.

[0071] In another alternative embodiment, adjusting the user perception model based on a pre-configured benchmark can also be achieved by increasing or decreasing the latency corresponding to each user perception score in the acquired user perception model, based on the parameters updated by the software or the received device performance requirement parameters, using the pre-configured benchmark as a reference, thereby obtaining a user perception model adjusted based on the benchmark.

[0072] For example, with Figure 4 Using the user perception model shown as a baseline, when the pre-configured baseline is shifted to the left based on software update parameters or received device performance requirement parameters, the latency corresponding to each user perception score in the user perception model is reduced, thus obtaining a result based on... Figure 4 The user perception model shifts to the left; when the pre-configured baseline is adjusted to the right based on software update parameters or received device performance requirement parameters, the latency corresponding to each user perception score in the user perception model is increased, thus obtaining a model based on... Figure 4 User perception model for moving to the right.

[0073] In another alternative embodiment, adjusting the user perception model based on a pre-configured reference point can also be achieved by scaling up or down the acquired user perception model based on the latency sensitivity parameters of the target user identified by the terminal device or the received device performance requirement parameters, using the pre-configured reference point as a benchmark, to obtain a user perception model adjusted based on the reference point.

[0074] It should be noted that adjusting the user perception model based on a pre-configured baseline is not limited to adjusting the user perception model based on the latency sensitivity parameters, product update parameters, software update parameters, and / or device performance update parameters of the target user identified by the terminal device in this application. Specifically, the adjustment can be made according to the actual situation, and no specific limitation is made in this application.

[0075] In this embodiment of the application, after adjusting the user perception model based on the latency sensitivity parameters, product update parameters, software update parameters and / or device performance update parameters of the target user identified by the terminal device, and obtaining the adjusted user perception model, it is necessary to perform model transformation based on the pre-configured benchmark points in the adjusted user perception model.

[0076] It should be noted that the model transformation process for the adjusted user perception model is described in step S102, and will not be repeated here.

[0077] S103. Based on the relationship between performance factors and time delay, determine the real-time performance factors corresponding to the real-time time delay, so as to adjust the performance capability of the frequency modulation scheduling mechanism when the real-time time delay arrives.

[0078] In this embodiment of the application, after obtaining the correspondence between performance factors and latency according to the above implementation method, the terminal device can determine the real-time performance factor corresponding to each real-time latency according to the correspondence between performance factors and latency. When the real-time latency arrives, the terminal can adjust the frequency modulation speed of the frequency modulation scheduling mechanism according to the performance factor corresponding to the real-time latency.

[0079] In this embodiment, the frequency modulation speed of the frequency modulation scheduling mechanism is adjusted by multiplying the performance factor with the frequency modulation speed in the frequency modulation scheduling mechanism.

[0080] For example, the performance factor can correspond to a number between 0 and 1. This indicates that the frequency modulation scheduling mechanism needs to slow down the frequency modulation speed according to the value corresponding to the performance factor. Suppose that the performance factor corresponding to the real-time delay is 0.3. In this case, when the frequency modulation scheduling mechanism adjusts the frequency modulation speed, it multiplies the original frequency modulation speed by a coefficient of 0.3, which is equivalent to slowing down the frequency modulation speed based on the original frequency modulation speed. This indicates that the system does not need excessive resources in the current application scenario, and unnecessary resource waste can be reduced.

[0081] For example, the performance factor can correspond to a number greater than 1. In this case, it indicates that the frequency modulation scheduling mechanism needs to speed up the frequency modulation speed according to the value corresponding to the performance factor. Suppose that the performance factor corresponding to the real-time delay is 2. In this case, when the frequency modulation scheduling mechanism adjusts the frequency modulation speed, it multiplies the original frequency modulation speed by a coefficient of 2, which is equivalent to speeding up the frequency modulation speed on the basis of the original frequency modulation speed. This indicates that the system needs to invest more resources to meet the current performance requirements in the current application scenario. At this time, the frequency modulation scheduling mechanism can quickly invest the corresponding resources by using twice the speed of the original frequency modulation speed to improve the performance of the terminal device.

[0082] It is understood that in the resource scheduling method provided in this application embodiment, when the frequency modulation scheduling mechanism schedules resources, a user perception model is added. Based on the user perception model, the performance factors corresponding to each latency are determined for the user perception score in the user perception model, and the correspondence between performance factors and latency is obtained. In latency-related scenarios, the performance factors corresponding to user perception can be considered, and the speed of the frequency modulation scheduling mechanism can be adjusted in a timely manner, thereby changing the resource scheduling scheme. Thus, when the resources required by the current application scenario increase, the frequency modulation scheduling mechanism can schedule the corresponding resources in a timely manner to meet the performance requirements of the current application scenario; when the resources required by the current application scenario decrease, resources can be released in a timely manner to reduce unnecessary power consumption.

[0083] Based on the above embodiments, in another embodiment of this application, after obtaining the screen refresh rate or user demand parameters of the terminal device, and downsampling and quantizing the user perception model according to the screen refresh rate or user demand parameters to obtain a set of reference points corresponding to the user perception model; before performing calculations on a set of reference points based on the pre-configured benchmark points to obtain the performance factor and latency correspondence, that is, in step S102, a performance target point can also be selected from a set of reference points obtained by downsampling the user perception model.

[0084] In this embodiment of the application, the performance target point is selected by choosing one of the reference points from a set of reference points that are downsampled and quantized from the user perception model. Specifically, when selecting the performance target point, it is first necessary to identify the current usage status of the terminal device and select the performance target point from the set of reference points based on the usage status of the terminal device.

[0085] It should be noted that the usage status of a terminal device may include one or more of the following: the current usage mode of the terminal device, the usage time distribution within a preset time period, and the usage frequency of a preset usage scenario. Specifically, the appropriate option can be selected based on the actual situation, and this application embodiment does not impose specific limitations on the usage status of the terminal.

[0086] In this application embodiment, for example, the current usage mode of the terminal device can be a power saving mode or other application modes; the usage time distribution within a preset time period can be that the application usage frequency is higher during the day and the terminal device usage frequency is lower at night; the usage frequency of the preset usage scenario can be that the gaming scenario has a higher usage frequency and the photography scenario has a lower usage frequency, etc., which can be selected according to the actual situation. In this application, the specific form of the terminal device usage state is not specifically limited.

[0087] Based on the above, the performance target point is selected. Figure 7 This is a schematic diagram illustrating the adjustment of the correspondence between initial performance factors and latency based on a performance target point, as provided in an embodiment of this application. Figure 7 As shown in this embodiment, after obtaining the correspondence between the initial performance factors and latency, the correspondence between the initial performance factors and latency can be adjusted according to the performance target point. Specifically, the following steps can be performed:

[0088] S201. Determine the user perception score comparison results between the performance target point and the pre-configured baseline point.

[0089] In this embodiment of the application, the difference between the user perception score corresponding to the performance target point and the pre-configured benchmark point is determined as the user perception score comparison result. The user perception score comparison result can indicate the magnitude relationship between the user perception scores corresponding to the performance target point and the pre-configured benchmark point, as well as the degree of magnitude between the two.

[0090] In the embodiments of this application, exemplarily, such as Figure 8 As shown, in the user perception score comparison results, the perceived score corresponding to the performance target point is greater than the user perception score corresponding to the pre-configured baseline point. The magnitude of the difference between the two is the perceived score corresponding to the performance target point minus the user perception score corresponding to the pre-configured baseline point; for example... Figure 9 As shown, in the comparison results of user perception scores, the perception score corresponding to the performance target point is less than the user perception score corresponding to the pre-configured baseline point. The magnitude of the difference between the two is the user perception score corresponding to the pre-configured baseline point minus the perception score corresponding to the performance target point.

[0091] S202. Based on the user perception rating comparison results, determine the adjustment degree value, and increase or decrease the adjustment degree value of the initial performance factor corresponding to each delay in the initial performance factor and delay correspondence relationship to obtain the performance factor and delay correspondence relationship.

[0092] In this embodiment of the application, based on the above user perception score comparison results, the value of the initial performance factor corresponding to each delay can be determined in the correspondence between the initial performance factor and the delay based on the size relationship between the user perception score between the performance target point and the pre-configured benchmark point. The degree of increase or decrease of the value of the initial performance factor corresponding to each delay can be determined based on the size relationship between the two.

[0093] Assuming the perceived score corresponding to the performance target point is greater than the user perceived score corresponding to the pre-configured baseline point, then the value of the initial performance factor corresponding to each latency in the initial performance factor-latency correspondence can be increased by a corresponding adjustment value. Conversely, if the perceived score corresponding to the performance target point is less than the user perceived score corresponding to the pre-configured baseline point, then the value of the initial performance factor corresponding to each latency in the initial performance factor-latency correspondence can be decreased by a corresponding adjustment value. Using the above implementation method, the adjusted performance factor-latency correspondence can be obtained.

[0094] Exemplarily, in the embodiments of this application, such as Figure 8 , Figure 10 As shown, Figure 8The image shows a set of reference points obtained after downsampling and quantization. This set of reference points includes a pre-configured baseline point and a performance target point. By determining the relationship between the pre-configured baseline point and the performance target point, it can be determined that the user perception score corresponding to the performance target point is greater than the user perception score corresponding to the pre-configured baseline point. After the user perception score corresponding to the performance target point is mirrored by difference, the corresponding performance factor is obtained. Then, the performance influencing factor corresponding to the average user perception score is obtained. The difference between the performance influencing factor corresponding to the performance target point and the performance influencing factor corresponding to the average user perception score is used as the adjustment degree value. The performance factor corresponding to each latency is increased by the magnitude of the adjustment degree value, resulting in the adjusted performance factor and latency correspondence. The adjusted performance factor and latency correspondence is shifted upwards overall compared to the initial performance factor and latency correspondence. Figure 10 As shown, Figure 10 This indicates the relationship between the adjusted performance factor and latency.

[0095] Exemplarily, in the embodiments of this application, such as Figure 9 , Figure 11 As shown, Figure 9 The image shows a set of reference points obtained after downsampling and quantization. This set of reference points includes a pre-configured baseline point and a performance target point. By determining the relationship between the pre-configured baseline point and the performance target point, it can be determined that the user perception score corresponding to the performance target point is less than the user perception score corresponding to the pre-configured baseline point. After the user perception score corresponding to the performance target point is mirrored by difference, the corresponding performance factor is obtained. Then, the performance influencing factor corresponding to the average user perception score is obtained. The difference between the performance influencing factor corresponding to the average user perception score and the performance influencing factor corresponding to the performance target point is used as the adjustment degree value. The performance factor corresponding to each latency is reduced by the magnitude of the adjustment degree value, resulting in the adjusted performance factor and latency correspondence. The adjusted performance factor and latency correspondence is shifted downwards overall compared to the initial performance factor and latency correspondence. Figure 11 As shown, Figure 11 This indicates the relationship between the adjusted performance factor and latency.

[0096] Based on the above embodiments, a resource scheduling method is provided in this application, such as... Figure 12 As shown, the specific steps include:

[0097] Step 1: Obtain the current application scenario information and select the corresponding user perception model based on the current application scenario information;

[0098] Step 2: Determine the pre-configured baseline point corresponding to the user perception model;

[0099] Step 3: Based on the pre-configured baseline, adjust the user perception model according to the latency sensitivity parameters, product update parameters, software update parameters and / or device performance requirement parameters of the target user identified by the terminal device to obtain the adjusted user perception model.

[0100] Step 4: Perform downsampling and quantization processing on the adjusted user perception model based on the screen refresh rate of the terminal device or user demand parameters to obtain a set of reference points corresponding to the adjusted user perception model.

[0101] Step 5: Identify the usage status of the terminal device and determine the performance target point from the above set of reference points based on the usage status;

[0102] Step 6: Using the pre-configured reference point as a reference, perform differential mirroring operation on the above set of reference points to obtain the correspondence between the initial performance factor and the latency.

[0103] Step 7: Determine the comparison result between the user perception score corresponding to the performance target point and the user perception score corresponding to the pre-configured baseline point;

[0104] Step 8: Based on the above comparison results, determine the adjustment degree value, and increase or decrease the adjustment degree value of the initial performance factor corresponding to each delay in the initial performance factor and delay correspondence to obtain the adjusted performance factor and delay correspondence.

[0105] Step 9: When the real-time delay arrives, adjust the performance capability of the frequency modulation scheduling mechanism using the real-time performance factors corresponding to the real-time delay.

[0106] It should be noted that the execution order of steps 5 and 6 is not limited. Step 5 can be executed first and then step 6, or step 6 can be executed first and then step 5, or steps 5 and 6 can be executed simultaneously. Specifically, the choice can be made according to the actual situation.

[0107] Based on the above embodiments, the terminal device provided in this application embodiment includes an application scenario identification module, a user perception model selection module, a parameter predefinition or update module, a terminal device usage status detection module, a performance target point adjustment module, a screen refresh rate acquisition module, a model conversion module, a real-time latency counting module, a performance factor determination module, and a kernel scheduler module. See details for further information. Figure 13 As shown.

[0108] Specifically, the application scenario recognition module is used to obtain information about the current application scenario.

[0109] The user perception model selection module is used to select the corresponding user perception model from multiple continuous curves in the terminal device based on the current application scenario information. The user perception model conforms to the changing trend of user perception score under different latency in the current application scenario.

[0110] The parameter predefinition or update module is used to increase or decrease the user perception score corresponding to each latency in the user perception model based on product update parameters or received device performance requirement parameters, using a pre-configured benchmark point as a reference, to obtain an adjusted user perception model; and / or, based on software update parameters or received device performance requirement parameters, using a pre-configured benchmark point as a reference, increase or decrease the latency corresponding to each user perception score in the user perception model, to obtain an adjusted user perception model; and / or, based on sensitivity parameters or received device performance requirement parameters, using a pre-configured benchmark point as a reference, amplify or reduce the user perception model to obtain an adjusted user perception model.

[0111] The terminal device usage status detection module is used to detect the current usage status of the terminal device.

[0112] The performance target point adjustment module is used to adjust the performance target points determined from a set of reference points according to the usage status.

[0113] The screen refresh rate acquisition module is used to acquire the current screen refresh rate.

[0114] The model conversion module is used to downsample and quantize the user perception model based on pre-configured benchmarks to obtain a set of reference points corresponding to the user perception model. The module then performs calculations on the set of reference points to obtain the relationship between performance factors and latency.

[0115] The real-time delay counting module is used to determine the real-time delay.

[0116] The performance factor determination module is used to transform the user perception model to obtain the correspondence between performance factors and latency; based on the correspondence between performance factors and latency, the real-time performance factors corresponding to the real-time latency are determined.

[0117] The kernel scheduler module is used to adjust the performance capabilities of the frequency modulation scheduling mechanism based on real-time performance factors when real-time delays occur.

[0118] Based on the above embodiments, another embodiment of this application provides a terminal device 1, such as... Figure 14 As shown, the terminal device 1 includes:

[0119] Application scenario identification module 10 is used to obtain current application scenario information.

[0120] The user perception model selection module 11 is used to select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency.

[0121] The pre-configured reference point determination module 12 is used to determine the pre-configured reference point corresponding to the user perception model.

[0122] The model conversion module 13 is used to perform downsampling and quantization processing on the user perception model based on the pre-configured benchmark points to obtain a set of reference points corresponding to the user perception model, and to perform calculations on the set of reference points to obtain the relationship between performance factors and latency.

[0123] The real-time delay counting module 14 is used to determine the real-time delay.

[0124] The performance factor determination module 15 is used to perform model transformation on the user perception model to obtain the correspondence between performance factors and latency; and to determine the real-time performance factors corresponding to the real-time latency based on the correspondence between performance factors and latency.

[0125] The kernel scheduler module 16 is used to adjust the performance capability of the frequency modulation scheduling mechanism using the real-time performance factors when the real-time delay arrives.

[0126] Optionally, terminal device 1 may further include: a screen refresh rate acquisition module.

[0127] The screen refresh rate acquisition module is used to acquire the screen refresh rate of the terminal device or user-required parameters.

[0128] Optionally, the model conversion module 13 is further configured to downsample and quantize the user perception model according to the screen refresh rate or user demand parameters to obtain a set of reference points corresponding to the user perception model, and to perform calculations on the set of reference points based on the pre-configured benchmark points to obtain the relationship between performance factors and latency.

[0129] Optionally, terminal device 1 may further include: a terminal device usage status detection module.

[0130] The terminal device usage status detection module is used to identify the usage status of the terminal device.

[0131] Optionally, terminal device 1 may further include: a performance target point determination module.

[0132] The performance target point determination module is used to determine the performance target point from the set of reference points based on the usage status.

[0133] Optionally, the model conversion module 13 is further configured to perform calculations on the set of reference points based on the pre-configured benchmark points to obtain the initial performance factor and time delay correspondence.

[0134] Optionally, the performance factor determination module 15 is further configured to determine the user perception score comparison result between the performance target point and the pre-configured benchmark point; determine the adjustment degree value based on the user perception score comparison result; and increase or decrease the adjustment degree value for each initial performance factor value corresponding to each delay in the initial performance factor and delay correspondence relationship to obtain the performance factor and delay correspondence relationship.

[0135] Optionally, terminal device 1 may further include: a parameter predefinition or update module.

[0136] The parameter predefinition or update module is used to obtain latency sensitivity parameters, product update parameters, software update parameters, and / or device performance requirement parameters based on the target user identified by the terminal device.

[0137] Optionally, the terminal device 1 may further include: a user perception model adjustment module.

[0138] The user perception model adjustment module is used to adjust the user perception model based on the pre-configured benchmark point to obtain the adjusted user perception model.

[0139] Optionally, the model conversion module 13 is further configured to perform model conversion on the adjusted user perception model based on the reference point in the adjusted user perception model, so as to obtain the correspondence between the performance factors and the latency.

[0140] Optionally, the pre-configured benchmark point determination module 12 is further configured to determine the average value of the user perception score in the user perception model; and determine the point corresponding to the average value in the user perception model as the pre-configured benchmark point.

[0141] Optionally, the user perception model adjustment module is further configured to, based on the product update parameters or received device performance requirement parameters, and using the pre-configured benchmark point as a reference, increase or decrease the user perception score corresponding to each latency in the user perception model to obtain an adjusted user perception model; and / or, based on the software update parameters or received device performance requirement parameters, and using the pre-configured benchmark point as a reference, increase or decrease the latency corresponding to each user perception score in the user perception model to obtain an adjusted user perception model; and / or, based on the sensitivity parameters or received device performance requirement parameters, and using the pre-configured benchmark point as a reference, enlarge or reduce the user perception model to obtain an adjusted user perception model.

[0142] This application provides a terminal device that acquires current application scenario information; selects a corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency; determines a pre-configured benchmark point corresponding to the user perception model; and performs model transformation on the user perception model based on the pre-configured benchmark point to obtain the correspondence between performance factors and latency; and determines the real-time performance factor corresponding to the real-time latency based on the correspondence between performance factors and latency, so as to adjust the performance capability of the frequency modulation scheduling mechanism using the real-time performance factor when the real-time latency arrives. Therefore, the terminal device proposed in this application embodiment, when the frequency modulation scheduling mechanism schedules resources, adds a user perception model. Based on the user perception model, it determines the performance factor corresponding to each latency in the user perception score of the user perception model, and obtains the correspondence between performance factors and latency. In latency-related scenarios, it can consider the performance factors corresponding to user perception, adjust the speed of the frequency modulation scheduling mechanism in a timely manner, and then change the resource scheduling scheme. Thus, when the resources required by the current application scenario increase, the frequency modulation scheduling mechanism can schedule the corresponding resources in a timely manner to meet the performance requirements of the current application scenario; when the resources required by the current application scenario decrease, it can release resources in a timely manner to reduce unnecessary power consumption.

[0143] Figure 15 This is a schematic diagram of the composition structure of a terminal device 1 provided in an embodiment of this application. In practical applications, based on the same disclosed concept of the above embodiments, such as... Figure 15 As shown, the terminal device 1 in this embodiment includes a processor 17, a memory 18, and a communication bus 19.

[0144] In specific embodiments, the application scenario identification module 10, user perception model selection module 11, pre-configuration benchmark point determination module 12, model conversion module 13, real-time delay counting module 14, performance factor determination module 15, kernel scheduler module 16, screen refresh rate acquisition module, terminal device usage status detection module, performance target point determination module, parameter predefinition or update module, and user perception model adjustment module can be implemented by a processor 17 located on the terminal device 1. The processor 17 can be at least one of an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a CPU, a controller, a microcontroller, or a microprocessor. It is understood that for different devices, the electronic devices used to implement the above processor functions can also be other types; this embodiment does not specifically limit them.

[0145] In this embodiment, the communication bus 19 is used to establish communication between the processor 17 and the memory 18; when the processor 17 executes the running program stored in the memory 18, it implements the following resource scheduling method:

[0146] Obtain current application scenario information; select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency; determine the pre-configured benchmark point corresponding to the user perception model; and based on the pre-configured benchmark point, perform model transformation on the user perception model to obtain the correspondence between performance factors and latency; based on the correspondence between performance factors and latency, determine the real-time performance factor corresponding to the real-time latency, so as to adjust the performance capability of the frequency modulation scheduling mechanism using the real-time performance factor when the real-time latency arrives.

[0147] Furthermore, the processor 17 is also used to obtain the screen refresh rate or user demand parameters of the terminal device, and to downsample and quantize the user perception model according to the screen refresh rate or user demand parameters to obtain a set of reference points corresponding to the user perception model; and to perform calculations on the set of reference points based on the pre-configured reference points to obtain the relationship between the performance factor and the latency.

[0148] Furthermore, the processor 17 is also configured to identify the usage state of the terminal device; and determine a performance target point from the set of reference points based on the usage state; perform calculations on the set of reference points using the pre-configured benchmark point as a reference to obtain the initial performance factor and latency correspondence; determine the user perception score comparison result between the performance target point and the pre-configured benchmark point; determine the adjustment degree value based on the user perception score comparison result; and increase or decrease the adjustment degree value for the initial performance factor value corresponding to each latency in the initial performance factor and latency correspondence to obtain the performance factor and latency correspondence.

[0149] Furthermore, the processor 17 is also configured to adjust the user perception model based on the pre-configured reference point according to the latency sensitivity parameters, product update parameters, software update parameters and / or device performance requirement parameters of the target user identified by the terminal device, to obtain an adjusted user perception model; and to perform model transformation on the adjusted user perception model based on the reference point in the adjusted user perception model to obtain the correspondence between performance factors and latency.

[0150] Furthermore, the processor 17 is also configured to, based on the product update parameters or received device performance requirement parameters, and using the pre-configured benchmark point as a reference, increase or decrease the user perception score corresponding to each latency in the user perception model to obtain an adjusted user perception model; and / or, based on the software update parameters or received device performance requirement parameters, and using the pre-configured benchmark point as a reference, increase or decrease the latency corresponding to each user perception score in the user perception model to obtain an adjusted user perception model; and / or, based on the sensitivity parameters or received device performance requirement parameters, and using the pre-configured benchmark point as a reference, enlarge or reduce the user perception model to obtain an adjusted user perception model.

[0151] Furthermore, the processor 17 is also used to determine the average value of the user perception score in the user perception model; and to determine the point corresponding to the average value in the user perception model as the pre-configured benchmark point.

[0152] Based on the above embodiments, this application provides a storage medium storing a computer program thereon. The computer-readable storage medium stores one or more programs, which can be executed by one or more processors and applied in a terminal device. The computer program implements the resource scheduling method described above.

[0153] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0154] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause an image display device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this disclosure.

[0155] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A resource scheduling method, characterized in that, The method includes: Obtain current application scenario information; and select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency; Determine the pre-configured benchmark point corresponding to the user perception model; and based on the pre-configured benchmark point, perform model transformation on the user perception model to obtain the correspondence between performance factors and latency; wherein, the performance factor is a coefficient for adjusting the performance capability of the frequency modulation scheduling mechanism; Based on the relationship between the performance factors and the delay, the real-time performance factors corresponding to the real-time delay are determined, so that when the real-time delay arrives, the performance capabilities of the frequency modulation scheduling mechanism can be adjusted using the real-time performance factors.

2. The method according to claim 1, characterized in that, The step of transforming the user perception model based on the pre-configured benchmark to obtain the correspondence between performance factors and latency includes: The screen refresh rate or user demand parameters of the terminal device are obtained, and the user perception model is downsampled and quantized according to the screen refresh rate or user demand parameters to obtain a set of reference points corresponding to the user perception model. Using the pre-configured benchmark as a reference, calculations are performed on the set of reference points to obtain the relationship between the performance factor and the latency.

3. The method according to claim 2, characterized in that, After obtaining the screen refresh rate or user demand parameters of the terminal device, and downsampling and quantizing the user perception model based on the screen refresh rate or user demand parameters to obtain a set of reference points corresponding to the user perception model; before obtaining the performance factor and latency correspondence by calculating the set of reference points based on the pre-configured benchmark points, the method further includes: Identify the usage status of the terminal device; and determine the performance target point from the set of reference points based on the usage status; Accordingly, the step of using the pre-configured benchmark point as a reference to perform calculations on the set of reference points to obtain the performance factor and latency correspondence includes: Using the pre-configured benchmark as a reference, calculations are performed on the set of reference points to obtain the initial performance factor and latency correspondence. Determine the user perception score comparison result between the performance target point and the pre-configured baseline point; Based on the user perception score comparison results, an adjustment degree value is determined, and the adjustment degree value is increased or decreased for each initial performance factor value corresponding to each delay in the initial performance factor and delay correspondence relationship to obtain the performance factor and delay correspondence relationship.

4. The method according to claim 3, characterized in that, The usage state includes at least one of the following: The terminal device's current usage mode, usage time distribution within a preset time period, and usage frequency of preset usage scenarios.

5. The method according to claim 2 or 3, characterized in that, The calculation of the set of reference points includes: Perform differential mirroring, table lookup, and / or quantization operations on the set of reference points.

6. The method according to claim 1, characterized in that, After determining the pre-configured baseline point corresponding to the user perception model, the method further includes: Based on the latency sensitivity parameters, product update parameters, software update parameters, and / or device performance requirement parameters of the target user identified by the terminal device, the user perception model is adjusted based on the pre-configured benchmark point to obtain the adjusted user perception model. Accordingly, based on the pre-configured benchmark, the user perception model is transformed to obtain the correspondence between performance factors and latency, including: Based on the benchmark in the adjusted user perception model, the adjusted user perception model is transformed to obtain the correspondence between performance factors and latency.

7. The method according to claim 6, characterized in that, The step of adjusting the user perception model based on the latency sensitivity parameters, product update parameters, software update parameters, and / or device performance update parameters of the target user identified by the terminal device, and on the pre-configured benchmark point, to obtain the adjusted user perception model, includes: Based on the product update parameters or the received device performance requirement parameters, and using the pre-configured benchmark as a reference, the user perception score corresponding to each time delay in the user perception model is increased or decreased to obtain the adjusted user perception model. And / or, based on the software update parameters or the received device performance requirement parameters, and using the pre-configured benchmark as a reference, the latency corresponding to each user perception score in the user perception model is increased or decreased to obtain an adjusted user perception model; And / or, based on the sensitivity parameter or the received device performance requirement parameter, using the pre-configured benchmark as a reference, the user perception model is amplified or reduced to obtain an adjusted user perception model.

8. The method according to claim 1, characterized in that, Determining the pre-configured baseline point corresponding to the user perception model includes: Determine the average value of the user perception scores in the user perception model; The point corresponding to the average value in the user perception model is determined as the pre-configured benchmark point.

9. A terminal device, characterized in that, The terminal device includes: The application scenario recognition module is used to obtain information about the current application scenario; The user perception model selection module is used to select the corresponding user perception model based on the current application scenario information; the user perception model represents the correspondence between user perception score and latency. A parameter pre-definition or update module is used to determine the pre-configured benchmark point corresponding to the user perception model; The model conversion module is used to perform downsampling and quantization processing on the user perception model based on the pre-configured benchmark points to obtain a set of reference points corresponding to the user perception model, and to perform calculations on the set of reference points to obtain the relationship between performance factors and latency; wherein, the performance factors are coefficients for adjusting the performance capability of the frequency modulation scheduling mechanism. The real-time delay counting module is used to determine the real-time delay. The performance factor determination module is used to perform model transformation on the user perception model to obtain the correspondence between performance factors and latency; and to determine the real-time performance factors corresponding to the real-time latency based on the correspondence between performance factors and latency. The kernel scheduler module is used to adjust the frequency modulation speed of the frequency modulation scheduling mechanism using the real-time performance factors when the real-time delay arrives.

10. A terminal device, characterized in that, The terminal device includes: a processor, a memory, and a communication bus; when the processor executes the running program stored in the memory, it implements the method as described in any one of claims 1-8.

11. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

Images