Computing task distribution system based on cloud side-end collaborative edge gateway
By designing a computing task allocation system based on cloud-edge-end collaborative edge gateway in an edge computing system, and using particle swarm algorithm to optimize task allocation, the problem of unbalanced cloud, edge and end resource allocation is solved, and the computing resource utilization efficiency and tasks are improved Execution efficiency.
Patent Information
- Application Number
- CN202410988790.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, resource allocation of cloud, edge and end elements in edge computing systems is difficult to balance, resulting in low computing resource utilization efficiency, resource waste and task delay problems.
Design a computing task allocation system based on cloud edge collaborative edge gateway, including task management module, resource management module, collaborative processing module and task scheduling module, use particle swarm algorithm to optimize task allocation decisions, monitor equipment resources in real time, and coordinate cloud, Task distribution and results summary between edge and end devices.
It realizes the rational arrangement of edge gateway computing resources, avoid resource vacancies, ensure data synchronization and consistency, and improve the execution efficiency and system performance of computing tasks.
Smart Images

Figure CN118945125A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of computing resource allocation, and specifically relates to a computing task allocation system based on a cloud-edge-end collaborative edge gateway. Background Art
[0002] The edge gateway is a gateway deployed at the edge of the network. It connects the physical and digital worlds through network connection, protocol conversion and other functions, and provides lightweight connection management, real-time data analysis and application management functions. The three elements of cloud, edge and end in the edge computing system cooperate with each other to complete the logical architecture of the system. How to balance the three elements of cloud, edge and end has become the key to the rational use of edge gateway computing resources. Summary of the invention
[0003] In order to solve the above problems existing in the prior art, the present invention provides a computing task allocation system based on cloud-edge-end collaborative edge gateway.
[0004] The purpose of the present invention can be achieved through the following technical solutions:
[0005] A computing task allocation system based on cloud-edge-device collaborative edge gateway, comprising: a task management module, a resource management module, a collaborative processing module and a task scheduling module;
[0006] The task management module is used to receive, store and manage the computing tasks to be processed;
[0007] The resource management module is used to monitor and manage computing resources of the cloud, edge and end devices;
[0008] The collaborative processing module is used for the edge gateway to coordinate the task distribution and result aggregation between the cloud, edge and terminal devices;
[0009] The task scheduling module is used to assign tasks to corresponding devices according to task characteristics and device resource conditions;
[0010] The task characteristics include: task type, task priority, task data volume and task time limit;
[0011] The computing resource usage includes: processing power, load, and current task computing progress;
[0012] The computing task allocation system operates as follows:
[0013] S1: Task submission, the user submits the computing task to be processed to the system;
[0014] S2: Task reception: the system receives the computing tasks to be processed submitted by the user and records the characteristics and priority of the tasks;
[0015] S3: Resource monitoring, real-time monitoring of computing resource usage in the cloud, edge, and end devices;
[0016] S4: Task allocation. The system uses the particle swarm algorithm to continuously optimize task allocation decisions and allocate tasks to corresponding devices based on task characteristics and equipment resource conditions.
[0017] S5: Task distribution: tasks assigned to the same device are distributed to the pending task queue of the device according to priority;
[0018] S6; collaborative processing, the edge gateway coordinates the task distribution and result aggregation between the cloud, edge and end devices;
[0019] S7: Task completion: After the device completes the task, it returns the result to the edge gateway;
[0020] S8: Result summary: The edge gateway collects the task execution results of all devices and summarizes and organizes them;
[0021] S9: Result return: The edge gateway returns the task execution result to the task submitter.
[0022] Preferably, in step S4, the particle swarm algorithm is used to optimize the task allocation decision process as follows:
[0023] S41: Initialize the particle swarm, randomly generate a group of particles, each particle represents a possible task allocation scheme, and each particle contains a position and a speed;
[0024] S42: Calculate fitness, and calculate the fitness value of each example according to the performance index of the task allocation scheme;
[0025] S43: Update the global optimal position, according to the fitness value of the particle, update the global optimal position, and record the current optimal task allocation plan;
[0026] S44: updating the speed and position of the particle, according to the current position, speed and global optimal position of the particle, updating the speed and position of the particle;
[0027] S45: Determine whether the termination condition is met;
[0028] S46: Repeat steps S42 to S45 until the termination condition is met.
[0029] Preferably, in step S44, the updating formula of the speed and position of each particle is as follows:
[0030] Speed update formula:
[0031] V i (t+1)=w*Vi (t)+c1*r1*(P i (t)-X i (t))+c2*r2*(P g (t)-X i (t));
[0032] Position update formula:
[0033] X i (t+1)=X i (t)+V i (t+1);
[0034] Among them, V i (t) represents the velocity of particle i at time t, X i (t) represents the position of particle i at time t, P i (t) represents the individual optimal position of particle i, P g (t) represents the global optimal position, w is the inertia weight, c1 and c2 are acceleration factors, and r1 and r2 are random numbers.
[0035] Preferably, in step S45, the termination condition includes reaching a maximum number of iterations, reaching a maximum running time, or finding a task allocation solution that meets the requirements.
[0036] Preferably, the task management module includes a task receiving unit, a task parsing unit and a task status management unit;
[0037] The task accepting unit is used to receive a task request from a cloud or edge device;
[0038] The task parsing unit is used to parse the task request and extract task characteristics;
[0039] The task status management unit is used to track and manage the execution status of the task, including the update and recording of the submission, execution and completion status of the task.
[0040] Preferably, the resource management module includes a device registration unit and a resource detection unit;
[0041] The device registration unit is used to register device resource information in the system;
[0042] The resource detection unit is used to detect the resource utilization of the edge device in real time, including CPU, memory and network bandwidth.
[0043] Preferably, the collaborative processing module includes a task collaboration unit, a data synchronization unit and a collaborative decision-making unit;
[0044] The task coordination unit is used to coordinate the segmentation, transmission and result merging of tasks that require collaborative processing by multiple edge devices;
[0045] The data synchronization unit is used to ensure data synchronization and consistency during the collaborative processing process to avoid data conflicts and losses;
[0046] The collaborative decision unit is used to decide which tasks need to be collaboratively processed to improve computing efficiency based on task characteristics and device resource conditions.
[0047] Preferably, the task scheduling module includes a task allocation algorithm unit, a task priority unit, a task queue management unit and a task scheduling strategy unit;
[0048] The task allocation algorithm unit is used to select an appropriate task allocation decision based on factors such as task characteristics, device resource status and network status;
[0049] The task priority unit is used to schedule high-priority tasks to be executed first according to the priorities of the tasks;
[0050] The task queue management unit is used to maintain the task queue and sort and schedule the tasks according to the requirements of the scheduling algorithm;
[0051] The task scheduling strategy unit is used to select a suitable scheduling strategy according to the type and characteristics of the task.
[0052] The beneficial effects of the present invention are:
[0053] 1. The system uses particle swarm algorithm to continuously optimize task allocation decisions and rationally arrange the computing resources of the edge gateway.
[0054] 2. A resource detection unit is set up to detect the resource utilization of edge devices in real time to avoid resource vacancies and waste.
[0055] 3. A collaborative processing module is set up to coordinate the segmentation, transmission and result merging of tasks that require collaborative processing of multiple edge devices, while ensuring data synchronization and consistency during the collaborative processing process to avoid data conflicts and losses. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.
[0057] Figure 1 A system structure diagram provided in one embodiment of the present invention;
[0058] Figure 2 The system operation steps provided in one embodiment of the present invention;
[0059] Figure 3 An algorithm optimization task allocation decision process provided in one embodiment of the present invention;
[0060] Figure 4 A structural diagram of a task management module provided in one embodiment of the present invention;
[0061] Figure 5 A structural diagram of a resource management module provided in one embodiment of the present invention;
[0062] Figure 6 A structural diagram of a collaborative processing module provided in one embodiment of the present invention;
[0063] Figure 7 This is a structural diagram of a task scheduling module provided in one embodiment of the present invention. DETAILED DESCRIPTION
[0064] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.
[0065] Example 1
[0066] Please refer to Figure 1 to Figure 2 The present embodiment provides a computing task allocation system based on a cloud-edge-device collaborative edge gateway, including: a task management module, a resource management module, a collaborative processing module, and a task scheduling module;
[0067] The task management module is used to receive, store and manage the computing tasks to be processed;
[0068] The resource management module is used to monitor and manage computing resources of the cloud, edge and end devices;
[0069] The collaborative processing module is used for the edge gateway to coordinate the task distribution and result aggregation between the cloud, edge and terminal devices;
[0070] The task scheduling module is used to assign tasks to corresponding devices according to task characteristics and device resource conditions;
[0071] The task characteristics include: task type, task priority, task data volume and task time limit;
[0072] The computing resource usage includes: processing power, load, and current task computing progress;
[0073] The computing task allocation system operates as follows:
[0074] S1: Task submission, the user submits the computing task to be processed to the system;
[0075] S2: Task reception: the system receives the computing tasks to be processed submitted by the user and records the characteristics and priority of the tasks;
[0076] S3: Resource monitoring, real-time monitoring of computing resource usage in the cloud, edge, and end devices;
[0077] S4: Task allocation. The system uses the particle swarm algorithm to continuously optimize task allocation decisions and allocate tasks to corresponding devices based on task characteristics and equipment resource conditions.
[0078] S5: Task distribution: tasks assigned to the same device are distributed to the pending task queue of the device according to priority;
[0079] S6; collaborative processing, the edge gateway coordinates the task distribution and result aggregation between the cloud, edge and end devices;
[0080] S7: Task completion: After the device completes the task, it returns the result to the edge gateway;
[0081] S8: Result summary: The edge gateway collects the task execution results of all devices and summarizes and organizes them;
[0082] S9: Result return: The edge gateway returns the task execution result to the task submitter.
[0083] In this embodiment, the system uses a particle swarm algorithm to continuously optimize task allocation decisions and rationally arrange computing resources of the edge gateway.
[0084] The system is equipped with a resource detection unit to detect the resource utilization of edge devices in real time to avoid resource vacancies and waste.
[0085] The system is equipped with a collaborative processing module to coordinate the segmentation, transmission and result merging of tasks that require collaborative processing by multiple edge devices, while ensuring data synchronization and consistency during the collaborative processing to avoid data conflicts and losses.
[0086] Example 2
[0087] Please refer to Figure 3 According to a computing task allocation system based on cloud-edge-device collaborative edge gateway provided in Example 1, in step S4, the particle swarm algorithm is used to optimize the task allocation decision process as follows:
[0088] S41: Initialize the particle swarm, randomly generate a group of particles, each particle represents a possible task allocation scheme, and each particle contains a position and a speed;
[0089] S42: Calculate fitness, and calculate the fitness value of each example according to the performance index of the task allocation scheme;
[0090] S43: Update the global optimal position, according to the fitness value of the particle, update the global optimal position, and record the current optimal task allocation plan;
[0091] S44: updating the speed and position of the particle, according to the current position, speed and global optimal position of the particle, updating the speed and position of the particle;
[0092] S45: Determine whether a termination condition is met, such as reaching a maximum number of iterations or finding a task allocation solution that meets the requirements;
[0093] S46: Repeat steps S42 to S45 until the termination condition is met.
[0094] In this embodiment, the particle swarm algorithm is used to continuously optimize the task allocation decision and reasonably arrange the computing resources of the edge gateway. The particle swarm algorithm is an optimization algorithm based on swarm intelligence, which simulates the behavior of groups such as bird flocks or fish schools. In the computing task allocation system, the role of the particle swarm algorithm is to find the optimal task allocation solution through the optimization algorithm to maximize the performance and efficiency of the system.
[0095] Example 3
[0096] According to a computing task allocation system based on cloud-edge-device collaborative edge gateway provided in Example 2, in step S44, the update formula of the speed and position of each particle is as follows:
[0097] Speed update formula:
[0098] V i (t+1)=w*V i (t)+c1*r1*(P i (t)-X i (t))+c2*r2*(P g (t)-X i (t));
[0099] Position update formula:
[0100] X i (t+1)=X i (t)+V i (t+1);
[0101] Among them, V i (t) represents the velocity of particle i at time t, X i (t) represents the position of particle i at time t, P i (t) represents the individual optimal position of particle i, P g(t) represents the global optimal position, w is the inertia weight, c1 and c2 are acceleration factors, and r1 and r2 are random numbers.
[0102] In step S45, the termination conditions include reaching the maximum number of iterations, reaching the maximum running time, or finding a task allocation solution that meets the requirements.
[0103] The termination conditions of the particle swarm algorithm are usually the following:
[0104] Reaching the maximum number of iterations: Set a predetermined maximum number of iterations. When the number of algorithm iterations reaches this value, the algorithm is terminated.
[0105] Reaching the maximum running time: Set a predetermined maximum running time. When the algorithm running time exceeds this value, the algorithm is terminated.
[0106] Reaching the objective function threshold: Set a predetermined objective function threshold. When the objective function value of the optimal solution obtained by the algorithm reaches or exceeds the threshold, the algorithm is terminated.
[0107] Reaching convergence conditions: By monitoring the convergence of the particle swarm algorithm, when the convergence degree of the algorithm meets certain conditions, the algorithm is terminated. Common convergence conditions include that the difference between the global optimal solution and the local optimal solution of the particle swarm is less than a certain threshold, or the speed and position changes of the particle swarm are less than a certain threshold.
[0108] Reaching the predetermined stopping criteria: According to the characteristics of the specific problem, set some specific stopping criteria, such as reaching a certain target accuracy, satisfying certain constraints, etc.
[0109] In this embodiment, in the computing task allocation system, the position of the particle represents the task allocation scheme, and the speed represents the changing direction and speed of the task allocation scheme. By continuously updating the speed and position of the particle, the system can search for the optimal task allocation scheme.
[0110] In addition to the updating formulas for speed and position, the particle swarm algorithm also needs to set appropriate parameters, such as inertia weight w, acceleration factors c1 and c2, and determine the fitness function to evaluate the performance of the task allocation scheme.
[0111] In summary, the role of the particle swarm algorithm in the computing task allocation system based on cloud-edge collaborative edge gateway is to find the optimal task allocation solution through iterative search, thereby improving the performance and efficiency of the system.
[0112] Example 4
[0113] Please refer to Figure 4, according to a computing task allocation system based on cloud-edge-device collaborative edge gateway provided in Example 1, the task management module includes a task receiving unit, a task parsing unit and a task status management unit;
[0114] The task accepting unit is used to receive a task request from a cloud or edge device;
[0115] This unit is responsible for receiving task requests from the cloud or other edge devices. It acts as an interface between the task allocation system and external communications. After receiving the task request, it passes it to the task parsing unit for processing. The main function of the task receiving unit is to ensure the reliable reception of task requests, including the reception, verification and decapsulation of data packets.
[0116] The task parsing unit is used to parse the task request and extract task characteristics;
[0117] This unit is responsible for parsing and analyzing the received task requests. It extracts necessary information from the task request, such as task type, priority, resource requirements, etc., and prioritizes tasks and formulates allocation strategies based on task characteristics. The goal of the task parsing unit is to effectively allocate tasks to appropriate edge nodes or the cloud to achieve efficient execution of tasks and resource utilization.
[0118] The task status management unit is used to track and manage the execution status of the task, including the update and recording of the submission, execution and completion status of the task.
[0119] This unit is responsible for managing the execution status and progress of tasks. It tracks and records the execution of each task and updates the task status information in a timely manner. The task status management unit can also detect the availability and load of edge nodes, and dynamically schedule and migrate tasks according to the real-time resource status to achieve balanced distribution and load balancing of tasks. In addition, it is also responsible for processing feedback and result reporting after the task is completed.
[0120] In this embodiment, the system also sets a task management module to ensure that the task allocation system can efficiently and reliably receive, parse and manage task requests. Through reasonable task allocation and state management, high concurrent processing of tasks, dynamic scheduling and optimization of resources can be achieved, and the execution efficiency of computing tasks and the overall performance of the system can be improved.
[0121] Example 5
[0122] Please refer to Figure 5 , According to a computing task allocation system based on cloud-edge-device collaborative edge gateway provided in Example 1, the resource management module includes a device registration unit and a resource detection unit;
[0123] The device registration unit is used to register device resource information in the system;
[0124] This unit is responsible for the registration and authentication of edge devices. When an edge device is connected to the system, the device registration unit will authenticate and authorize it and assign it a unique identifier. The device registration process ensures that only legitimate devices are allowed to access the system and participate in computing task allocation. The registration unit also maintains the device registration information and provides it to other modules for reference.
[0125] The resource detection unit is used to detect the resource utilization of the edge device in real time, including CPU, memory and network bandwidth.
[0126] This unit is responsible for detecting and managing the resource status of edge nodes. It periodically monitors the available resources of edge nodes, including computing power, storage capacity, bandwidth, etc. The resource detection unit generates a resource report based on the detected resource status and provides it to the task analysis unit as a basis for task allocation. It can also obtain the load status of edge nodes in real time for load balancing and resource scheduling.
[0127] In this embodiment, the role of the system resource management module is to ensure that the computing task allocation system can effectively manage and utilize the resources of the edge device. The device registration unit is used to manage the permissions and identification of device access to ensure the security and compliance of the system. The resource detection unit is used to monitor and report the resource status of the edge node in real time, and provide reliable resource information to the task parsing unit for reasonable task scheduling and allocation. By optimizing resource management, the system performance and resource utilization can be maximized, and the execution efficiency of computing tasks and the overall responsiveness of the system can be improved.
[0128] Example 6
[0129] Please refer to Figure 6 , according to a computing task allocation system based on cloud-edge-device collaborative edge gateway provided in Example 1, the collaborative processing module includes a task collaboration unit, a data synchronization unit and a collaborative decision unit;
[0130] The task coordination unit is used to coordinate the segmentation, transmission and result merging of tasks that require collaborative processing by multiple edge devices;
[0131] This unit is responsible for coordinating and managing the collaborative execution of tasks between multiple edge nodes. The task coordination unit assigns tasks to appropriate edge nodes for execution based on the type, priority, and resource requirements of the task. At the same time, it is also responsible for monitoring and managing the execution status of the task to ensure the smooth execution of the collaborative task. The significance of the task coordination unit is to improve the parallel processing capability of computing tasks and improve computing efficiency and system performance through the collaborative execution of tasks.
[0132] The data synchronization unit is used to ensure data synchronization and consistency during the collaborative processing process to avoid data conflicts and losses;
[0133] This unit is responsible for processing data synchronization and interaction between edge nodes. The data synchronization unit can synchronize, aggregate, and update the intermediate results generated during task execution to maintain data consistency. At the same time, it can also handle data interaction between different edge nodes to achieve data sharing required for distributed computing and collaborative decision-making. The significance of the data synchronization unit is to provide data consistency and reliability to support collaborative processing between multiple nodes.
[0134] The collaborative decision unit is used to decide which tasks need to be collaboratively processed to improve computing efficiency based on task characteristics and device resource conditions.
[0135] This unit is responsible for making decisions on key issues that arise during the collaborative execution of multiple nodes. The collaborative decision unit dynamically schedules tasks and allocates resources based on information such as task priority, node load, and task execution progress. It can make the best decision based on real-time conditions to ensure efficient execution of computing tasks and system performance optimization. The significance of the collaborative decision unit is to provide intelligent decision-making functions, optimize task allocation and resource scheduling, and achieve system adaptability and efficiency.
[0136] In this embodiment, the system can realize the collaborative execution of tasks between multiple edge nodes through the functions and synergy of the collaborative processing module, and the computing task allocation system based on the cloud-edge collaborative edge gateway can improve the parallel processing capability and efficiency of computing tasks. At the same time, through data synchronization and collaborative decision-making, distributed processing of tasks and optimal allocation of resources can be achieved to improve the performance and responsiveness of the system. This will help meet the needs of large-scale computing tasks and provide efficient edge computing solutions.
[0137] Example 7
[0138] Please refer to Figure 7 , according to a computing task allocation system based on cloud-edge-device collaborative edge gateway provided in Example 1, the task scheduling module includes a task allocation algorithm unit, a task priority unit, a task queue management unit and a task scheduling strategy unit;
[0139] The task allocation algorithm unit is used to select an appropriate task allocation decision based on factors such as task characteristics, device resource status and network status;
[0140] This unit is responsible for selecting the appropriate edge node for task execution. The task allocation algorithm unit uses a suitable scheduling algorithm to determine which edge node to assign the task to based on the system's resource status and task characteristics. The goal of the task allocation algorithm unit is to achieve balanced distribution of tasks and rational use of resources, thereby improving the overall performance and efficiency of the system.
[0141] The task priority unit is used to schedule high-priority tasks to be executed first according to the priorities of the tasks;
[0142] This unit is responsible for setting priorities for tasks. The Task Priority Unit assigns a priority to each task based on factors such as the importance, urgency, and deadline of the task. Based on the priority of the task, the Task Scheduling Module can reasonably arrange the execution order of the tasks to ensure that urgent and important tasks are processed first.
[0143] The task queue management unit is used to maintain the task queue and sort and schedule the tasks according to the requirements of the scheduling algorithm;
[0144] This unit is responsible for managing the task queue, which is used to store tasks to be executed. The task queue management unit receives the tasks assigned by the task allocation algorithm unit and queues the tasks according to certain rules to ensure that the tasks are executed in a certain order. It can realize the functions of task buffering and queuing, so as to better perform task scheduling and resource management.
[0145] The task scheduling strategy unit is used to select a suitable scheduling strategy according to the type and characteristics of the task.
[0146] This unit is responsible for formulating task scheduling strategies and implementing real-time scheduling and execution of tasks. The task scheduling strategy unit considers the current resource status of the system, the priority of the task and the results of the scheduling algorithm, and dynamically adjusts the execution order and allocation scheme of the tasks. It can allocate, schedule and reschedule tasks according to actual conditions to achieve system adaptability and efficiency.
[0147] In this embodiment, the role of the task scheduling module is to achieve reasonable scheduling and allocation of computing tasks to maximize the performance and resource utilization of the system. Through effective task allocation algorithms, task priority management, task queue management and task scheduling strategies, the execution load of tasks can be balanced, task scheduling can be optimized, and the execution efficiency of computing tasks and the overall responsiveness of the system can be improved.
[0148] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.
Claims
1. A computing task allocation system based on cloud-edge-device collaborative edge gateway, characterized in that: The computing task allocation system includes: a task management module, a resource management module, a collaborative processing module and a task scheduling module; The task management module is used to receive, store and manage the computing tasks to be processed; The resource management module is used to monitor and manage computing resources of the cloud, edge and end devices; The collaborative processing module is used for the edge gateway to coordinate the task distribution and result aggregation between the cloud, edge and terminal devices; The task scheduling module is used to assign tasks to corresponding devices according to task characteristics and device resource conditions; The task characteristics include: task type, task priority, task data volume and task time limit; The computing resource usage includes: processing power, load, and current task computing progress; The computing task allocation system operates as follows: S1: Task submission, the user submits the computing task to be processed to the system; S2: Task reception: the system receives the computing tasks to be processed submitted by the user and records the characteristics and priority of the tasks; S3: Resource monitoring, real-time monitoring of computing resource usage in the cloud, edge, and end devices; S4: Task allocation. The system uses the particle swarm algorithm to continuously optimize task allocation decisions and allocate tasks to corresponding devices based on task characteristics and equipment resource conditions. S5: Task distribution: tasks assigned to the same device are distributed to the pending task queue of the device according to priority; S6; collaborative processing, the edge gateway coordinates the task distribution and result aggregation between the cloud, edge and end devices; S7: Task completion: After the device completes the task, it returns the result to the edge gateway; S8: Result summary: The edge gateway collects the task execution results of all devices and summarizes and organizes them; S9: Result return: The edge gateway returns the task execution result to the task submitter.
2. According to claim 1, a computing task allocation system based on cloud-edge-device collaborative edge gateway is characterized in that: In step S4, the decision process of optimizing task allocation using particle swarm algorithm is as follows: S41: Initialize the particle swarm, randomly generate a group of particles, each particle represents a possible task allocation scheme, and each particle contains a position and a speed; S42: Calculate fitness, and calculate the fitness value of each example according to the performance index of the task allocation scheme; S43: Update the global optimal position, according to the fitness value of the particle, update the global optimal position, and record the current optimal task allocation plan; S44: updating the speed and position of the particle, according to the current position, speed and global optimal position of the particle, updating the speed and position of the particle; S45: Determine whether the termination condition is met; S46: Repeat steps S42 to S45 until the termination condition is met.
3. According to claim 2, a computing task allocation system based on cloud-edge-device collaborative edge gateway is characterized in that: In step S44, the updating formula of the speed and position of each particle is as follows: Speed update formula: V i (t+1)=w*V i (t)+c1*r1*(P i (t)-X i (t))+c2*r2*(P g (t)-X i (t)); Position update formula: X i (t+1)=X i (t)+V i (t+1); Among them, V i (t) represents the velocity of particle i at time t, X i (t) represents the position of particle i at time t, P i (t) represents the individual optimal position of particle i, P g (t) represents the global optimal position, w is the inertia weight, c1 and c2 are acceleration factors, and r1 and r2 are random numbers.
4. According to claim 3, a computing task allocation system based on cloud-edge-device collaborative edge gateway is characterized in that: In step S45, the termination conditions include reaching the maximum number of iterations, reaching the maximum running time, or finding a task allocation solution that meets the requirements.
5. According to claim 1, a computing task allocation system based on cloud-edge-device collaborative edge gateway is characterized in that: The task management module includes a task receiving unit, a task parsing unit and a task status management unit; The task accepting unit is used to receive a task request from a cloud or edge device; The task parsing unit is used to parse the task request and extract task characteristics; The task status management unit is used to track and manage the execution status of the task, including the update and recording of the submission, execution and completion status of the task.
6. The computing task allocation system based on cloud-edge-device collaborative edge gateway according to claim 1 is characterized in that: The resource management module includes a device registration unit and a resource detection unit; The device registration unit is used to register device resource information in the system; The resource detection unit is used to detect the resource utilization of the edge device in real time, including CPU, memory and network bandwidth.
7. The computing task allocation system based on cloud-edge-device collaborative edge gateway according to claim 1 is characterized in that: The collaborative processing module includes a task collaboration unit, a data synchronization unit and a collaborative decision-making unit; The task coordination unit is used to coordinate the segmentation, transmission and result merging of tasks that require collaborative processing by multiple edge devices; The data synchronization unit is used to ensure data synchronization and consistency during the collaborative processing process to avoid data conflicts and losses; The collaborative decision unit is used to decide which tasks need to be collaboratively processed to improve computing efficiency based on task characteristics and device resource conditions.
8. The computing task allocation system based on cloud-edge-device collaborative edge gateway according to claim 1 is characterized in that: The task scheduling module includes a task allocation algorithm unit, a task priority unit, a task queue management unit and a task scheduling strategy unit; The task allocation algorithm unit is used to select an appropriate task allocation decision based on factors such as task characteristics, device resource status and network status; The task priority unit is used to schedule high-priority tasks to be executed first according to the priorities of the tasks; The task queue management unit is used to maintain the task queue and sort and schedule the tasks according to the requirements of the scheduling algorithm; The task scheduling strategy unit is used to select a suitable scheduling strategy according to the type and characteristics of the task.
Citation Information
Cited By
Large model computing power distribution and scheduling system oriented to edge computing
CN119166369A
A large-scale computing power allocation and scheduling system for edge computing
CN119166369B
Computer and cloud computing power optimization method based on multi-task collaboration
CN119473631A
Decentralized network security detection online edge task scheduling method and system
CN119696852A
Decentralized Network Security Detection Online Edge Task Scheduling Method and System
CN119696852B
