Computing resource scheduling method, scheduler, Internet of Things system and computer readable medium

Abstract

The invention relates to distributed computing technology, and in particular to a computing resource scheduling method, a scheduler, an Internet of Things system and a computer-readable medium. The Internet of Things system provided by an embodiment of the present invention comprises: a sensor (10), a data acquisition module (20) connected to the sensor (10), at least one edge computing device (30) that provides computing resources required for data processing for the IoT system (100), and a scheduler (40). The scheduler (40) is configured to receive the data to be processed collected by the sensor (10) from the data acquisition module (20), determine the processing priority of the data to be processed, predict the amount of computing resources and duration required to process the data to be processed according to the processing priority, and schedule the computing resources of the at least one edge computing device (30) according to the predicted amount of computing resources and duration to process the data to be processed.

Description

Technical Field

[0001] The present invention relates to distributed computing technology, and in particular to a computing resource scheduling method, a scheduler, an Internet of Things system and a computer-readable medium. Background Art

[0002] Distributed computing is a computing method that, compared with centralized computing, divides computing tasks into multiple small parts and assigns them to multiple devices to complete separately.

[0003] In traditional Internet of Things (IoT) systems, centralized computing is usually used for data processing. The distance between IoT devices and cloud platforms is relatively long, resulting in large data transmission latency. Therefore, for application scenarios with high latency requirements such as industrial IoT systems, if distributed computing is used and the computing resources of edge computing devices are used, data transmission latency can be effectively reduced, and the occupation of network bandwidth and cloud resources can be reduced.

[0004] In edge computing, how to effectively manage computing resources is crucial to improving computing efficiency. Task scheduling, as a major way to manage computing resources, can improve system efficiency and service quality. Summary of the invention

[0005] The present disclosure provides a computing resource scheduling method, a scheduler, an IoT system, and a computer-readable medium in distributed computing.

[0006] In a first aspect, an Internet of Things (IoT) system is provided, comprising a sensor, a data acquisition module connected to the sensor, and at least one edge computing device, configured to provide computing resources required for data processing for the IoT system; a scheduler connected to the data acquisition module, configured to receive the data to be processed collected by the sensor from the data acquisition module, determine the processing priority of the data to be processed, predict the amount of computing resources and duration required to process the data to be processed according to the determined processing priority, and schedule the computing resources of the at least one edge computing device according to the predicted amount of computing resources and duration to process the data to be processed.

[0007] In the second aspect, an edge distributed data processing system is provided for processing data collected by sensors in an Internet of Things (IoT) system. The system may include at least one edge computing device configured to provide the IoT system with computing resources required for data processing; a scheduler configured to: receive data to be processed collected by a sensor from a data acquisition module in the IoT system, determine the processing priority of the data to be processed, predict the amount of computing resources and duration required to process the data to be processed based on the determined processing priority, and schedule the computing resources of at least one edge computing device to process the data to be processed according to the predicted amount of computing resources and duration.

[0008] In a third aspect, a resource scheduling method is provided. The method may include the following steps: receiving data to be processed collected by a sensor in an Internet of Things (IoT) system; determining the processing priority of the data to be processed; predicting the amount of computing resources and duration required to process the data to be processed according to the determined processing priority; and scheduling the computing resources of the edge computing device in the IoT system to process the data to be processed according to the predicted amount of computing resources and duration.

[0009] In a fourth aspect, a scheduler is provided. The scheduler may include: a receiving module configured to receive data to be processed collected by a sensor in an Internet of Things (IoT) system; a priority determination module configured to determine the processing priority of the data to be processed; and a scheduling module configured to: predict the amount of computing resources and duration required to process the data to be processed according to the processing priority determined by the priority determination module, and schedule the computing resources of the edge computing device in the IoT system to process the data to be processed according to the predicted amount of computing resources and duration.

[0010] In a fifth aspect, a scheduler is provided, comprising: at least one memory for storing computer-readable code; and at least one processor for executing the computer-readable code stored in the at least one memory to execute the method provided in the third aspect.

[0011] According to a sixth aspect, a storage medium is provided, wherein the storage medium stores computer-readable code, and when the computer-readable code is executed by at least one processor, the method provided in the third aspect is executed.

[0012] Among them, the scheduling method is used to manage computing resources, determine the processing priority of the data to be processed, predict the amount of computing resources and duration required for data processing based on the processing priority, and schedule the computing resources of the edge computing device for data processing based on this. It realizes precise scheduling based on priority, takes into account the amount of computing resources and duration required for data processing, and can achieve optimal scheduling of resources and efficient completion of data processing tasks.

[0013] In the above aspects, optionally, the greater the fluctuation amplitude of the value of the physical quantity represented by the data to be processed, the higher the processing priority of the data to be processed. In the judgment of the processing priority, the fluctuation of the value of the physical quantity represented by the data to be processed is taken into account. When the fluctuation amplitude is large, it is considered that the data to be processed should be processed first, so that real-time processing of emergency situations such as equipment failure is realized, and computing resources can be effectively allocated to problems that need to be solved urgently.

[0014] In the above aspects, optionally, the higher the processing priority, the earlier the data to be processed is predicted to have the amount of computing resources and duration required to process the data to be processed. Processing based on the FIFO mechanism further ensures timely processing of high-priority data to be processed.

[0015] In the above aspects, optionally, when predicting the amount of computing resources and duration required to process the data to be processed according to the determined processing priority, the amount of computing resources and duration occupied when the historical data similar to the length of the data to be processed, the type of the sensor and the processing priority are found, and the amount of computing resources and duration required to process the data to be processed are predicted based on the amount of computing resources and duration occupied when the historical data is processed. Since the information of the historical data is referenced, the prediction result is more accurate, providing accurate and reliable information for further allocation of computing resources.

[0016] In the above aspects, optionally, computing resources of edge computing devices that meet the predicted computing resource amount and duration can be selected from each edge computing device in the IoT system to process the data to be processed. In this way, the processing requirements of the data to be processed can be met.

[0017] In the above aspects, optionally, after the data to be processed is processed, the amount of computing resources and duration used to process the data to be processed are recorded to serve as historical data for subsequent data processing, providing a reference for predicting the amount of computing resources and duration. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the structure of an IoT system and an edge distributed data processing system provided in an embodiment of the present invention.

[0019] Figure 2 A schematic diagram of the structure of a scheduler provided in an embodiment of the present invention.

[0020] Figure 3 Another structural diagram of a scheduler provided in an embodiment of the present invention.

[0021] Figure 4 The process of various modules in the scheduler provided by the embodiment of the present invention cooperating to complete task scheduling is shown.

[0022] Figure 5 A flowchart of a resource scheduling method provided by an embodiment of the present invention.

[0023] Figure 6 The figure is a schematic diagram of a scenario to which the embodiments of the present invention can be applied.

[0024] List of reference numerals:

[0025] 100: IoT system 10: Sensor 20: Data acquisition module

[0026] 30: Edge computing device 40: Scheduler 50: Cloud

[0027] 200: Edge distributed data processing system 300: Resource scheduling method

[0028] 401: Receiving module 402: Priority determination module 403: Scheduling module

[0029] 404: recording module 405: at least one memory 406: at least one processor

[0030] 407: Communication module 408: Bus 409: Historical data

[0031] 410: Available computing resource information

[0032] S301: receiving data to be processed

[0033] S302: Determine the priority of data to be processed

[0034] S303: Predicting computing resources and duration

[0035] S305: Record the amount of computing resources and duration used to process the data to be processed

[0036] S306: Obtaining available computing resource information

[0037] S3031: Determine whether there is historical data similar to the data to be processed

[0038] S3032: Predict the amount of computing resources and duration of data to be processed based on historical data

[0039] S304: Scheduling edge computing devices that meet the predicted computing resource volume and duration

[0040] S304': scheduling edge computing devices with minimal delay and computing resource availability exceeding a threshold DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail with reference to the accompanying drawings. The embodiments described below are only part of the embodiments of the present invention, but not all of them.

[0042] Figure 1 A schematic diagram of the structure of an IoT system and an edge distributed data processing system provided in an embodiment of the present invention.

[0043] like Figure 1 As shown, the IoT system 100 may include:

[0044] At least one sensor 10, the sensor 10 is used as a field device to collect data of the connected device for monitoring the operating status of the device, etc.

[0045] The data acquisition module 20 connected to each sensor 10 is configured to collect data collected by the connected sensor 10, such as temperature data collected by a temperature sensor, pressure data collected by a pressure sensor, liquid flow rate data collected by a water flow sensor, etc. The data collected by the sensor can be used to monitor the operating status of the device to which the sensor is connected. This is crucial for the normal operation of the device and the timely discovery of faults. Among them, one sensor 10 can be connected to at least one data acquisition module 20, and one data acquisition module 20 can also be connected to at least one sensor 10. The data acquisition module 20 can be a system on a chip (SOC) device, a microprocessor, etc. The sensor data collected by the data acquisition module 20 is sent to the scheduler 40.

[0046] At least one edge computing device 30 is configured to provide the IoT system 100 with computing resources required for data processing, and the computing resources may include a central processing unit, memory, etc. The amount of available computing resources can be measured by the central processing unit occupancy rate and the memory occupancy rate. Each edge computing device 30 can be organized in the form of a device cluster, and each edge computing device 30 is a node in the cluster and is scheduled by the scheduler 40.

[0047] At least one scheduler 40. The scheduler 40 may be an intelligent edge device with processing capabilities. On the one hand, it receives the data to be processed from the data acquisition module 20, and on the other hand, it is connected to the edge computing device 30, and is configured to: receive the data to be processed collected by the sensor 10 from the data acquisition module 20, determine the processing priority of the data to be processed, predict the amount of computing resources and duration required to process the data to be processed according to the determined processing priority, and schedule the computing resources of at least one edge computing device 30 according to the predicted amount of computing resources and duration to process the data to be processed.

[0048] The IoT system 100 may further include a cloud 50. The cloud 50 generally has more computing resources than the edge devices, and tasks that do not require high latency can be processed by the cloud 50.

[0049] Figure 1 In the example, the edge computing device 30 and the scheduler 40 together constitute an edge distributed data processing system 200. It should be noted that the edge distributed data processing system 200 can be regarded as a part of the IoT system 100, or exist independently of the IoT system 100.

[0050] In the embodiment of the present invention, the scheduler 40 determines the priority of the data to be processed, predicts the amount of computing resources and duration required to process the data to be processed, and schedules the computing resources accordingly. This achieves accurate scheduling based on priority, taking into account the amount of computing resources and duration required for data processing, and can achieve optimal resource scheduling and efficient completion of data processing tasks. Figure 2 , Figure 3 and Figure 4 The scheduler 40 is further described.

[0051] Figure 2 FIG. 4 is a schematic diagram of a structure of the scheduler 40. Figure 2 As shown, the scheduler 40 may include:

[0052] A receiving module 401 is configured to receive data to be processed collected by a sensor 10 in an Internet of Things IoT system 100;

[0053] a priority determination module 402, configured to determine the processing priority of the data to be processed;

[0054] A scheduling module 403 is configured to: predict the amount of computing resources and duration required to process the data to be processed according to the processing priority determined by the priority determination module 402, and schedule the computing resources of the edge computing device in the IoT system to process the data to be processed according to the predicted amount of computing resources and duration.

[0055] In addition, the scheduler 40 may also include a recording module 404, which is configured to record the amount of computing resources and duration used to process the data to be processed after the data to be processed is processed, which can be used as historical data of subsequent data to be processed and as a reference when predicting the amount of computing resources and duration required for subsequent data to be processed.

[0056] Next, combine Figure 4 The processing flow of each module in the scheduler 40 on the data to be processed is described.

[0057] In step S301 , the receiving module 401 receives data to be processed collected by a sensor from a data collection module 20 in the IoT system 100 .

[0058] In step S302, the priority determination module 402 determines the processing priority of the data to be processed. Optionally, the greater the fluctuation range of the value of the physical quantity represented by the data to be processed, the higher the processing priority of the data to be processed.

[0059] Specifically, the priority determination module 402 may generate an upper limit and a lower limit of the physical quantity value represented by the data according to the previous data collected by the sensor 10. Here, the t distribution may be used to estimate the upper limit and the lower limit.

[0060] For t distribution, if there are n data samples (n is a positive integer), x is the sample of data collected by the sensor 10. Then, the upper limit v u and the lower limit v l It can be expressed as:

[0061]

[0062]

[0063] here, represents the mean of the t distribution, S n represents the standard deviation of the t distribution, t (a，n-1) It represents the standard t distribution coefficient under the condition that the sample size is n.

[0064] At a given time t, the value of the physical quantity represented by the data collected by the sensor 10 is expressed as v t Then, at a given time t, the processing priority p of the data collected by the sensor can be calculated by the following formula:

[0065]

[0066] Among them, the larger the p value, the higher the processing priority of the data, and the more the data should be processed first.

[0067] The above formula is only an example. Its physical meaning is that if the fluctuation amplitude of the physical quantity represented by the data collected by a sensor 10 is larger, it means that the possibility of failure of the equipment monitored by the sensor 10 is greater. In this case, the data should be processed first.

[0068] like Figure 4 As shown in the queue at the rear of the box 402, the data to be processed collected by each sensor 10 can be sorted in the queue according to their respective processing priorities, and can be processed using a First In First Out (FIFO) strategy. The data to be processed with a high processing priority will be processed first in the next step S303, that is, the higher the processing priority, the earlier the data to be processed will be predicted for the amount of computing resources and time required to process the data to be processed.

[0069] In step S303: the scheduling module 403 predicts the amount of computing resources and duration required to process the data to be processed according to the determined processing priority, and schedules the computing resources of the edge computing device in the IoT system to process the data to be processed according to the predicted amount of computing resources and duration.

[0070] Specifically, refer to Figure 5 In sub-step S3031 of step S303, the scheduling module 403 may search for historical data similar to the data to be processed. For example, the amount of computing resources and duration occupied by historical data similar to the length, sensor type and processing priority of the data to be processed are searched. If found, the scheduling module 403 further executes sub-step S3032, otherwise the scheduling module 403 executes step S304'. Among them, the rules for determining "similar" can be pre-set, for example, if the length of the data to be processed does not exceed 110% of the length of the historical data and is not shorter than 90% of the length of the historical data, it is determined that the data to be processed is similar to the historical data in length. If the type of sensor 10 that collects the data to be processed is the same as that of the sensor that collects the historical data, it is determined that the data to be processed is similar to the historical data in sensor type. If the processing priority of the data to be processed is 1 or 2 levels higher than the processing priority of the historical data, or the processing priority of the data to be processed is the same as the processing priority of the historical data, or the processing priority of the data to be processed is 1 level lower than the processing priority of the historical data, it is determined that the data to be processed is similar to the historical data in processing priority. If the length, processing priority and sensor type are similar, the data to be processed is determined to be similar to the historical data. The above determination rules are only for illustration and can be flexibly set according to specific application scenarios in actual applications.

[0071] In sub-step S3032, the scheduling module 403 can predict the amount of computing resources and duration required to process the data to be processed based on the amount of computing resources and duration occupied when the found historical data was processed. And further, in step S304, the scheduling module 403 can schedule the computing resources of the edge computing devices in the IoT system according to the predicted amount of computing resources and duration to process the data to be processed, such as: selecting the computing resources of the edge computing device 30 that meets the predicted amount of computing resources and duration from each edge computing device 30 in the IoT system 100 to process the data to be processed. Optionally, if there are multiple edge computing devices 30 that meet the requirements, the device with the smallest network delay to the scheduler 40 can be further selected to reduce the data processing delay.

[0072] In step S304', the scheduling module 403 may select an edge computing device 30 with the smallest network delay to the scheduler 40 and an available computing resource amount greater than a given threshold (e.g., 30%) of the total computing resource amount of the edge computing device 30 itself to process the data to be processed. If such an edge computing device 30 is not found, an alarm may be issued, optionally.

[0073] Further, in step S305, after the data to be processed is processed, the recording module 404 can record the amount of computing resources and duration used to process the data to be processed in the database of historical data 409. In the database, the length of the data to be processed, the processing priority, and the required amount of computing resources and duration corresponding to each sensor type can be stored. In this way, when the scheduling module 403 receives a data to be processed, it can search the historical data 409 according to the type of the sensor 10 that collected the data to be processed, the processing priority, and the length of the data to be processed, and find similar records to predict the amount of computing resources and duration.

[0074] Furthermore, the scheduling module 403 may use a heartbeat mechanism to obtain the status of computing resources in each edge computing device 30, such as "available", "unavailable", "occupied", etc., and the amount of available computing resources (such as CPU occupancy, memory occupancy, etc.), so as to schedule computing resources in appropriate edge computing devices 30 when processing the data to be processed. The scheduling module 403 may store the above information obtained in the available computing resource information 410. Since this information needs to be frequently accessed, a nosql mechanism (such as redis) may be optionally used for information storage.

[0075] Figure 3 Another structural diagram of the scheduler provided in the embodiment of the present invention. Figure 3As shown, the scheduler 40 may include at least one memory 405 for storing computer-readable codes; at least one processor 406 for executing the computer-readable codes stored in the memory 405, thereby executing the aforementioned method 300. Among them, the at least one memory 405 and the at least one processor 406 and the communication module 407 may be connected through a bus 408. The communication module 407 communicates with external devices such as the data acquisition module 20, the edge computing device 30 and the cloud 50 under the control of the at least one processor 406. Figure 2 and Figure 4 Each module included in the scheduler 40 can be regarded as a program module written by a computer-readable code stored in the memory 405. When these program modules are called by the processor 406, the aforementioned method 300 can be executed.

[0076] Figure 5 Flow chart of a resource scheduling method provided by an embodiment of the present invention. The method 300 may include the following steps:

[0077] S301: receiving data to be processed;

[0078] S302: Determine the priority of the data to be processed;

[0079] S303: predicting computing resource amount and duration;

[0080] Wherein, step S303 may include sub-steps S3031 and S3032:

[0081] S3031: Determine whether there is historical data similar to the data to be processed. If there is historical data similar to the data to be processed, execute sub-step S3032; otherwise, execute step S304';

[0082] S3032: predicting the amount of computing resources and duration of the data to be processed based on historical data, and then executing step S304;

[0083] S304: Schedule edge computing devices that meet the predicted computing resource volume and duration, and then execute step S305;

[0084] S304': scheduling edge computing devices with minimal delay and computing resource availability exceeding a threshold

[0085] S305: Record the amount of computing resources and duration used to process the data to be processed.

[0086] Other optional implementations of the above-mentioned method steps can refer to Figure 4 The description of part is not repeated here.

[0087] Figure 6Schematic diagram of a scenario to which various embodiments of the present invention can be applied. Among them, the edge distributed data processing system 200 deployed at the edge can realize data monitoring of an IoT system (such as a building or factory). The data is collected by various sensors 10 and processed at the edge, and the cloud 50 can be used as a data warehouse to store business logic (such as data reports).

[0088] In addition, an embodiment of the present invention further provides a computer-readable medium, which stores a computer-readable code. When the computer-readable code is executed by at least one processor, the aforementioned method 300 is implemented.

[0089] In summary, the embodiments of the present invention provide a computing resource scheduling method, a scheduler, an Internet of Things system, and a computer-readable medium. Computing resources are scheduled according to the processing priority of the data to be processed, and accurate scheduling based on priority is achieved, taking into account the amount of computing resources and duration required for data processing, and can achieve optimal resource scheduling and efficient completion of data processing tasks. In the judgment of processing priority, the fluctuation of the physical value represented by the data to be processed is taken into account. When the fluctuation amplitude is large, it is considered that the data to be processed should be processed first, and real-time processing of emergency situations such as equipment failure is achieved, and computing resources can be effectively allocated to problems that need to be solved urgently. When computing resources are predicted and scheduled, the data to be processed can be sorted according to the processing priority, and processed based on the FIFO mechanism, further ensuring the timely processing of the data to be processed with high priority. The heartbeat mechanism is further adopted to obtain the latest computing resource status, thereby ensuring the effective scheduling of computing resources. In addition, when predicting the amount of computing resources and duration required for processing the data to be processed, the information of historical data is referenced, so that the prediction result is more accurate, and accurate and reliable information is provided for further allocation of computing resources.

[0090] It should be noted that not all steps and modules in the above-mentioned processes and system structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by at least two physical entities, or some components in at least two independent devices can be implemented together.

[0091] In the above embodiments, the hardware unit can be realized by mechanical or electrical means. For example, a hardware unit can include permanent dedicated circuits or logic (such as special processors, field programmable gate arrays (Field-Programmable Gate Array, FPGA) or application specific integrated circuits (Application Specific Integrated Circuits, ASIC) etc.) to complete the corresponding operation. The hardware unit can also include programmable logic or circuits (such as general-purpose processors or other programmable processors), which can be temporarily set by software to complete the corresponding operation. Specific implementation (mechanical means or dedicated permanent circuits or temporarily set circuits) can be determined based on cost and time considerations.

[0092] The embodiments of the present invention are shown and described in detail above through the accompanying drawings and preferred embodiments. However, the embodiments of the present invention are not limited to these disclosed embodiments. Based on the above embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present invention, and these embodiments are also within the protection scope of the embodiments of the present invention.

Claims

1. An Internet of Things (IoT) system (100), characterized in that: include: A sensor (10); A data acquisition module (20) connected to the sensor (10); At least one edge computing device (30) configured to provide computing resources required for data processing for the IoT system (100); A scheduler (40) connected to the data acquisition module (20) is configured to: receiving the data to be processed collected by the sensor (10) from the data collection module (20); Determining the processing priority of the data to be processed; wherein, the greater the fluctuation amplitude of the value of the physical quantity represented by the data to be processed, the higher the processing priority of the data to be processed; Predicting the amount of computing resources and duration required to process the data to be processed based on the determined processing priority; wherein searching for historical data similar to the data to be processed; predicting the amount of computing resources and duration required to process the data to be processed based on the amount of computing resources and duration occupied when the historical data was processed; The computing resources of the at least one edge computing device (30) are scheduled according to the predicted computing resource amount and the duration to process the data to be processed.

2. An edge distributed data processing system (200), for processing data collected by sensors in an Internet of Things (IoT) system (100), characterized in that: include: At least one edge computing device (30) configured to provide computing resources required for data processing for the IoT system (100); A scheduler (40) is configured to: Receiving data to be processed collected by a sensor (10) from a data collection module (20) in the IoT system (100); Determining the processing priority of the data to be processed; wherein, the greater the fluctuation amplitude of the value of the physical quantity represented by the data to be processed, the higher the processing priority of the data to be processed; Predicting the amount of computing resources and duration required to process the data to be processed based on the determined processing priority; wherein searching for historical data similar to the data to be processed; predicting the amount of computing resources and duration required to process the data to be processed based on the amount of computing resources and duration occupied when the historical data was processed; The computing resources of the at least one edge computing device (30) are scheduled according to the predicted computing resource amount and the duration to process the data to be processed.

3. Resource scheduling method (300), characterized in that: include: Receiving ( S301 ) data to be processed collected by a sensor in an Internet of Things (IoT) system; Determining (S302) a processing priority of the data to be processed; wherein the greater the fluctuation amplitude of the value of the physical quantity represented by the data to be processed, the higher the processing priority of the data to be processed; Predicting (S303) the amount of computing resources and duration required to process the data to be processed based on the determined processing priority; wherein searching for historical data similar to the data to be processed; predicting the amount of computing resources and duration required to process the data to be processed based on the amount of computing resources and duration occupied when the historical data found was processed; The computing resources of the edge computing device in the IoT system are scheduled ( S304 ) according to the predicted computing resource amount and the duration to process the data to be processed.

4. The method according to claim 3, characterized in that Predicting (S303) the amount of computing resources and the duration required to process the data to be processed according to the determined processing priority, including: predicting the amount of computing resources and the duration required to process the data to be processed according to the determined processing priority, wherein the higher the processing priority, the earlier the data to be processed is predicted to have the amount of computing resources and the duration required to process the data to be processed.

5. The method according to claim 3, characterized in that Predicting (S303) the amount of computing resources and duration required to process the to-be-processed data according to the determined processing priority includes: Finding the amount of computing resources and duration occupied when processing historical data similar to the length of the data to be processed, the type of the sensor, and the processing priority; The amount of computing resources and duration required to process the data to be processed are predicted based on the amount of computing resources and duration occupied when the found historical data was processed.

6. The method according to claim 5, characterized in that Scheduling (S304) computing resources of the edge computing device in the IoT system according to the predicted computing resource amount and duration to process the data to be processed includes: The computing resources of an edge computing device that meets the predicted amount of computing resources and the duration are selected from each edge computing device in the IoT system to process the data to be processed.

7. The method according to claim 5 or 6, characterized in that Also includes: After the data to be processed is processed, the amount of computing resources and the duration used to process the data to be processed are recorded ( S305 ).

8. A scheduler (40), characterized in that include: A receiving module (401) is configured to receive data to be processed collected by a sensor (10) in an Internet of Things (IoT) system (100); A priority determination module (402) configured to determine the processing priority of the data to be processed; A scheduling module (403) is configured to: The amount of computing resources and the duration required for processing the data to be processed are predicted according to the processing priority determined by the priority determination module (402); wherein the greater the fluctuation amplitude of the value of the physical quantity represented by the data to be processed, the higher the processing priority of the data to be processed; wherein historical data similar to the data to be processed is searched; and the amount of computing resources and the duration required for processing the data to be processed are predicted according to the amount of computing resources and the duration occupied when the historical data is processed; The computing resources of the edge computing device in the IoT system are scheduled according to the predicted computing resource amount and duration to process the data to be processed.

9. The scheduler (40) according to claim 8, characterized in that The scheduling module (403) is specifically configured to: predict the amount of computing resources and time required to process the data to be processed according to the determined processing priority, wherein the higher the processing priority, the earlier the data to be processed is predicted to have the amount of computing resources and time required to process the data to be processed.

10. The scheduler (40) according to claim 8, characterized in that When the scheduling module (403) predicts the amount of computing resources and duration required to process the data to be processed according to the determined processing priority, it is specifically configured as follows: Finding the amount of computing resources and duration occupied when processing historical data similar to the length of the data to be processed, the type of the sensor, and the processing priority; The amount of computing resources and duration required to process the data to be processed are predicted based on the amount of resources and duration occupied when the found historical data was processed.

11. The scheduler (40) according to claim 10, characterized in that When the scheduling module (403) schedules the computing resources of the edge computing device in the IoT system to process the data to be processed according to the predicted computing resource amount and the duration, the scheduling module (403) is specifically configured as follows: The computing resources of an edge computing device that meets the predicted amount of computing resources and the duration are selected from each edge computing device in the IoT system to process the data to be processed.

12. The scheduler (40) according to claim 10 or 11, characterized in that Also included is a recording module (404) configured to: After the data to be processed is processed, the amount of computing resources and the duration used to process the data to be processed are recorded.

13. A scheduler (40), characterized in that include: at least one memory (405) for storing computer readable code; At least one processor (406) is used to execute the computer-readable code stored in the at least one memory (405) to execute the method according to any one of claims 3 to 7.

14. A storage medium, characterized in that The storage medium stores computer-readable codes, and when the computer-readable codes are executed by at least one processor, the method according to any one of claims 3 to 7 is executed.

Images