5G gateway resource optimization system and method

Abstract

The invention discloses a 5G gateway resource optimization system and method. The system comprises an acquisition module, a calculation module I, a judgment module, a calculation module II, a calculation module III, a processing module, a feedback module and the like. The invention relates to the technical field of resource optimization, in particular to a 5G gateway resource optimization system and method. The technical problem to be solved by the invention is to provide the 5G gateway resource optimization system and method, so that dynamic task allocation can be performed, the utilization rate of resource allocation is improved, and the management efficiency of gateway resources is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of resource optimization, and in particular, to a 5G gateway resource optimization system and method. Background Art

[0002] With the development of the Internet of Things and 5G communication technology, edge gateways, as an important carrier of edge computing, have become a core component in the data processing and decision-making process. Edge computing can perform pre-analysis and local decision-making at the data source and communicate with the cloud, thereby reducing dependence on cloud resources and effectively reducing response time. Its data resource optimization plays an important role in improving the efficiency of the overall system.

[0003] At present, 5G edge gateways usually perform resource optimization and task management according to the system's pre-set rules and fixed thresholds. By setting the constraints for the edge gateway's resource optimization and allocation, a resource allocation model is obtained. When it is obtained that there is a certain amount of resources to be allocated at a certain node, the task computing load data is obtained and loaded into the resource allocation model, thereby scheduling the idle computing power of the edge gateway and front-end equipment to share the burden of computing resources.

[0004] Regarding the above technical solutions, these strategies are difficult to cope with dynamically changing task loads and changing network conditions. For example, when a node is under high concurrency pressure or overload risk, dynamic task allocation cannot be performed in a timely manner, resulting in a lot of resource waste and reducing the management efficiency of gateway resources. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a 5G gateway resource optimization system and method, so as to enable dynamic task allocation, improve the utilization rate of resource allocation, and improve the management efficiency of gateway resources.

[0006] In a first aspect, the present invention provides a 5G gateway resource optimization system, which adopts the following technical solutions to achieve the purpose of the invention: A 5G gateway resource optimization system includes the following modules: Acquisition module: The output end is connected to the input end of calculation module I, and is used to obtain the task data to be processed in I gateway nodes. The tasks to be processed include real-time tasks and N non-real-time tasks. The calculation amount of the real-time task of the i-th node is , the computational amount of the nth non-real-time task of the ith node is , the maximum single-task computation amount of the i-th node is , the average task calculation rate of the i-th node is ; Calculation module I: The input end is connected to the output end of the acquisition module, and the output end is connected to the input end of the judgment module, which is used to calculate the processing time of the task to be processed at the i-th node , The calculation model is: ; Judgment module: The input end is connected to the output end of calculation module I, and the output end is connected to the input end of calculation module II and the input end of feedback module respectively, which is used to obtain the deadline of the pending task of the i-th node. , the system real time is ,like , then it is judged that the resources are sufficient and the feedback module is executed. Otherwise, it is judged that the resources are insufficient and the calculation II step is executed; Computation module II: The input end is connected to the output end of the judgment module, and the output end is connected to the input end of the calculation module III, which is used to calculate the non-real-time task allocation weight of the i-th node according to the real-time task calculation amount, the non-real-time task calculation amount and the task type. , The calculation model is: ; in, is the priority weight of the i-th node, ; Computing module III: The input end is connected to the output end of computing module II, and the output end is connected to the input end of the processing module, which is used to calculate the task resource allocation ratio of the i-th node: , the task resource allocation of the i-th node is ; Processing module: The input end is connected to the output end of the computing module III, and the output end is connected to the input end of the processing module, which is used to allocate the task resources according to the ratio of the i-th node. Arrange non-real-time tasks in descending order according to the task resource allocation amount Distribute to other nodes for calculation; Feedback module: The input end is connected to the output end of the judgment module and the output end of the processing module respectively, and is used to upload the distribution result of the task data to be processed to the management end.

[0007] As a further limitation of the present technical solution, in the calculation module II, The calculation model is updated as follows: ; Among them, α is the first weight coefficient, β is the second weight coefficient, and γ is the third weight coefficient.

[0008] As a further limitation of the present technical solution, in the calculation module II, The calculation model is updated as follows: ; in, is the real-time coefficient of task processing of the i-th node, The calculation model is: .

[0009] As a further limitation of the present technical solution, in the calculation module I, The calculation model is updated as follows: ; in, is the node load factor, and Y is the maximum concurrent task capacity.

[0010] As a further limitation of this technical solution, in the processing module, the minimum task resource allocation amount is also set to , the node numbers after descending order are 1, 2, 3, ..., m, ..., M, where M = I. If according to The task resource allocation of the mth node in descending order is less than or equal to , after allocating to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated.

[0011] In a second aspect, the present invention provides a 5G gateway resource optimization method, which adopts the following technical solutions to achieve the purpose of the invention: A 5G gateway resource optimization method comprises the following steps: Get data: Get the pending task data in I gateway nodes. The pending tasks include real-time tasks and N non-real-time tasks. The real-time task calculation amount of the i-th node is , the computational amount of the nth non-real-time task of the ith node is , the maximum single-task computation amount of the i-th node is , the average task calculation rate of the i-th node is ; Calculation I: Calculate the processing time of the task to be processed at the i-th node , The calculation model is: ; Judgment: Get the deadline for the pending tasks of the i-th node. , the system real time is ,like , then it is judged that the resources are sufficient and the feedback step is executed; otherwise, it is judged that the resources are insufficient and the calculation II step is executed; Calculation II: Calculate the non-real-time task allocation weight of the i-th node based on the real-time task calculation amount, the non-real-time task calculation amount and the task type , The calculation model is: ; in, is the priority weight of the i-th node, ; Calculation III: Calculate the task resource allocation ratio of the i-th node: , the task resource allocation of the i-th node is ; Processing: According to the task resource allocation ratio of the i-th node Arrange non-real-time tasks in descending order according to the task resource allocation amount Distribute to other nodes for calculation; Feedback: Upload the data allocation results of pending tasks to the management end.

[0012] As a further limitation of the present technical solution, in step II, The calculation model is updated as follows: ; Among them, α is the first weight coefficient, β is the second weight coefficient, and γ is the third weight coefficient.

[0013] As a further limitation of the present technical solution, in step II, The calculation model is updated as follows: ; in, is the real-time coefficient of task processing of the i-th node, The calculation model is: .

[0014] As a further limitation of the present technical solution, in the calculation step I, The calculation model is updated as follows: ; in, is the node load factor, and Y is the maximum concurrent task capacity.

[0015] As a further limitation of this technical solution, in the processing step, the minimum task resource allocation amount is also set to , the node numbers after descending order are 1, 2, 3, ..., m, ..., M, where M = I. If according to The task resource allocation of the mth node in descending order is less than or equal to , after allocating to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: 1. By calculating the processing time of the pending tasks of the i-th node, it is judged whether the computing resources of the node are sufficient. When it is judged that the resources of the node are insufficient, the non-real-time task allocation weight of the i-th node is calculated according to the computing amount of the real-time tasks, the computing amount of the non-real-time tasks and the task type of the node, so as to achieve multi-objective optimization. By introducing priority weights, the scheduling priority of high-value business nodes is significantly higher than that of ordinary task nodes. The task resource allocation ratio of the i-th node is calculated, and then the task resource allocation amount of the i-th node is calculated. According to the task resource allocation ratio of the i-th node, the non-real-time tasks are allocated to other nodes in descending order for calculation, so as to perform dynamic task allocation, improve the utilization rate of resource allocation, and improve the management efficiency of gateway resources.

[0017] 2. By designing the first weight coefficient, the second weight coefficient, and the third weight coefficient, it is possible to adapt to the precise adjustment of node priority, resource utilization, and real-time performance in different environments, and further adapt to different business requirements in differentiated scenarios, thereby improving practicality and reliability.

[0018] 3. By calculating the real-time coefficient of the node's task processing, it is helpful to ensure the priority of real-time task processing. For example, when facing high real-time services such as emergency braking instructions and collaborative perception, the real-time task priority scheduling mechanism can effectively reduce the task interruption rate. At the same time, the resource release rate of non-real-time tasks is significantly improved, supporting dynamic expansion and contraction needs, enabling dynamic task allocation, improving the utilization rate of resource allocation, and improving the management efficiency of gateway resources.

[0019] 4. By introducing the node load factor and the maximum concurrent task capacity, the processing time of the pending tasks of the i-th node is calculated, so as to quantify the concurrent pressure of the gateway node in real time and accurately identify the risk of node overload. Compared with the traditional static threshold detection, it can predict resource bottlenecks in advance and avoid system crashes caused by task accumulation. In the case of burst traffic, the dynamic correction mechanism of the load factor significantly improves the elastic scaling capability of node resources, so that key tasks can obtain resources first.

[0020] 5. By setting the minimum resource allocation amount and descending allocation rules, resource fragmentation can be effectively curbed. While maintaining allocation efficiency, the resource waste caused by the allocation of fragmented tasks can be reduced. This makes it suitable for batch processing scenarios of computing-intensive non-real-time tasks, thereby improving the utilization rate of resource allocation and the management efficiency of gateway resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. 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. In the drawings: Figure 1 This is a system diagram of Embodiment 1 of the present invention; Figure 2 This is a flow chart of Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the present invention Figure 1 and Figure 2 , the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] It should be noted that the directional terms such as left, right, up, down, front and back in the embodiments of the present invention are merely relative concepts or are based on the normal use state of the product, that is, the direction of movement of the product, and should not be considered as limiting.

[0024] When a component is referred to as being "located" or "disposed on" another component, it may be on the other component or there may be an intervening component. When a component is referred to as being "connected to" another component, it may be directly connected to the other component or there may be an intervening component.

[0025] Embodiment 1: A 5G gateway resource optimization system includes the following modules: Acquisition module: The output end is connected to the input end of calculation module I, and is used to obtain the task data to be processed in I gateway nodes. The tasks to be processed include real-time tasks and N non-real-time tasks. The calculation amount of the real-time task of the i-th node is , the computational amount of the nth non-real-time task of the ith node is , the maximum single-task computation amount of the i-th node is , the average task calculation rate of the i-th node is ; Calculation module I: The input end is connected to the output end of the acquisition module, and the output end is connected to the input end of the judgment module, which is used to calculate the processing time of the task to be processed at the i-th node , The calculation model is: ; in, is the node load factor, Y is the maximum concurrent task capacity; Judgment module: The input end is connected to the output end of calculation module I, and the output end is connected to the input end of calculation module II and the input end of feedback module respectively, which is used to obtain the deadline of the pending task of the i-th node. , the system real time is ,like , then it is judged that the resources are sufficient and the feedback module is executed. Otherwise, it is judged that the resources are insufficient and the calculation II step is executed; Computation module II: The input end is connected to the output end of the judgment module, and the output end is connected to the input end of the calculation module III, which is used to calculate the non-real-time task allocation weight of the i-th node according to the real-time task calculation amount, the non-real-time task calculation amount and the task type. , The calculation model is: ; in, is the priority weight of the i-th node, , α is the first weight coefficient, β is the second weight coefficient, γ is the third weight coefficient, is the real-time coefficient of task processing of the i-th node, The calculation model is: ; Computing module III: The input end is connected to the output end of computing module II, and the output end is connected to the input end of the processing module, which is used to calculate the task resource allocation ratio of the i-th node: , the task resource allocation of the i-th node is ; Processing module: The input end is connected to the output end of the computing module III, and the output end is connected to the input end of the processing module, which is used to allocate the task resources according to the ratio of the i-th node. Arrange non-real-time tasks in descending order according to the task resource allocation amount Distribute to other nodes for calculation; Set the minimum task resource allocation to , the node numbers after descending order are 1, 2, 3, ..., m, ..., M, where M = I. If according to The task resource allocation of the mth node in descending order is less than or equal to , after being allocated to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated; Feedback module: The input end is connected to the output end of the judgment module and the output end of the processing module respectively, and is used to upload the distribution result of the task data to be processed to the management end.

[0026] The working principle of this embodiment is: A total of I gateway nodes are set up in the area where resource allocation can be performed. The acquisition module obtains the pending task data in the I gateway nodes. The pending tasks include real-time tasks and N non-real-time tasks, and obtains the real-time task computing amount of the ith node, the nth non-real-time task computing amount of the ith node, the maximum single task computing amount of the ith node, and the average task computing rate of the ith node. The calculation module I calculates the processing time of the pending task of the ith node by introducing the node load factor and the maximum concurrent task carrying capacity, thereby quantifying the concurrent pressure of the gateway node in real time and accurately identifying the node overload risk. Compared with traditional static threshold detection, it can predict resource bottlenecks in advance and avoid system crashes caused by task accumulation. In the burst traffic scenario, the dynamic correction mechanism of the load factor significantly improves the elastic scaling capability of node resources, so that key tasks can obtain resources first.

[0027] The judgment module obtains the task deadline of the pending task of the i-th node and the system real-time time. If the judgment module determines that the resources are sufficient, the feedback module uploads the data allocation result of the pending task to the management end for recording. If the judgment module determines that the resources are insufficient, the calculation module II calculates the non-real-time task allocation weight of the i-th node according to the real-time task calculation amount, non-real-time task calculation amount and task type of the node, thereby realizing multi-objective optimization. By introducing priority weights, the scheduling priority of high-value business nodes is significantly higher than that of ordinary task nodes. By designing the first weight coefficient, the second weight coefficient, and the third weight coefficient, it is possible to adapt to the precise adjustment of node priority, resource utilization, and real-time performance in different environments, and further adapt to different business requirements in differentiated scenarios, thereby improving practicality and reliability; by calculating the real-time coefficient of the task processing of the node It is conducive to ensuring the priority of real-time task processing. For example, when facing high real-time services such as emergency braking instructions and collaborative perception, the real-time task priority scheduling mechanism can effectively reduce the task interruption rate. At the same time, the resource release rate of non-real-time tasks is significantly improved, supporting dynamic expansion and contraction needs, and can perform dynamic task allocation, improve the utilization rate of resource allocation, and improve the management efficiency of gateway resources.

[0028] The calculation module III calculates the task resource allocation ratio of the i-th node, and then calculates the task resource allocation amount of the i-th node. The processing module allocates the non-real-time tasks to other nodes in descending order according to the task resource allocation amount of the i-th node for calculation, thereby performing dynamic task allocation, improving the utilization rate of resource allocation, and improving the management efficiency of gateway resources; Set the minimum task resource allocation to If the task resource allocation amount of the mth node in descending order is less than or equal to that of the m+1th node, the allocation to the next node will be stopped after being allocated to the m+1th node, and the remaining non-real-time tasks to be allocated will be completed by the m+1th node. The feedback module uploads the allocation results of the pending task data to the management end. By setting the minimum resource allocation amount and descending allocation rules, resource fragmentation can be effectively curbed. While maintaining the allocation efficiency, the resource waste caused by the allocation of fragmented tasks can be reduced. This makes it suitable for batch processing scenarios of computing-intensive non-real-time tasks, thereby improving the utilization rate of resource allocation and the management efficiency of gateway resources.

[0029] Embodiment 2: A 5G gateway resource optimization method comprises the following steps: S1. Obtain data: Obtain the pending task data in I gateway nodes. The pending tasks include real-time tasks and N non-real-time tasks. The calculation amount of the real-time task of the i-th node is , the computational amount of the nth non-real-time task of the ith node is , the maximum single-task computation amount of the i-th node is , the average task calculation rate of the i-th node is ; S2. Calculation I: Calculate the processing time of the task to be processed at the i-th node , The calculation model is: ; in, is the node load factor, Y is the maximum concurrent task capacity; S3. Judgment: Get the deadline of the pending task of the i-th node as , the system real time is ,like , then it is judged that the resources are sufficient and the feedback module is executed. Otherwise, it is judged that the resources are insufficient and the calculation II step is executed; S4, Calculation II: Calculate the non-real-time task allocation weight of the i-th node based on the real-time task calculation amount, the non-real-time task calculation amount and the task type , The calculation model is: ; in, is the priority weight of the i-th node, , α is the first weight coefficient, β is the second weight coefficient, γ is the third weight coefficient, is the real-time coefficient of task processing of the i-th node, The calculation model is: ; S5. Calculation III: Calculate the task resource allocation ratio of the i-th node: , the task resource allocation of the i-th node is ; S6. Processing: According to the task resource allocation ratio of the i-th node Arrange non-real-time tasks in descending order according to the task resource allocation amount Distribute to other nodes for calculation; Set the minimum task resource allocation to , the node numbers after descending order are 1, 2, 3, ..., m, ..., M, where M = I. If according to The task resource allocation of the mth node in descending order is less than or equal to , after being allocated to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated; S7. Feedback: Upload the data allocation results of the pending tasks to the management end.

[0030] The working principle of this embodiment is: There are I gateway nodes in the area where resource allocation can be performed. The task data to be processed in I gateway nodes is obtained. The tasks to be processed include real-time tasks and N non-real-time tasks. The real-time task calculation amount of the i-th node is obtained. , the computational amount of the nth non-real-time task of the ith node , the maximum single-task computation of the i-th node , the average task calculation rate of the i-th node , by introducing the node load factor and the maximum concurrent task load Y, calculate the processing time of the task to be processed at the i-th node , thereby quantifying the concurrent pressure of the gateway node in real time and accurately identifying the risk of node overload. Compared with traditional static threshold detection, it can predict resource bottlenecks in advance and avoid system crashes caused by task accumulation. In burst traffic scenarios, the dynamic correction mechanism of the load factor significantly improves the elastic scaling capability of node resources, allowing key tasks to obtain resources first.

[0031] Get the task deadline of the pending task of the i-th node , system real time ,like , then it is judged that the resources are sufficient and the allocation result of the task data to be processed is uploaded to the management end for recording. Otherwise, it is judged that the resources are insufficient. When it is judged that the node resources are insufficient, the non-real-time task allocation weight of the i-th node is calculated according to the real-time task computing amount, non-real-time task computing amount and task type of the node. , thereby achieving multi-objective optimization, through priority weights The introduction of makes the scheduling priority of high-value business nodes significantly higher than that of ordinary task nodes; By designing the first weight coefficient, the second weight coefficient, and the third weight coefficient, it is possible to adapt to the precise adjustment of node priority, resource utilization, and real-time performance in different environments, and further adapt to different business requirements in differentiated scenarios, thereby improving practicality and reliability; by calculating the real-time coefficient of the task processing of the node It is conducive to ensuring the priority of real-time task processing. For example, when facing high real-time services such as emergency braking instructions and collaborative perception, the real-time task priority scheduling mechanism can effectively reduce the task interruption rate. At the same time, the resource release rate of non-real-time tasks is significantly improved, supporting dynamic expansion and contraction needs, and can perform dynamic task allocation, improve the utilization rate of resource allocation, and improve the management efficiency of gateway resources.

[0032] Calculate the task resource allocation ratio of the i-th node , and then calculate the task resource allocation of the i-th node , according to the task resource allocation ratio of the i-th node Arrange non-real-time tasks in descending order according to the task resource allocation amount Allocate the data to other nodes for calculation, so as to dynamically allocate tasks, improve the utilization rate of resource allocation, and improve the management efficiency of gateway resources; Set the minimum task resource allocation to , if according to The task resource allocation of the mth node in descending order is less than or equal to , after being allocated to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated, and uploads the allocation results of the pending task data to the management end. By setting the minimum resource allocation amount and descending allocation rules, resource fragmentation can be effectively curbed. While maintaining the allocation efficiency, the resource waste caused by the allocation of fragmented tasks can be reduced. This makes it suitable for batch processing scenarios of computing-intensive non-real-time tasks, thereby improving the utilization rate of resource allocation and the management efficiency of gateway resources.

[0033] The above disclosure is only a specific embodiment of the present invention, but the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A 5G gateway resource optimization system, characterized in that: Includes the following modules: Acquisition module: The output end is connected to the input end of calculation module I, and is used to obtain the task data to be processed in I gateway nodes. The tasks to be processed include real-time tasks and N non-real-time tasks. The calculation amount of the real-time task of the i-th node is , the computational amount of the nth non-real-time task of the ith node is , the maximum single-task computation amount of the i-th node is , the average task calculation rate of the i-th node is ; Calculation module I: The input end is connected to the output end of the acquisition module, and the output end is connected to the input end of the judgment module, which is used to calculate the processing time of the task to be processed at the i-th node , The calculation model is: ； Judgment module: The input end is connected to the output end of calculation module I, and the output end is connected to the input end of calculation module II and the input end of feedback module respectively, which is used to obtain the deadline of the pending task of the i-th node. , the system real time is ,like , then it is judged that the resources are sufficient and the feedback module is executed. Otherwise, it is judged that the resources are insufficient and the calculation II step is executed; Computation module II: The input end is connected to the output end of the judgment module, and the output end is connected to the input end of the calculation module III, which is used to calculate the non-real-time task allocation weight of the i-th node according to the real-time task calculation amount, the non-real-time task calculation amount and the task type. , The calculation model is: ； in, is the priority weight of the i-th node, ; Computing module III: The input end is connected to the output end of computing module II, and the output end is connected to the input end of the processing module, which is used to calculate the task resource allocation ratio of the i-th node: , the task resource allocation of the i-th node is ; Processing module: The input end is connected to the output end of the computing module III, and the output end is connected to the input end of the processing module, which is used to allocate the task resources according to the ratio of the i-th node. Arrange non-real-time tasks in descending order according to the task resource allocation amount Distribute to other nodes for calculation; Feedback module: The input end is connected to the output end of the judgment module and the output end of the processing module respectively, and is used to upload the distribution result of the task data to be processed to the management end.

2. A 5G gateway resource optimization system according to claim 1, characterized in that: In the computational module II, The calculation model is updated as follows: ； Among them, α is the first weight coefficient, β is the second weight coefficient, and γ is the third weight coefficient.

3. A 5G gateway resource optimization system according to claim 2, characterized in that: In the computational module II, The calculation model is updated as follows: ； in, is the real-time coefficient of task processing of the i-th node, The calculation model is: 。 4. A 5G gateway resource optimization system according to claim 1, characterized in that: In the calculation module I, The calculation model is updated as follows: ； in, is the node load factor, and Y is the maximum concurrent task capacity.

5. A 5G gateway resource optimization system according to any one of claims 1 to 4, characterized in that: In the processing module, the minimum task resource allocation is also set to , the node numbers after descending order are 1, 2, 3, ..., m, ..., M, where M = I. If according to The task resource allocation of the mth node in descending order is less than or equal to , after allocating to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated.

6. A 5G gateway resource optimization method, characterized in that: The following steps are involved: Get data: Get the pending task data in I gateway nodes. The pending tasks include real-time tasks and N non-real-time tasks. The real-time task calculation amount of the i-th node is , the computational amount of the nth non-real-time task of the ith node is , the maximum single-task computation amount of the i-th node is , the average task calculation rate of the i-th node is ; Calculation I: Calculate the processing time of the task to be processed at the i-th node , The calculation model is: ； Judgment: Get the deadline for the pending tasks of the i-th node. , the system real time is ,like , then it is judged that the resources are sufficient and the feedback step is executed; otherwise, it is judged that the resources are insufficient and the calculation II step is executed; Calculation II: Calculate the non-real-time task allocation weight of the i-th node based on the real-time task calculation amount, the non-real-time task calculation amount and the task type , The calculation model is: ； in, is the priority weight of the i-th node, ; Calculation III: Calculate the task resource allocation ratio of the i-th node: , the task resource allocation of the i-th node is ; Processing: According to the task resource allocation ratio of the i-th node Arrange non-real-time tasks in descending order according to the task resource allocation amount Distribute to other nodes for calculation; Feedback: Upload the data allocation results of pending tasks to the management end.

7. A 5G gateway resource optimization method according to claim 6, characterized in that: In step II of the calculation, The calculation model is updated as follows: ； Among them, α is the first weight coefficient, β is the second weight coefficient, and γ is the third weight coefficient.

8. A 5G gateway resource optimization method according to claim 7, characterized in that: In step II of the calculation, The calculation model is updated as follows: ； in, is the real-time coefficient of task processing of the i-th node, The calculation model is: 。 9. A 5G gateway resource optimization method according to claim 6, characterized in that: In the calculation step I, The calculation model is updated as follows: ； in, is the node load factor, and Y is the maximum concurrent task capacity.

10. A 5G gateway resource optimization method according to any one of claims 6 to 9, characterized in that: In the processing step, the minimum task resource allocation is also set to , the node numbers after descending order are 1, 2, 3, ..., m, ..., M, where M = I. If according to The task resource allocation of the mth node in descending order is less than or equal to , after allocating to the m+1th node, the allocation to the next node stops, and the m+1th node completes the remaining non-real-time tasks to be allocated.

Images