Task scheduling method and system applied to Internet of Things scene

Abstract

The invention provides a task scheduling method and system applied to an Internet of Things scene, and the method comprises the steps: recognizing a task uploaded by a cloud computer user side, and a virtual machine and a server subordinate to an Internet of Things operating system platform, obtaining task information, virtual machine information and server information, and determining a task scheduling strategy and a resource scheduling strategy; on the basis of a resource scheduling strategy, resources subordinate to the virtual machine are scheduled to corresponding servers, a task scheduling receiving list about all the servers is generated, and optimal configuration of different types of resources in various task scheduling scenes is achieved; the method also comprises the steps of segmenting an uploaded task into a set containing a plurality of task blocks based on a task scheduling strategy, carrying out task unloading processing on the set based on a task scheduling receiving list, generating task block calculation queues facing different types of servers, and intelligently carrying out task scheduling in a distributed architecture in a complex service scene. The execution efficiency of the Internet of Things is optimized, and the operation reliability of the Internet of Things is improved.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things, and in particular to a task scheduling method and system applied to Internet of Things scenarios. Background Art

[0002] Existing IoT applications can usually only run on the central cloud server or edge side. Due to the computing limitations of the central cloud server and the edge side themselves, the IoT has problems such as large network latency, large bandwidth consumption, high risk of single point failure, low privacy security, and limited local processing capabilities. These problems and defects will have an adverse impact on the real-time performance and operating costs of the IoT, reduce the operational reliability of the IoT, and increase the risk of privacy information leakage, and cannot improve the overall execution efficiency of the IoT. Summary of the invention

[0003] The purpose of the present invention is to provide a task scheduling method and system applied to an Internet of Things scenario, to identify the tasks uploaded by a cloud computer user end, the virtual machines and servers under the Internet of Things operating system platform, to obtain task information, virtual machine information and server information, thereby determining the task scheduling strategy and resource scheduling strategy, and providing a reliable basis for the scheduling and allocation of tasks and resources; also based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server, and a task scheduling receiving list for all servers is generated to achieve the best configuration of different types of resources in various task scheduling scenarios; also based on the task scheduling strategy, the uploaded tasks are divided into a collection containing a plurality of task blocks, and based on the task scheduling receiving list, the collection is task unloaded, a task block calculation queue for different types of servers is generated, and the calculation processing results are summarized and integrated, and each task block is sent to the cloud-edge-end queue of the Internet of Things through the task offloading mechanism, so as to improve the resource utilization and performance of the Internet of Things, intelligently perform task scheduling in a distributed architecture under complex business scenarios, optimize the execution efficiency of the Internet of Things, and improve the operation reliability of the Internet of Things.

[0004] The present invention is achieved through the following technical solutions: The task scheduling method applied to the IoT scenario includes: Identify all tasks uploaded to the IoT operating system platform by different cloud computing clients to obtain corresponding task information; identify all virtual machines and all servers under the IoT operating system platform to obtain corresponding virtual machine information and server information; determine task scheduling strategies for all tasks and resource scheduling strategies for all virtual machines based on the task information, the virtual machine information and the server information; Based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server; and based on the working status of all the servers to which the resources are scheduled, all the servers with the authority to receive the task are determined, and a task scheduling receiving list for all the servers with the authority to receive the task is generated; Based on the task scheduling strategy, the tasks uploaded by the cloud computing user end are processed in blocks to obtain a collection of several task blocks; based on the task scheduling receiving list, the collection is subjected to task offloading processing to generate task block calculation queues for different types of servers; the task block calculation queues are then allocated to corresponding servers for calculation processing, and the calculation processing results are summarized and integrated.

[0005] Optionally, all tasks uploaded by different cloud computing clients to the Internet of Things operating system platform are identified to obtain corresponding task information; all virtual machines and all servers under the Internet of Things operating system platform are identified to obtain corresponding virtual machine information and server information; based on the task information, the virtual machine information and the server information, task scheduling strategies for all tasks and resource scheduling strategies for all virtual machines are determined, including: Identify task upload requests from different cloud computing clients to obtain identity information of the cloud computing clients; based on the identity information, determine whether the cloud computing client has the authority to upload tasks to the Internet of Things operating system platform; if the cloud computing client has the authority to upload tasks, perform data identification on the tasks uploaded by the cloud computing client to the Internet of Things operating system platform to obtain corresponding task data structure information and task data content information as the task information; Based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, which are used as the virtual machine information and the server information respectively; Based on the task information, determine the scheduling order and scheduling time interval for scheduling all tasks, so as to obtain the task scheduling strategy for all tasks; based on the virtual machine information and the server information, determine the mode in which the virtual machine allocates its own computing resources and / or memory resources to the server, and use this as the resource scheduling strategy.

[0006] Optionally, the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform is monitored for recognition quality, including: Extracting the amount of data corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform; Extract the data corresponding to each task to identify the response time; Compare the data recognition response time corresponding to the task with a preset data recognition response time threshold; When the data recognition response time corresponding to the task exceeds the preset data recognition response time threshold, the data volume value of the task corresponding to the data recognition response time exceeding the preset data recognition response time threshold is retrieved; The data volume value of the task corresponding to the data recognition response time that exceeds the preset data recognition response time threshold is used as the observed data volume; The data recognition response coefficient is obtained by combining the observed data volume with the data volume corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform and the data recognition response time corresponding to each task; The data recognition response coefficient is obtained by the following formula: Wherein, E represents the data recognition response coefficient; n represents the number of tasks uploaded by the cloud computing client to the IoT operating system platform; C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g represents the data recognition response time corresponding to the observed data volume; k represents the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Where k represents the adjustment coefficient C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g Indicates the data recognition response time corresponding to the observed data volume; ΔT indicates the theoretical change in data response time corresponding to each preset unit change in data volume; comparing the data identification response coefficient with a preset coefficient threshold; When the data recognition response coefficient exceeds a preset coefficient threshold, it is determined that there is a risk of abnormal operation in the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform, and a risk warning is issued.

[0007] Optionally, based on the resource scheduling strategy, resources under the virtual machine are scheduled to corresponding servers; and based on the respective working states of all servers to which resources are scheduled, all servers with authority to receive tasks are determined, and a task scheduling receiving list for all servers with authority to receive tasks is generated, including: Based on the resource scheduling strategy, select matching computing power resources and / or memory resources from the computing power resource library and / or memory resource library under the virtual machine; and based on the connection link between the virtual machine and the server for resource scheduling within the Internet of Things, schedule the selected matching computing power resources and / or memory resources to the corresponding server; Analyze the work logs of all servers with resources scheduled to obtain the workload of tasks to be processed of all servers with resources scheduled to obtain the workload of tasks to be processed; based on the workload of tasks to be processed, determine whether the server is in an overloaded working state; if so, determine that the server has the authority to receive tasks; if not, determine that the server does not have the authority to receive tasks; Based on the order of the workload of the tasks to be processed of all the servers with the authority to receive the tasks from small to large, a task scheduling receiving list of all the servers with the authority to receive the tasks is generated.

[0008] Optionally, based on the task scheduling strategy, the tasks uploaded by the cloud computing client are processed in blocks to obtain a set of several task blocks; based on the task scheduling receiving list, the set is processed for task offloading to generate task block calculation queues for different types of servers; the task block calculation queues are then allocated to corresponding servers for calculation processing, and the calculation processing results are summarized and integrated, including: Based on the task scheduling strategy, extract corresponding tasks from all tasks uploaded to the Internet of Things operating system platform in a matching scheduling time interval, and based on the task data structure information of the extracted tasks, block the tasks to obtain a set including a plurality of task blocks; Based on the task scheduling receiving list, cloud servers and edge servers connected to the collection are selected from all servers with the authority to receive tasks, and task offloading processing is performed on the collection to generate task block cloud computing queues and task block edge computing queues for the cloud server and the edge server respectively; the task block cloud computing queues and the task block edge computing queues are then respectively allocated to the cloud server and the edge server for computing processing, and the computing processing results of the edge server are transmitted to the cloud server, so that all computing processing results are summarized and integrated on the cloud server side to obtain a complete processing result matching the task.

[0009] The task scheduling system applied to IoT scenarios includes: The task information identification module is used to identify all tasks uploaded to the IoT operating system platform by different cloud computing user terminals and obtain corresponding task information; The terminal identification module is used to identify all virtual machines and all servers under the IoT operating system platform to obtain corresponding virtual machine information and server information; A scheduling strategy generation module, used to determine the task scheduling strategy for all tasks and the resource scheduling strategy for all virtual machines based on the task information, the virtual machine information and the server information; A resource scheduling execution module, used to schedule the resources under the virtual machine to the corresponding server based on the resource scheduling policy; The task scheduling determination module is used to determine all servers with the authority to receive tasks based on the respective working states of all servers with resources scheduled, and generate a task scheduling receiving list for all servers with the authority to receive tasks; A task block module, used for processing the tasks uploaded by the cloud computing client in blocks based on the task scheduling strategy, and obtaining a set including a plurality of task blocks; A task scheduling execution module, used to perform task offloading processing on the collection based on the task scheduling receiving list, and generate task block calculation queues for different types of servers; The task processing result integration module is used to distribute the task block calculation queue to the corresponding server for calculation processing, and summarize and integrate the calculation processing results.

[0010] Optionally, the task information identification module is used to identify all tasks uploaded by different cloud computing clients to the Internet of Things operating system platform to obtain corresponding task information, including: Identify task upload requests from different cloud computing clients to obtain identity information of the cloud computing clients; based on the identity information, determine whether the cloud computing client has the authority to upload tasks to the Internet of Things operating system platform; if the cloud computing client has the authority to upload tasks, perform data identification on the tasks uploaded by the cloud computing client to the Internet of Things operating system platform to obtain corresponding task data structure information and task data content information as the task information; The terminal identification module is used to identify all virtual machines and all servers under the Internet of Things operating system platform, and obtain corresponding virtual machine information and server information, including: Based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, which are used as the virtual machine information and the server information respectively; The scheduling strategy generation module is used to determine the task scheduling strategy for all tasks and the resource scheduling strategy for all virtual machines based on the task information, the virtual machine information and the server information, including: Based on the task information, determine the scheduling order and scheduling time interval for scheduling all tasks, so as to obtain the task scheduling strategy for all tasks; based on the virtual machine information and the server information, determine the mode in which the virtual machine allocates its own computing resources and / or memory resources to the server, and use this as the resource scheduling strategy.

[0011] Optionally, the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform is monitored for recognition quality, including: Extracting the amount of data corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform; Extract the data corresponding to each task to identify the response time; Compare the data recognition response time corresponding to the task with a preset data recognition response time threshold; When the data recognition response time corresponding to the task exceeds the preset data recognition response time threshold, the data volume value of the task corresponding to the data recognition response time exceeding the preset data recognition response time threshold is retrieved; The data volume value of the task corresponding to the data recognition response time that exceeds the preset data recognition response time threshold is used as the observed data volume; The data recognition response coefficient is obtained by combining the observed data volume with the data volume corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform and the data recognition response time corresponding to each task; The data recognition response coefficient is obtained by the following formula: Wherein, E represents the data recognition response coefficient; n represents the number of tasks uploaded by the cloud computing client to the IoT operating system platform; C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T grepresents the data recognition response time corresponding to the observed data volume; k represents the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Where k represents the adjustment coefficient C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g Indicates the data recognition response time corresponding to the observed data volume; ΔT indicates the theoretical change in data response time corresponding to each preset unit change in data volume; comparing the data identification response coefficient with a preset coefficient threshold; When the data recognition response coefficient exceeds a preset coefficient threshold, it is determined that there is a risk of abnormal operation in the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform, and a risk warning is issued.

[0012] Optionally, the resource scheduling execution module is used to schedule the resources under the virtual machine to the corresponding server based on the resource scheduling policy, including: Based on the resource scheduling strategy, select matching computing power resources and / or memory resources from the computing power resource library and / or memory resource library under the virtual machine; and based on the connection link between the virtual machine and the server for resource scheduling within the Internet of Things, schedule the selected matching computing power resources and / or memory resources to the corresponding server; The task scheduling determination module is used to determine all servers with the authority to receive tasks based on the respective working states of all servers scheduled with resources, and generate a task scheduling receiving list for all servers with the authority to receive tasks, including: Analyze the work logs of all servers with resources scheduled to obtain the workload of tasks to be processed of all servers with resources scheduled to obtain the workload of tasks to be processed; based on the workload of tasks to be processed, determine whether the server is in an overloaded working state; if so, determine that the server has the authority to receive tasks; if not, determine that the server does not have the authority to receive tasks; Based on the order of the workload of the tasks to be processed of all the servers with the authority to receive the tasks from small to large, a task scheduling receiving list of all the servers with the authority to receive the tasks is generated.

[0013] Optionally, the task block module is used to perform block processing on the task uploaded by the cloud computing client based on the task scheduling strategy to obtain a set of several task blocks, including: Based on the task scheduling strategy, extract corresponding tasks from all tasks uploaded to the Internet of Things operating system platform in a matching scheduling time interval, and based on the task data structure information of the extracted tasks, block the tasks to obtain a set including a plurality of task blocks; The task scheduling execution module is used to perform task offloading processing on the collection based on the task scheduling receiving list, and generate task block calculation queues for different types of servers, including: Based on the task scheduling receiving list, a cloud server and an edge server connected to the collection are selected from all servers with the authority to receive tasks, and the collection is subjected to task offloading processing to generate a task block cloud computing queue and a task block edge computing queue for the cloud server and the edge server respectively; The task processing result integration module is used to distribute the task block calculation queue to the corresponding server for calculation processing, and summarize and integrate the calculation processing results, including: The task block cloud computing queue and the task block edge computing queue are respectively allocated to the cloud server and the edge server for computing and processing, and the computing and processing results of the edge server are transmitted to the cloud server, so that all computing and processing results are summarized and integrated on the cloud server side to obtain a complete processing result matching the task.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The task scheduling method and system provided in the present application for application in the Internet of Things scenario identify the tasks uploaded by the cloud computer user end, the virtual machines and servers under the Internet of Things operating system platform, and obtain task information, virtual machine information and server information, so as to determine the task scheduling strategy and resource scheduling strategy, and provide a reliable basis for the scheduling and allocation of tasks and resources; based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server, and a task scheduling reception list for all servers is generated to achieve the optimal configuration of different types of resources in various task scheduling scenarios; based on the task scheduling strategy, the uploaded tasks are divided into a collection containing several task blocks, and based on the task scheduling reception list, the collection is task unloaded, a task block calculation queue for different types of servers is generated, and the calculation processing results are summarized and integrated, and each task block is sent to the cloud-edge-end queue of the Internet of Things through the task offloading mechanism, thereby improving the resource utilization and performance of the Internet of Things, intelligently scheduling tasks in a distributed architecture under complex business scenarios, optimizing the execution efficiency of the Internet of Things, and improving the operational reliability of the Internet of Things. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them: Figure 1 A flowchart of a task scheduling method applied to an Internet of Things scenario provided by the present invention.

[0016] Figure 2 This is a schematic diagram of the structure of a task scheduling system applied to an Internet of Things scenario provided by the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0018] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0019] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0020] See also Figure 1 As shown, an embodiment of the present application provides a task scheduling method applied to an Internet of Things scenario. The task scheduling method applied to an Internet of Things scenario includes: Identify all tasks uploaded to the IoT operating system platform by different cloud computing clients to obtain corresponding task information; identify all virtual machines and all servers under the IoT operating system platform to obtain corresponding virtual machine information and server information; determine task scheduling strategies for all tasks and resource scheduling strategies for all virtual machines based on the task information, the virtual machine information and the server information; Based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server; and based on the working status of all the servers to which the resources are scheduled, all the servers with the authority to receive the task are determined, and a task scheduling receiving list for all the servers with the authority to receive the task is generated; Based on the task scheduling strategy, the tasks uploaded by the cloud computing user are processed in blocks to obtain a collection of several task blocks; based on the task scheduling receiving list, the collection is task unloaded to generate task block calculation queues for different types of servers; then the task block calculation queues are allocated to the corresponding servers for calculation processing, and the calculation processing results are summarized and integrated.

[0021] The beneficial effects of the above embodiments are as follows: the task scheduling method applied to the Internet of Things scenario identifies the tasks uploaded by the cloud computer user end, the virtual machines and servers under the Internet of Things operating system platform, and obtains task information, virtual machine information and server information, thereby determining the task scheduling strategy and resource scheduling strategy, and providing a reliable basis for the scheduling and allocation of tasks and resources; based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server, and a task scheduling reception list for all servers is generated to achieve the optimal configuration of different types of resources in various task scheduling scenarios; based on the task scheduling strategy, the uploaded tasks are divided into a collection containing several task blocks, and based on the task scheduling reception list, the collection is task unloaded, a task block calculation queue for different types of servers is generated, and the calculation processing results are summarized and integrated, and each task block is sent to the cloud-edge-end queue of the Internet of Things through the task offloading mechanism, thereby improving the resource utilization and performance of the Internet of Things, intelligently scheduling tasks in a distributed architecture under complex business scenarios, optimizing the execution efficiency of the Internet of Things, and improving the operation reliability of the Internet of Things.

[0022] In another embodiment, all tasks uploaded by different cloud computing clients to the Internet of Things operating system platform are identified to obtain corresponding task information; all virtual machines and all servers under the Internet of Things operating system platform are identified to obtain corresponding virtual machine information and server information; based on the task information, the virtual machine information and the server information, a task scheduling strategy for all tasks and a resource scheduling strategy for all virtual machines are determined, including: Identify task upload requests from different cloud computing clients to obtain the identity information of the cloud computing client; based on the identity information, determine whether the cloud computing client has the authority to upload tasks to the IoT operating system platform; if the cloud computing client has the authority to upload tasks, perform data identification on the tasks uploaded by the cloud computing client to the IoT operating system platform to obtain corresponding task data structure information and task data content information as the task information; Based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, which are used as the virtual machine information and the server information respectively; Based on the task information, determine the scheduling order and scheduling time interval for scheduling all tasks, so as to obtain the task scheduling strategy for all tasks; based on the virtual machine information and the server information, determine the mode in which the virtual machine allocates its own computing resources and / or memory resources to the server, and use this as the resource scheduling strategy.

[0023] The beneficial effect of the above embodiment is that the Internet of Things operating system platform is connected to multiple virtual machines, multiple cloud servers and multiple edge servers, and can match and process tasks uploaded by different cloud computing user terminals, ensuring that the uploaded tasks can be accurately and efficiently calculated and processed in a timely manner. In actual work, if all the tasks uploaded by cloud computing user terminals are centrally scheduled and assigned to several designated virtual machines, cloud servers and edge servers, the workload of the corresponding virtual machines, cloud servers and edge servers will be increased. Therefore, it is necessary to flexibly and effectively schedule tasks and resources for all virtual machines, cloud servers and edge servers under the Internet of Things operating system platform for different uploaded tasks, ensuring that all resources of the Internet of Things operating system platform can be efficiently and accurately utilized. Specifically, the task upload request from different cloud computing user terminals is identified, and the identity information of the cloud computing user terminal is obtained. If the identity information exists in the preset identity information list, it is determined that the cloud computing user terminal has the authority to upload the task. Otherwise, it is determined that the cloud computing user terminal does not have the authority to upload the task. Then, the cloud computing user with the authority to upload the task is uploaded to the Internet of Things operating system platform. Data identification is performed, and the corresponding task data structure information and task data content information are obtained, so that the data attributes of the uploaded task can be accurately calibrated. In addition, based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, providing a reliable and accurate basis for subsequent resource scheduling and task scheduling. Based on the task information, the scheduling order and scheduling time interval for scheduling all tasks are determined, so as to obtain the task scheduling strategy for all tasks, so that it is possible to quantitatively determine when to schedule the tasks uploaded by the cloud computing user end; and based on the virtual machine information and the server information, the mode of the virtual machine scheduling and allocating its own computing resources and / or memory resources to the server is determined, and this is used as the resource scheduling strategy, so that the computing resources and memory resources in the virtual machine can be accurately and quantitatively counted, ensuring the effective scheduling and utilization of computing resources and memory resources, and avoiding the maximum scheduling and utilization of computing resources and memory resources.

[0024] In another embodiment, the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform is monitored for recognition quality, including: Extracting the amount of data corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform; Extract the data corresponding to each task to identify the response time; Compare the data recognition response time corresponding to the task with a preset data recognition response time threshold; When the data recognition response time corresponding to the task exceeds the preset data recognition response time threshold, the data volume value of the task corresponding to the data recognition response time exceeding the preset data recognition response time threshold is retrieved; The data volume value of the task corresponding to the data recognition response time that exceeds the preset data recognition response time threshold is used as the observed data volume; The data recognition response coefficient is obtained by combining the observed data volume with the data volume corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform and the data recognition response time corresponding to each task; The data recognition response coefficient is obtained by the following formula: Wherein, E represents the data recognition response coefficient; n represents the number of tasks uploaded by the cloud computing client to the IoT operating system platform; C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g represents the data recognition response time corresponding to the observed data volume; k represents the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Where k represents the adjustment coefficient C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g Indicates the data recognition response time corresponding to the observed data volume; ΔT indicates the theoretical change in data response time corresponding to each preset unit change in data volume; comparing the data identification response coefficient with a preset coefficient threshold; When the data recognition response coefficient exceeds a preset coefficient threshold, it is determined that there is a risk of abnormal operation in the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform, and a risk warning is issued.

[0025] The beneficial effect of the above embodiment is that the solution can timely discover abnormalities in the data recognition process by real-time monitoring the data recognition process of tasks uploaded by the cloud computing user end to the Internet of Things operating system platform. When the data recognition response time exceeds the preset threshold or the data recognition response coefficient exceeds the preset coefficient threshold, the system will issue a risk warning, thereby effectively avoiding or reducing system performance degradation or failure caused by data recognition delays or errors.

[0026] Accurately evaluate data recognition efficiency: By extracting the data volume and data recognition response time corresponding to each task and calculating the data recognition response coefficient, the solution can accurately evaluate the efficiency of data recognition. This helps system administrators or developers understand the current operating status of the system so as to make targeted optimization and adjustments. The solution introduces an adjustment coefficient k, which is dynamically calculated based on the data volume of the task, the data recognition response time, and the observed data volume and its corresponding response time. This design enables the system to automatically adjust the evaluation criteria according to the current task load and data characteristics, improving the adaptability and flexibility of the system. Through real-time monitoring and evaluation of the data recognition process, the system can allocate resources more reasonably. By timely warning and optimizing the data recognition process, the solution can reduce user waiting time, improve the response speed and stability of the system, and thus enhance the overall user experience.

[0027] On the other hand, by real-time monitoring the data recognition process of tasks uploaded from the cloud computing user end to the IoT operating system platform, the system can quickly respond to the needs of data recognition and reduce recognition delays. When the data recognition response time exceeds the preset threshold, the system can immediately identify and process it to avoid bottlenecks and congestion in the recognition process. The calculation of the data recognition response coefficient and the adjustment coefficient mentioned in the scheme helps the system to more accurately evaluate the efficiency of data recognition, thereby optimizing the data processing process. By dynamically adjusting resource allocation, the system can ensure the efficient operation of the data recognition process. When the data recognition response coefficient exceeds the preset coefficient threshold, the system will issue a risk warning to prompt the system administrator or developer to deal with the abnormal situation in time. This early warning mechanism helps to timely discover and solve potential system failures and improve the stability and reliability of the system. The real-time monitoring and evaluation mechanism in the scheme enables the system to allocate and manage resources more reasonably. By avoiding resource overload or idleness, the system can maintain efficient operation and improve overall performance. At the same time, the adjustment coefficient k introduced in the scheme enables the system to automatically adjust the evaluation criteria according to the current task load and data characteristics. This dynamic adjustment mechanism helps the system adapt to different application scenarios and user needs and enhances the scalability and flexibility of the system. By optimizing the data recognition and processing process, the system can support larger-scale data processing needs. This is critical for IoT operating system platforms because with the continuous increase in IoT devices and the continuous accumulation of data, the demand for large-scale data processing capabilities is also increasing. In addition, by improving data recognition efficiency and system stability, the system can reduce user waiting time and improve response speed. This is critical to improving user experience because users usually expect the system to respond to their needs quickly. By enhancing system stability and scalability, the system can maintain efficient operation and reduce failures and downtime. This helps to improve system availability and ensure that users can access and use the system at any time.

[0028] In summary, the technical effects of the above technical solutions in terms of performance indicators are mainly reflected in improving data recognition efficiency, improving system stability, enhancing system scalability, and optimizing user experience. These technical effects together constitute the significant advantages of the technical solution in terms of performance indicators. At the same time, the technical solution effectively improves the data recognition efficiency and system stability of the Internet of Things operating system platform through real-time monitoring, precise evaluation, dynamic adjustment and resource optimization, reduces the risk of abnormal operation, and is of great significance for ensuring the normal operation of the system and improving user experience.

[0029] In another embodiment, based on the resource scheduling policy, resources under the virtual machine are scheduled to corresponding servers; and based on the working status of all servers to which resources are scheduled, all servers with the authority to receive tasks are determined, and a task scheduling receiving list for all servers with the authority to receive tasks is generated, including: Based on the resource scheduling strategy, select matching computing power resources and / or memory resources from the computing power resource library and / or memory resource library under the virtual machine; and based on the connection link between the virtual machine and the server for resource scheduling within the Internet of Things, schedule the selected matching computing power resources and / or memory resources to the corresponding server; Analyze the work logs of all servers with resources scheduled to obtain the workload of tasks to be processed of all servers with resources scheduled to obtain the workload of tasks to be processed; based on the workload of tasks to be processed, determine whether the server is in an overloaded working state; if so, determine that the server has the authority to receive tasks; if not, determine that the server does not have the authority to receive tasks; Based on the order of the workload of the tasks to be processed of all the servers with the authority to receive the tasks from small to large, a task scheduling receiving list of all the servers with the authority to receive the tasks is generated.

[0030] The beneficial effect of the above embodiment is that, based on the resource scheduling strategy, the computing power resource library and / or memory resource library under the virtual machine are identified and screened for resources, and computing power resources and / or memory resources that are in an idle state and have matching resource quantities are obtained, and based on the connection link between the virtual machine and the server that performs resource scheduling within the Internet of Things, the selected matching computing power resources and / or memory resources are scheduled to the corresponding server, so as to ensure the control and utilization authority of the server to the scheduled computing power resources and / or memory resources. In addition, the work logs of all servers with resources scheduled are analyzed to obtain the workload of pending tasks of all servers with resources scheduled; based on the workload of pending tasks, it is judged whether the server is in an overloaded working state; if so, it is judged that the server has the authority to receive tasks; if not, it is judged that the server does not have the authority to receive tasks, so that it can be ensured that the server can technically process the tasks assigned by the scheduling and avoid the situation of delay in task processing. Based on the order of the workload of pending tasks of all servers with the authority to receive tasks from small to large, a task scheduling reception list of all servers with the authority to receive tasks is generated, thereby providing a reliable scheduling basis for scheduling tasks to different servers.

[0031] In another embodiment, based on the task scheduling strategy, the task uploaded by the cloud computing client is processed in blocks to obtain a set of several task blocks; based on the task scheduling receiving list, the set is processed for task offloading to generate task block calculation queues for different types of servers; the task block calculation queues are then allocated to corresponding servers for calculation processing, and the calculation processing results are summarized and integrated, including: Based on the task scheduling strategy, extract corresponding tasks from all tasks uploaded to the IoT operating system platform in a matching scheduling time interval, and based on task data structure information of the extracted tasks, perform block processing on the tasks to obtain a set including a plurality of task blocks; Based on the task scheduling receiving list, the cloud server and edge server connected to the collection are selected from all servers with the authority to receive tasks, and the collection is subjected to task offloading processing to generate a task block cloud computing queue and a task block edge computing queue for the cloud server and the edge server respectively; the task block cloud computing queue and the task block edge computing queue are respectively allocated to the cloud server and the edge server for computing processing, and the computing processing result of the edge server is transmitted to the cloud server, so that all computing processing results are summarized and integrated on the cloud server side to obtain a complete processing result matching the task.

[0032] The beneficial effect of the above embodiment is that, based on the task scheduling strategy, the corresponding task is extracted from all tasks uploaded to the Internet of Things operating system platform in the matching scheduling time interval, and based on the task data structure information of the extracted task, the task is processed in blocks to obtain a set containing several task blocks. By dividing the task slices into several task blocks, it is possible to avoid the situation where a single task is processed as a whole and the processing is stuck. Also based on the task scheduling receiving list, a cloud server and an edge server connected to the set are selected from all servers with the authority to receive tasks, and the set is task unloaded to generate a task block cloud computing queue and a task block edge computing queue for the cloud server and the edge server respectively; then the task block cloud computing queue and the task block edge computing queue are respectively assigned to the cloud server and the edge server for computing and processing, and two different types of servers, cloud servers and edge servers, are used to process the task blocks under the corresponding computing queues, which can optimize the task processing performance of the server, realize the cloud-edge-end mode processing of the task blocks, ensure the intelligent scheduling of tasks in the distributed architecture in complex business scenarios, and optimize the execution efficiency of the Internet of Things.

[0033] See also Figure 2 As shown, an embodiment of the present application provides a task scheduling system applied to an Internet of Things scenario. The task scheduling system applied to an Internet of Things scenario includes: The task information identification module is used to identify all tasks uploaded to the IoT operating system platform by different cloud computing user terminals and obtain corresponding task information; The terminal identification module is used to identify all virtual machines and all servers under the IoT operating system platform and obtain corresponding virtual machine information and server information; A scheduling strategy generation module, used to determine a task scheduling strategy for all tasks and a resource scheduling strategy for all virtual machines based on the task information, the virtual machine information and the server information; A resource scheduling execution module is used to schedule the resources under the virtual machine to the corresponding server based on the resource scheduling policy; The task scheduling determination module is used to determine all servers with the authority to receive tasks based on the respective working states of all servers with resources scheduled, and generate a task scheduling receiving list for all servers with the authority to receive tasks; A task block module is used to process the task uploaded by the cloud computing client in blocks based on the task scheduling strategy to obtain a set of several task blocks; A task scheduling execution module is used to perform task offloading processing on the collection based on the task scheduling receiving list, and generate task block computing queues for different types of servers; The task processing result integration module is used to distribute the task block calculation queue to the corresponding server for calculation processing, and summarize and integrate the calculation processing results.

[0034] The beneficial effects of the above embodiments are as follows: the task scheduling system applied to the Internet of Things scenario identifies the tasks uploaded by the cloud computer user end, the virtual machines and servers under the Internet of Things operating system platform, and obtains task information, virtual machine information and server information, thereby determining the task scheduling strategy and resource scheduling strategy, and providing a reliable basis for the scheduling and allocation of tasks and resources; based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server, and a task scheduling reception list for all servers is generated to achieve the optimal configuration of different types of resources in various task scheduling scenarios; based on the task scheduling strategy, the uploaded tasks are divided into a collection containing several task blocks, and based on the task scheduling reception list, the collection is task unloaded, a task block calculation queue for different types of servers is generated, and the calculation processing results are summarized and integrated, and each task block is sent to the cloud-edge-end queue of the Internet of Things through the task offloading mechanism, thereby improving the resource utilization and performance of the Internet of Things, intelligently scheduling tasks in a distributed architecture under complex business scenarios, optimizing the execution efficiency of the Internet of Things, and improving the operation reliability of the Internet of Things.

[0035] In another embodiment, the task information identification module is used to identify all tasks uploaded to the Internet of Things operating system platform by different cloud computing clients to obtain corresponding task information, including: Identify task upload requests from different cloud computing clients to obtain the identity information of the cloud computing client; based on the identity information, determine whether the cloud computing client has the authority to upload tasks to the IoT operating system platform; if the cloud computing client has the authority to upload tasks, perform data identification on the tasks uploaded by the cloud computing client to the IoT operating system platform to obtain corresponding task data structure information and task data content information as the task information; The terminal identification module is used to identify all virtual machines and all servers under the IoT operating system platform, and obtain corresponding virtual machine information and server information, including: Based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, which are used as the virtual machine information and the server information respectively; The scheduling strategy generation module is used to determine the task scheduling strategy for all tasks and the resource scheduling strategy for all virtual machines based on the task information, the virtual machine information and the server information, including: Based on the task information, determine the scheduling order and scheduling time interval for scheduling all tasks, so as to obtain the task scheduling strategy for all tasks; based on the virtual machine information and the server information, determine the mode in which the virtual machine allocates its own computing resources and / or memory resources to the server, and use this as the resource scheduling strategy.

[0036] The beneficial effect of the above embodiment is that the Internet of Things operating system platform is connected to multiple virtual machines, multiple cloud servers and multiple edge servers, and can match and process tasks uploaded by different cloud computing user terminals, ensuring that the uploaded tasks can be accurately and efficiently calculated and processed in a timely manner. In actual work, if all the tasks uploaded by cloud computing user terminals are centrally scheduled and assigned to several designated virtual machines, cloud servers and edge servers, the workload of the corresponding virtual machines, cloud servers and edge servers will be increased. Therefore, it is necessary to flexibly and effectively schedule tasks and resources for all virtual machines, cloud servers and edge servers under the Internet of Things operating system platform for different uploaded tasks, ensuring that all resources of the Internet of Things operating system platform can be efficiently and accurately utilized. Specifically, the task upload request from different cloud computing user terminals is identified, and the identity information of the cloud computing user terminal is obtained. If the identity information exists in the preset identity information list, it is determined that the cloud computing user terminal has the authority to upload the task. Otherwise, it is determined that the cloud computing user terminal does not have the authority to upload the task. Then, the cloud computing user with the authority to upload the task is uploaded to the Internet of Things operating system platform. Data identification is performed, and the corresponding task data structure information and task data content information are obtained, so that the data attributes of the uploaded task can be accurately calibrated. In addition, based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, providing a reliable and accurate basis for subsequent resource scheduling and task scheduling. Based on the task information, the scheduling order and scheduling time interval for scheduling all tasks are determined, so as to obtain the task scheduling strategy for all tasks, so that it is possible to quantitatively determine when to schedule the tasks uploaded by the cloud computing user end; and based on the virtual machine information and the server information, the mode of the virtual machine scheduling and allocating its own computing resources and / or memory resources to the server is determined, and this is used as the resource scheduling strategy, so that the computing resources and memory resources in the virtual machine can be accurately and quantitatively counted, ensuring the effective scheduling and utilization of computing resources and memory resources, and avoiding the maximum scheduling and utilization of computing resources and memory resources.

[0037] In another embodiment, the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform is monitored for recognition quality, including: Extracting the amount of data corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform; Extract the data corresponding to each task to identify the response time; Compare the data recognition response time corresponding to the task with a preset data recognition response time threshold; When the data recognition response time corresponding to the task exceeds the preset data recognition response time threshold, the data volume value of the task corresponding to the data recognition response time exceeding the preset data recognition response time threshold is retrieved; The data volume value of the task corresponding to the data recognition response time that exceeds the preset data recognition response time threshold is used as the observed data volume; The data recognition response coefficient is obtained by combining the observed data volume with the data volume corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform and the data recognition response time corresponding to each task; The data recognition response coefficient is obtained by the following formula: Wherein, E represents the data recognition response coefficient; n represents the number of tasks uploaded by the cloud computing client to the IoT operating system platform; C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g represents the data recognition response time corresponding to the observed data volume; k represents the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Where k represents the adjustment coefficient C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g Indicates the data recognition response time corresponding to the observed data volume; ΔT indicates the theoretical change in data response time corresponding to each preset unit change in data volume; comparing the data identification response coefficient with a preset coefficient threshold; When the data recognition response coefficient exceeds a preset coefficient threshold, it is determined that there is a risk of abnormal operation in the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform, and a risk warning is issued.

[0038] The beneficial effect of the above embodiment is that the solution can timely discover abnormalities in the data recognition process by real-time monitoring the data recognition process of tasks uploaded by the cloud computing user end to the Internet of Things operating system platform. When the data recognition response time exceeds the preset threshold or the data recognition response coefficient exceeds the preset coefficient threshold, the system will issue a risk warning, thereby effectively avoiding or reducing system performance degradation or failure caused by data recognition delays or errors.

[0039] Accurately evaluate data recognition efficiency: By extracting the data volume and data recognition response time corresponding to each task and calculating the data recognition response coefficient, the solution can accurately evaluate the efficiency of data recognition. This helps system administrators or developers understand the current operating status of the system so as to make targeted optimization and adjustments. The solution introduces an adjustment coefficient k, which is dynamically calculated based on the data volume of the task, the data recognition response time, and the observed data volume and its corresponding response time. This design enables the system to automatically adjust the evaluation criteria according to the current task load and data characteristics, improving the adaptability and flexibility of the system. Through real-time monitoring and evaluation of the data recognition process, the system can allocate resources more reasonably. By timely warning and optimizing the data recognition process, the solution can reduce user waiting time, improve the response speed and stability of the system, and thus enhance the overall user experience.

[0040] On the other hand, by real-time monitoring the data recognition process of tasks uploaded from the cloud computing user end to the IoT operating system platform, the system can quickly respond to the needs of data recognition and reduce recognition delays. When the data recognition response time exceeds the preset threshold, the system can immediately identify and process it to avoid bottlenecks and congestion in the recognition process. The calculation of the data recognition response coefficient and the adjustment coefficient mentioned in the scheme helps the system to more accurately evaluate the efficiency of data recognition, thereby optimizing the data processing process. By dynamically adjusting resource allocation, the system can ensure the efficient operation of the data recognition process. When the data recognition response coefficient exceeds the preset coefficient threshold, the system will issue a risk warning to prompt the system administrator or developer to deal with the abnormal situation in time. This early warning mechanism helps to timely discover and solve potential system failures and improve the stability and reliability of the system. The real-time monitoring and evaluation mechanism in the scheme enables the system to allocate and manage resources more reasonably. By avoiding resource overload or idleness, the system can maintain efficient operation and improve overall performance. At the same time, the adjustment coefficient k introduced in the scheme enables the system to automatically adjust the evaluation criteria according to the current task load and data characteristics. This dynamic adjustment mechanism helps the system adapt to different application scenarios and user needs and enhances the scalability and flexibility of the system. By optimizing the data recognition and processing process, the system can support larger-scale data processing needs. This is critical for IoT operating system platforms because with the continuous increase in IoT devices and the continuous accumulation of data, the demand for large-scale data processing capabilities is also increasing. In addition, by improving data recognition efficiency and system stability, the system can reduce user waiting time and improve response speed. This is critical to improving user experience because users usually expect the system to respond to their needs quickly. By enhancing system stability and scalability, the system can maintain efficient operation and reduce failures and downtime. This helps to improve system availability and ensure that users can access and use the system at any time.

[0041] In summary, the technical effects of the above technical solutions in terms of performance indicators are mainly reflected in improving data recognition efficiency, improving system stability, enhancing system scalability, and optimizing user experience. These technical effects together constitute the significant advantages of the technical solution in terms of performance indicators. At the same time, the technical solution effectively improves the data recognition efficiency and system stability of the Internet of Things operating system platform through real-time monitoring, precise evaluation, dynamic adjustment and resource optimization, reduces the risk of abnormal operation, and is of great significance for ensuring the normal operation of the system and improving user experience.

[0042] In another embodiment, the resource scheduling execution module is used to schedule the resources under the virtual machine to the corresponding server based on the resource scheduling policy, including: Based on the resource scheduling strategy, select matching computing power resources and / or memory resources from the computing power resource library and / or memory resource library under the virtual machine; and based on the connection link between the virtual machine and the server for resource scheduling within the Internet of Things, schedule the selected matching computing power resources and / or memory resources to the corresponding server; The task scheduling determination module is used to determine all servers with the authority to receive tasks based on the respective working states of all servers with resources scheduled, and generate a task scheduling receiving list for all servers with the authority to receive tasks, including: Analyze the work logs of all servers with resources scheduled to obtain the workload of tasks to be processed of all servers with resources scheduled to obtain the workload of tasks to be processed; based on the workload of tasks to be processed, determine whether the server is in an overloaded working state; if so, determine that the server has the authority to receive tasks; if not, determine that the server does not have the authority to receive tasks; Based on the order of the workload of the tasks to be processed of all the servers with the authority to receive the tasks from small to large, a task scheduling receiving list of all the servers with the authority to receive the tasks is generated.

[0043] The beneficial effect of the above embodiment is that, based on the resource scheduling strategy, the computing power resource library and / or memory resource library under the virtual machine are identified and screened for resources, and computing power resources and / or memory resources that are in an idle state and have matching resource quantities are obtained, and based on the connection link between the virtual machine and the server that performs resource scheduling within the Internet of Things, the selected matching computing power resources and / or memory resources are scheduled to the corresponding server, so as to ensure the control and utilization authority of the server to the scheduled computing power resources and / or memory resources. In addition, the work logs of all servers with resources scheduled are analyzed to obtain the workload of pending tasks of all servers with resources scheduled; based on the workload of pending tasks, it is judged whether the server is in an overloaded working state; if so, it is judged that the server has the authority to receive tasks; if not, it is judged that the server does not have the authority to receive tasks, so that it can be ensured that the server can technically process the tasks assigned by the scheduling and avoid the situation of delay in task processing. Based on the order of the workload of pending tasks of all servers with the authority to receive tasks from small to large, a task scheduling reception list of all servers with the authority to receive tasks is generated, thereby providing a reliable scheduling basis for scheduling tasks to different servers.

[0044] In another embodiment, the task block module is used to block the task uploaded by the cloud computing client based on the task scheduling strategy to obtain a set of several task blocks, including: Based on the task scheduling strategy, extract corresponding tasks from all tasks uploaded to the IoT operating system platform in a matching scheduling time interval, and based on task data structure information of the extracted tasks, perform block processing on the tasks to obtain a set including a plurality of task blocks; The task scheduling execution module is used to perform task offloading processing on the collection based on the task scheduling receiving list, and generate task block calculation queues for different types of servers, including: Based on the task scheduling receiving list, a cloud server and an edge server connected to the collection are selected from all servers with the authority to receive tasks, and the collection is subjected to task offloading processing to generate a task block cloud computing queue and a task block edge computing queue for the cloud server and the edge server respectively; The task processing result integration module is used to distribute the task block calculation queue to the corresponding server for calculation processing, and summarize and integrate the calculation processing results, including: The task block cloud computing queue and the task block edge computing queue are respectively allocated to the cloud server and the edge server for computing and processing, and the computing and processing results of the edge server are transmitted to the cloud server, so that all computing and processing results are summarized and integrated on the cloud server side to obtain a complete processing result matching the task.

[0045] The beneficial effect of the above embodiment is that, based on the task scheduling strategy, the corresponding task is extracted from all tasks uploaded to the Internet of Things operating system platform in the matching scheduling time interval, and based on the task data structure information of the extracted task, the task is processed in blocks to obtain a set containing several task blocks. By dividing the task slices into several task blocks, it is possible to avoid the situation where a single task is processed as a whole and the processing is stuck. Also based on the task scheduling receiving list, a cloud server and an edge server connected to the set are selected from all servers with the authority to receive tasks, and the set is task unloaded to generate a task block cloud computing queue and a task block edge computing queue for the cloud server and the edge server respectively; then the task block cloud computing queue and the task block edge computing queue are respectively assigned to the cloud server and the edge server for computing and processing, and two different types of servers, cloud servers and edge servers, are used to process the task blocks under the corresponding computing queues, which can optimize the task processing performance of the server, realize the cloud-edge-end mode processing of the task blocks, ensure the intelligent scheduling of tasks in the distributed architecture in complex business scenarios, and optimize the execution efficiency of the Internet of Things.

[0046] In general, the task scheduling method and system applied to the Internet of Things scenario identify the tasks uploaded by the cloud computer user end, the virtual machines and servers under the Internet of Things operating system platform, and obtain task information, virtual machine information and server information, so as to determine the task scheduling strategy and resource scheduling strategy, and provide a reliable basis for the scheduling and allocation of tasks and resources; based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server, and a task scheduling reception list for all servers is generated to achieve the optimal configuration of different types of resources in various task scheduling scenarios; based on the task scheduling strategy, the uploaded tasks are divided into a collection containing several task blocks, and based on the task scheduling reception list, the collection is task unloaded, and a task block calculation queue for different types of servers is generated, and the calculation processing results are summarized and integrated, and each task block is sent to the cloud-edge-end queue of the Internet of Things through the task offloading mechanism, thereby improving the resource utilization and performance of the Internet of Things, intelligently scheduling tasks in a distributed architecture under complex business scenarios, optimizing the execution efficiency of the Internet of Things, and improving the operational reliability of the Internet of Things.

[0047] The above is only a specific implementation of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the protection scope of the present invention.

Claims

1. A task scheduling method applied to an Internet of Things scenario, characterized in that: include: Identify all tasks uploaded to the IoT operating system platform by different cloud computing clients and obtain corresponding task information; Identify all virtual machines and all servers under the IoT operating system platform to obtain corresponding virtual machine information and server information; determine task scheduling strategies for all tasks and resource scheduling strategies for all virtual machines based on the task information, the virtual machine information and the server information; Based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server; and based on the working status of all the servers to which the resources are scheduled, all the servers with the authority to receive the task are determined, and a task scheduling receiving list for all the servers with the authority to receive the task is generated; Based on the task scheduling strategy, the task uploaded by the cloud computing client is processed in blocks to obtain a set including a plurality of task blocks; Based on the task scheduling receiving list, the collection is task unloaded to generate task block calculation queues for different types of servers; the task block calculation queues are then allocated to corresponding servers for calculation processing, and the calculation processing results are summarized and integrated.

2. The task scheduling method applied to the Internet of Things scenario as claimed in claim 1, characterized in that: Identify all tasks uploaded to the IoT operating system platform by different cloud computing clients to obtain corresponding task information; identify all virtual machines and all servers under the IoT operating system platform to obtain corresponding virtual machine information and server information; Determining a task scheduling strategy for all tasks and a resource scheduling strategy for all virtual machines based on the task information, the virtual machine information, and the server information includes: Identify task upload requests from different cloud computing clients to obtain identity information of the cloud computing clients; based on the identity information, determine whether the cloud computing client has the authority to upload tasks to the Internet of Things operating system platform; if the cloud computing client has the authority to upload tasks, perform data identification on the tasks uploaded by the cloud computing client to the Internet of Things operating system platform to obtain corresponding task data structure information and task data content information as the task information; Based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, which are used as the virtual machine information and the server information respectively; Based on the task information, determine the scheduling order and scheduling time interval for scheduling all tasks, so as to obtain the task scheduling strategy for all tasks; based on the virtual machine information and the server information, determine the mode in which the virtual machine allocates its own computing resources and / or memory resources to the server, and use this as the resource scheduling strategy.

3. The task scheduling method applied to the Internet of Things scenario as claimed in claim 2 is characterized in that: Performing recognition quality monitoring on the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform includes: Extracting the amount of data corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform; Extract the data corresponding to each task to identify the response time; Compare the data recognition response time corresponding to the task with a preset data recognition response time threshold; When the data recognition response time corresponding to the task exceeds the preset data recognition response time threshold, the data volume value of the task corresponding to the data recognition response time exceeding the preset data recognition response time threshold is retrieved; The data volume value of the task corresponding to the data recognition response time that exceeds the preset data recognition response time threshold is used as the observed data volume; The data recognition response coefficient is obtained by combining the observed data volume with the data volume corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform and the data recognition response time corresponding to each task; The data recognition response coefficient is obtained by the following formula: Wherein, E represents the data recognition response coefficient; n represents the number of tasks uploaded by the cloud computing client to the IoT operating system platform; C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g represents the data recognition response time corresponding to the observed data volume; k represents the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Where k represents the adjustment coefficient C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g Indicates the data recognition response time corresponding to the observed data volume; ΔT indicates the theoretical change in data response time corresponding to each preset unit change in data volume; comparing the data identification response coefficient with a preset coefficient threshold; When the data recognition response coefficient exceeds a preset coefficient threshold, it is determined that there is a risk of abnormal operation in the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform, and a risk warning is issued.

4. The task scheduling method applied to the Internet of Things scenario as claimed in claim 2, characterized in that: Based on the resource scheduling strategy, the resources under the virtual machine are scheduled to the corresponding server; and based on the respective working states of all the servers to which the resources are scheduled, all the servers with the authority to receive the task are determined, and a task scheduling receiving list for all the servers with the authority to receive the task is generated, including: Based on the resource scheduling strategy, select matching computing power resources and / or memory resources from the computing power resource library and / or memory resource library under the virtual machine; and based on the connection link between the virtual machine and the server for resource scheduling within the Internet of Things, schedule the selected matching computing power resources and / or memory resources to the corresponding server; Analyze the work logs of all servers with resources scheduled to obtain the workload of tasks to be processed of all servers with resources scheduled to obtain the workload of tasks to be processed; based on the workload of tasks to be processed, determine whether the server is in an overloaded working state; if so, determine that the server has the authority to receive tasks; if not, determine that the server does not have the authority to receive tasks; Based on the order of the workload of the tasks to be processed of all the servers with the authority to receive the tasks from small to large, a task scheduling receiving list of all the servers with the authority to receive the tasks is generated.

5. The task scheduling method applied to the Internet of Things scenario as claimed in claim 3 is characterized in that: Based on the task scheduling strategy, the task uploaded by the cloud computing client is processed in blocks to obtain a set including a plurality of task blocks; Based on the task scheduling receiving list, the collection is subjected to task offloading processing to generate task block computing queues for different types of servers; Then the task block calculation queue is allocated to the corresponding server for calculation processing, and the calculation processing results are summarized and integrated, including: Based on the task scheduling strategy, extract corresponding tasks from all tasks uploaded to the Internet of Things operating system platform in a matching scheduling time interval, and based on the task data structure information of the extracted tasks, block the tasks to obtain a set including a plurality of task blocks; Based on the task scheduling receiving list, cloud servers and edge servers connected to the collection are selected from all servers with the authority to receive tasks, and task offloading processing is performed on the collection to generate task block cloud computing queues and task block edge computing queues for the cloud server and the edge server respectively; the task block cloud computing queues and the task block edge computing queues are then respectively allocated to the cloud server and the edge server for computing processing, and the computing processing results of the edge server are transmitted to the cloud server, so that all computing processing results are summarized and integrated on the cloud server side to obtain a complete processing result matching the task.

6. The task scheduling system applied to the Internet of Things scenario is characterized by: include: The task information identification module is used to identify all tasks uploaded to the IoT operating system platform by different cloud computing user terminals and obtain corresponding task information; The terminal identification module is used to identify all virtual machines and all servers under the IoT operating system platform to obtain corresponding virtual machine information and server information; A scheduling strategy generation module, used to determine the task scheduling strategy for all tasks and the resource scheduling strategy for all virtual machines based on the task information, the virtual machine information and the server information; A resource scheduling execution module, used to schedule the resources under the virtual machine to the corresponding server based on the resource scheduling policy; The task scheduling determination module is used to determine all servers with the authority to receive tasks based on the respective working states of all servers with resources scheduled, and generate a task scheduling receiving list for all servers with the authority to receive tasks; A task block module, used for processing the tasks uploaded by the cloud computing client in blocks based on the task scheduling strategy, and obtaining a set including a plurality of task blocks; A task scheduling execution module, used to perform task offloading processing on the collection based on the task scheduling receiving list, and generate task block calculation queues for different types of servers; The task processing result integration module is used to distribute the task block calculation queue to the corresponding server for calculation processing, and summarize and integrate the calculation processing results.

7. The task scheduling system applied to the Internet of Things scenario as claimed in claim 6, characterized in that: The task information identification module is used to identify all tasks uploaded to the IoT operating system platform by different cloud computing clients, and obtain corresponding task information, including: Identify task upload requests from different cloud computing clients to obtain identity information of the cloud computing clients; based on the identity information, determine whether the cloud computing client has the authority to upload tasks to the Internet of Things operating system platform; if the cloud computing client has the authority to upload tasks, perform data identification on the tasks uploaded by the cloud computing client to the Internet of Things operating system platform to obtain corresponding task data structure information and task data content information as the task information; The terminal identification module is used to identify all virtual machines and all servers under the Internet of Things operating system platform, and obtain corresponding virtual machine information and server information, including: Based on the address information of all virtual machines and all servers under the IoT operating system platform within the IoT, all virtual machines and all servers are identified respectively, and the available resource distribution status information of all virtual machines and the real-time task execution status information of all servers are obtained, which are used as the virtual machine information and the server information respectively; The scheduling strategy generation module is used to determine the task scheduling strategy for all tasks and the resource scheduling strategy for all virtual machines based on the task information, the virtual machine information and the server information, including: Based on the task information, determine the scheduling order and scheduling time interval for scheduling all tasks, so as to obtain the task scheduling strategy for all tasks; based on the virtual machine information and the server information, determine the mode in which the virtual machine allocates its own computing resources and / or memory resources to the server, and use this as the resource scheduling strategy.

8. The task scheduling system applied to the Internet of Things scenario as claimed in claim 7, characterized in that: Performing recognition quality monitoring on the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform includes: Extracting the amount of data corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform; Extract the data corresponding to each task to identify the response time; Compare the data recognition response time corresponding to the task with a preset data recognition response time threshold; When the data recognition response time corresponding to the task exceeds the preset data recognition response time threshold, the data volume value of the task corresponding to the data recognition response time exceeding the preset data recognition response time threshold is retrieved; The data volume value of the task corresponding to the data recognition response time that exceeds the preset data recognition response time threshold is used as the observed data volume; The data recognition response coefficient is obtained by combining the observed data volume with the data volume corresponding to each task uploaded by the cloud computing client to the Internet of Things operating system platform and the data recognition response time corresponding to each task; The data recognition response coefficient is obtained by the following formula: Wherein, E represents the data recognition response coefficient; n represents the number of tasks uploaded by the cloud computing client to the IoT operating system platform; C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g represents the data recognition response time corresponding to the observed data volume; k represents the adjustment coefficient, and the adjustment coefficient is obtained by the following formula: Where k represents the adjustment coefficient C i represents the amount of data corresponding to the i-th task uploaded; T i Indicates the data recognition response time corresponding to the i-th task uploaded; C g Indicates the amount of observation data; T g Indicates the data recognition response time corresponding to the observed data volume; ΔT indicates the theoretical change in data response time corresponding to each preset unit change in data volume; comparing the data identification response coefficient with a preset coefficient threshold; When the data recognition response coefficient exceeds a preset coefficient threshold, it is determined that there is a risk of abnormal operation in the data recognition process of the task uploaded by the cloud computing client to the Internet of Things operating system platform, and a risk warning is issued.

9. The task scheduling system applied to the Internet of Things scenario as claimed in claim 8, characterized in that: The resource scheduling execution module is used to schedule the resources under the virtual machine to the corresponding server based on the resource scheduling policy, including: Based on the resource scheduling strategy, select matching computing power resources and / or memory resources from the computing power resource library and / or memory resource library under the virtual machine; and based on the connection link between the virtual machine and the server for resource scheduling within the Internet of Things, schedule the selected matching computing power resources and / or memory resources to the corresponding server; The task scheduling determination module is used to determine all servers with the authority to receive tasks based on the respective working states of all servers scheduled with resources, and generate a task scheduling receiving list for all servers with the authority to receive tasks, including: Analyze the work logs of all servers with resources scheduled to obtain the workload of tasks to be processed of all servers with resources scheduled to obtain the workload of tasks to be processed; based on the workload of tasks to be processed, determine whether the server is in an overloaded working state; if so, determine that the server has the authority to receive tasks; if not, determine that the server does not have the authority to receive tasks; Based on the order of the workload of the tasks to be processed of all the servers with the authority to receive the tasks from small to large, a task scheduling receiving list of all the servers with the authority to receive the tasks is generated.

10. The task scheduling system applied to the Internet of Things scenario as claimed in claim 9, characterized in that: The task block module is used to process the task uploaded by the cloud computing client in blocks based on the task scheduling strategy to obtain a set of several task blocks, including: Based on the task scheduling strategy, extract corresponding tasks from all tasks uploaded to the Internet of Things operating system platform in a matching scheduling time interval, and based on the task data structure information of the extracted tasks, block the tasks to obtain a set including a plurality of task blocks; The task scheduling execution module is used to perform task offloading processing on the collection based on the task scheduling receiving list, and generate task block calculation queues for different types of servers, including: Based on the task scheduling receiving list, a cloud server and an edge server connected to the collection are selected from all servers with the authority to receive tasks, and the collection is subjected to task offloading processing to generate a task block cloud computing queue and a task block edge computing queue for the cloud server and the edge server respectively; The task processing result integration module is used to distribute the task block calculation queue to the corresponding server for calculation processing, and summarize and integrate the calculation processing results, including: The task block cloud computing queue and the task block edge computing queue are respectively allocated to the cloud server and the edge server for computing and processing, and the computing and processing results of the edge server are transmitted to the cloud server, so that all computing and processing results are summarized and integrated on the cloud server side to obtain a complete processing result matching the task.

Images