CN-120711033-B - Hierarchical distributed computing communication system and method based on cloud edge end integration

CN120711033BCN 120711033 BCN120711033 BCN 120711033BCN-120711033-B

Abstract

The invention belongs to the technical field of distributed computing and multistage network architecture, and discloses a hierarchical distributed computing communication system and method based on cloud edge end integration; the system comprises a task grading module, a node grading module and a node grading module, wherein the task grading module is used for grading an action computing task into a high-grade task, a medium-grade task and a low-grade task according to the emergency degree and the complexity of the task, the node grading module is used for grading the computing nodes into three types according to the geographic position, the computing capacity and the communication delay of each computing node in a network, the remote nodes, the edge nodes and the center nodes are used for improving the cooperation efficiency of the remote computing nodes and the edge computing nodes and realizing efficient real-time action data processing and action decision, and the coordination processing is carried out among the remote computing nodes, the edge computing nodes and the center nodes by grading the action computing task into different grades so as to ensure the real-time performance, the accuracy and the reliability of action computing.

Inventors

LAI GUOJUN
LIANG YI
Guo Yaze
WANG HONGYU
GUO FENG

Assignees

中国人民解放军陆军航空兵学院

Dates

Publication Date: 20260505
Application Date: 20250620

Claims (6)

1. Hierarchical distributed computing communication system based on cloud end is integrative, characterized by comprising: The task classification module is used for classifying the action calculation tasks into high-level tasks, medium-level tasks and low-level tasks according to the emergency degree and the complexity of the tasks, wherein the emergency degree is determined by the weighted calculation of time sensitivity, influence degree and resource demand urgency degree, and the complexity is determined by the weighted calculation of structural complexity, dynamic change rate and prejudgement degree; The node dividing module is used for dividing the computing nodes into three types, namely a remote node, an edge node and a central node according to the geographic position, the computing capacity and the communication delay of each computing node in the network; the system comprises a node classification module, an edge node, a central node and a central node, wherein the node classification module is used for classifying nodes according to a calculation formula, wherein the distance is 0.3+communication delay is 0.2+calculation capacity, and the node classification module is used for carrying out node classification according to a calculation formula, the remote node is positioned in a command center or a rear base, has strong data storage and analysis capacity and is mainly used for storing low-instantaneity tasks of historical data, data mining and strategy research, the edge node is deployed in an action front or an area close to the action, has strong real-time calculation capacity and is mainly responsible for processing high-instantaneity tasks, including fire striking calculation, target tracking and threat prediction, and the central node is positioned between the edge node and the remote node and serves as a central coordination system and is used for distributing tasks among all parts of the system and responsible for processing middle-level tasks in the parts; the task allocation and coordination calculation module comprises an initial task allocation unit and a task scheduling and coordination unit, wherein the initial task allocation unit is used for allocating calculation nodes according to the grades of the tasks, and particularly, the initial task allocation unit is used for allocating high-grade tasks to edge nodes for quick calculation, processing medium-grade tasks by a central node and the edge nodes together and allocating low-grade tasks to remote nodes for processing; The data synchronization and consistency maintenance module comprises a local data processing unit, an asynchronous data synchronization unit and a consistency maintenance unit, wherein the consistency maintenance unit ensures the data consistency among all nodes through a distributed database and a consistency algorithm; the self-adaptive scheduling and fault-tolerant mechanism module comprises a self-adaptive scheduling unit and a fault-tolerant mechanism unit, wherein the self-adaptive scheduling unit is used for autonomously scheduling tasks according to action environment changes, task priority changes and node load changes, the task priority changes comprise target adjustment, resource changes, external factors, risk assessment, stakeholder feedback, progress update and technical changes, and the node load changes comprise task allocation changes, data flow fluctuation, resource dynamic adjustment, hardware performance differences, network condition changes and faults; The data transmission communication module is used for transmitting the acquired data to the edge equipment according to a preset format, analyzing and processing the data by utilizing the cloud computing center and transmitting the data to the end user, and comprises a data acquisition and transmission unit, a communication architecture and model construction unit, an edge equipment communication and intention recognition unit and a data processing and matching unit, wherein the edge equipment communication and intention recognition unit is used for analyzing the communication behaviors among the edge equipment to infer intention.
2. The cloud-edge integration-based hierarchical distributed computing communication system of claim 1, wherein, at the data synchronization and consistency maintenance module: the local data processing unit is used for preprocessing and simplifying calculation of the edge nodes through data provided by the local sensor and the monitoring equipment, wherein the preprocessing comprises data cleaning, data conversion and data dimension reduction; The asynchronous data synchronization unit is used for asynchronously transmitting partial calculation results to the center and the remote nodes after the edge nodes complete calculation.
3. The cloud-edge integration-based hierarchical distributed computing communication system of claim 1, wherein the multiple copies store multiple copies for storing data at multiple locations or devices by synchronous replication or asynchronous replication; The hot backup is divided into online backup or dynamic backup, and the hot backup refers to backup performed under the condition that a database normally operates and can be accessed by a user; The node fault detection mechanism comprises heartbeat detection, timeout retransmission and confirmation mechanism, log-based and monitoring information analysis, fault injection test and distributed consistency protocol detection.
4. The hierarchical distributed computing communication system based on the cloud edge integration according to claim 1, wherein the data acquisition and transmission unit is used for completing data acquisition through a sensor, transmitting the data to the edge device in a wireless transmission mode according to a preset format, analyzing and processing the data by using a cloud computing center, and transmitting the processed data to an end user; the communication architecture and model construction unit is used for constructing a model by adopting a distributed data transmission framework, wherein the communication architecture is used for realizing communication between cloud computing and edge equipment and mobile terminals; the edge device communication and intention recognition unit analyzes communication behaviors between edge devices through machine learning or deep learning technologies to infer intention, including cluster analysis, timing analysis and anomaly detection.
5. The cloud-edge-integration-based hierarchical distributed computing communication system according to claim 4, wherein the data processing and matching unit is configured to perform data processing on the collected data, and control a plurality of edge devices to disperse massive data collection work on different devices by using a cloud computing center, where when data is filtered for the second time, data meeting requirements is screened out according to a matching rule or condition of exact matching, fuzzy matching, range matching, regular expression matching or similarity matching.
6. The cloud edge end integration-based hierarchical distributed computing communication method is characterized by being used for realizing the cloud edge end integration-based hierarchical distributed computing communication system as claimed in any one of claims 1 to 5, and comprises the following steps: S1, dividing an action calculation task into a high-level task, a medium-level task and a low-level task according to the emergency degree and the complexity of the task, wherein the emergency degree is calculated by setting weights for time sensitivity, influence degree and resource demand urgency degree, and the complexity is calculated by setting weights for structural complexity, dynamic change rate and prejudgement degree; S2, dividing the computing nodes into three types, namely a remote node, an edge node and a central node according to the geographic position, the computing capacity and the communication delay of each computing node in the network, and classifying the computing nodes according to a computing formula, namely the distance multiplied by 0.3+the communication delay multiplied by 0.2+the computing capacity multiplied by 0.5; S3, performing task allocation and cooperative calculation, wherein the task allocation and cooperative calculation comprises an initial task allocation unit and a task scheduling and cooperative unit, the initial task allocation unit is used for allocating calculation nodes according to the grades of tasks, and the task scheduling and cooperative unit is used for dynamically adjusting task allocation by monitoring the calculation load and the task emergency degree of each node; S4, carrying out data synchronization and consistency maintenance, wherein the data synchronization and consistency maintenance comprises a local data processing unit, an asynchronous data synchronization unit and a consistency maintenance unit, and specifically comprises local data processing, asynchronous data synchronization and consistency maintenance through a Paxos or Raft algorithm; S5, performing self-adaptive scheduling and fault tolerance, wherein the self-adaptive scheduling unit is used for autonomously scheduling tasks according to the change of the action environment, the change of the task priority and the change of the node load, and the fault tolerance mechanism unit is used for automatically performing task redistribution through a multi-copy storage, hot backup and node fault detection mechanism when part of nodes fail; S6, transmitting the acquired data to the edge equipment according to a preset format, analyzing and processing the data by utilizing the cloud computing center, and transmitting the data to the terminal user, wherein the data comprise edge equipment communication intention identification and data secondary filtering matching.

Description

Hierarchical distributed computing communication system and method based on cloud edge end integration Technical Field The invention relates to the technical field of distributed computing and multistage network architecture, in particular to a cloud side end integration-based hierarchical distributed computing communication system and method. Background With the development of science and technology, people enter into the era of 'everything interconnection', and mobile equipment and internet of things equipment which need to be connected show explosive growth. Devices such as smart home, vehicle network, health wear, smart manufacturing and the like need to complete complex operations through a cloud computing center, but cannot complete the operations by themselves. There are large data processing delays and data transmission delays between the cloud computing center and the intelligent devices, and in order to solve the problem, an edge server is added between the cloud computing center and the on-site intelligent devices. In this way, events can be handled with low latency and high efficiency. The edge computing industry alliance ECC and the industrial internet industry alliance AII jointly release an edge computing reference architecture 3.0, which is a three-layer architecture in total, namely a cloud computing layer, an edge computing layer and a field device computing layer. According to IDC report of International data company, global marginal calculation expenditure reaches 1760 hundred million dollars in 2022, annual growth rate reaches 14.8%, and China marginal calculation market reaches 1803.7 hundred million Yuanhong national coin scale in 2024; However, the existing mobile computing communication system is not efficient in real-time mobile data processing and mobile decision making, and is inconvenient to coordinate the mobile computing tasks among remote, edge and central nodes so as to ensure the real-time performance, accuracy and reliability of mobile computing; In view of the above, the present invention proposes a cloud-edge integration-based hierarchical distributed computing communication system and method to solve the above-mentioned problems. Disclosure of Invention In order to overcome the defects in the prior art and achieve the purposes, the invention provides a cloud edge end integration-based hierarchical distributed computing communication system and method, comprising the following steps: The task grading module is used for dividing the action calculation task into a high-grade task, a medium-grade task and a low-grade task according to the emergency degree and the complexity of the task; The node dividing module is used for dividing the computing nodes into three types, namely a remote node, an edge node and a central node according to the geographic position, the computing capacity and the communication delay of each computing node in the network; The task allocation and coordination calculation module comprises an initial task allocation unit and a task scheduling and coordination unit, wherein the initial task allocation unit is used for allocating calculation nodes according to the grades of the tasks, and the task scheduling and coordination unit is used for dynamically adjusting task allocation by monitoring the calculation load and the task emergency degree of each node; the data synchronization and consistency maintenance module comprises a local data processing unit, an asynchronous data synchronization unit and a consistency maintenance unit; The self-adaptive scheduling and fault-tolerant mechanism module comprises a self-adaptive scheduling unit and a fault-tolerant mechanism unit, wherein the self-adaptive scheduling unit is used for autonomously scheduling tasks according to the change of the action environment, the change of the task priority and the change of the node load, and the fault-tolerant mechanism unit is used for automatically reallocating the tasks when part of nodes fail; The data transmission communication module is used for transmitting the acquired data to the edge equipment according to a preset format, analyzing and processing the data by utilizing the cloud computing center and transmitting the data to the terminal user. Further, at the task ranking module: and performing rapid calculation on the action calculation tasks divided into high-level tasks at the edge nodes, processing the action calculation tasks divided into medium-level tasks by the central node and the edge nodes together, and distributing the action calculation tasks divided into low-level tasks to the remote nodes for processing. Further, at the node dividing module: The remote node is used for storing historical data, data mining and low real-time tasks of strategy research; the edge node is used for processing the task with high real-time performance; The central node is located between the edge node and the remote node and serves as a central coordination system, and task