CN115776445B - Traffic migration-oriented node identification method, device, equipment and storage medium - Google Patents
Traffic migration-oriented node identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN115776445B CN115776445B CN202310093567.7A CN202310093567A CN115776445B CN 115776445 B CN115776445 B CN 115776445B CN 202310093567 A CN202310093567 A CN 202310093567A CN 115776445 B CN115776445 B CN 115776445B
- Authority
- CN
- China
- Prior art keywords
- node
- traffic
- nodes
- request
- cluster
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Abstract
The invention belongs to the technical field of computers, and discloses a node identification method, device and equipment for traffic migration and a storage medium. The method comprises the following steps: determining the request processing capacity of each node according to a preset evaluation mode, and acquiring the request processing quantity of each node; determining a topology center node according to the request processing capacity of each node and the service requirements of users; determining a flow center node according to the request processing quantity of each node; and completing node identification according to the traffic center node and the topology center node. By the method, the identification of the key nodes for traffic migration in the edge X heterogeneous node network is realized according to the traffic center node and the topology center node, so that a user can adjust traffic according to the traffic center node and the topology center node in the follow-up process, and traffic identification and traffic migration tasks can be completed in real time by the edge X system in the process of traffic operation.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a storage medium for identifying a node for traffic migration.
Background
With the development of 5G technology, artificial intelligence, computing hardware, and other fields, a set of computing frameworks that are deployed closer to the production site, the data source (i.e., edge side), and that can provide low latency, real-time performance have emerged. Emerging edge computing technologies can provide low-latency real-time computing, processing services for industry users. The edge x is a "man-in-the-middle" interposed between the edge physical sensing device and the executing device, i.e., the cloud information system, and is specifically denoted as edge x foundation. The EdgeX platform supports and encourages fast-growing internet of things solution providers to work cooperatively in a unified ecosystem to reduce uncertainty in application development, speed deployment time, and promote scaling.
Edge service provider can monitor the equipment in the real world more easily, send executable instructions, collect edge sensor data, and complete cloud edge equipment interaction. The edge sensors and the cloud servers form a complex edge X heterogeneous node network, meanwhile, the process request flow for completing specific service flows in mutually heterogeneous network nodes, and the flow in the edge X may change in real time due to network fluctuation, equipment downtime and other reasons, so that the node request processing capacity and the request flow in the edge X network are not matched. Therefore, in the use process of edge x, it is necessary to monitor the flow request situation in the heterogeneous node network in real time and design an algorithm to identify the key nodes causing traffic anomalies.
The iteration time of the edge X framework is not long at present, and related theoretical researches are mainly focused on current situation analysis of application scenes, model selection, platform characteristics and the like of an edge computing platform, and a series of application system designs such as an Internet of things gateway monitoring system, an intelligent home system architecture design and the like are realized by utilizing the edge X edge platform. For the specific scene of edge X, the work research related to traffic migration is less in the key node identification. The work in the related direction is mainly oriented to the traditional internet scene, and is different from the heterogeneous node network of the edge X, and the heterogeneous property among the nodes is less prominent than the edge X nodes in the internet scene although the processing capacity among the nodes is different. Therefore, a method for identifying key nodes of the EdgeX heterogeneous node network facing traffic migration is needed.
Disclosure of Invention
The invention mainly aims to provide a method, a device, equipment and a storage medium for identifying nodes facing traffic migration, which aim to solve the technical problem of how to identify key nodes of an edge X heterogeneous node network facing traffic migration in the prior art.
In order to achieve the above object, the present invention provides a traffic migration-oriented node identification method, which includes:
Determining the request processing capacity of each node according to a preset evaluation mode, and acquiring the request processing quantity of each node;
determining a topology center node according to the request processing capacity of each node and the service requirements of users;
determining a flow center node according to the request processing quantity of each node;
and completing node identification according to the traffic center node and the topology center node.
Optionally, the determining the topology center node according to the request processing capability of each node and the user service requirement includes:
calculating the modularity according to the request processing capacity of each node, and determining the community modularity of each node;
determining a plurality of request capability clusters according to community modularity of each node;
calculating the compact weight of each request capacity cluster to obtain a cluster center node of each request capacity cluster;
and determining the topological center node according to the clustering center nodes of the request capacity clusters and the user service requirements.
Optionally, the calculating the tight weight of each request capability cluster to obtain a cluster center node of each request capability cluster includes:
acquiring the total number of nodes and the number of clustered nodes clustered by each request capability;
determining the compact weight of each node in each request capacity cluster according to the node cluster distance, the total number of the nodes and the cluster node number of each request capacity cluster;
Sorting the compact weights of the nodes in the request capacity clusters to obtain a sorting result;
and determining a clustering center node of each request capacity cluster according to the sequencing result.
Optionally, the determining the traffic center node according to the request processing number of each node includes:
initializing labels of all nodes to obtain node labels of all nodes;
clustering nodes according to the node labels of the nodes and the request processing quantity of the nodes to obtain a plurality of traffic clusters;
calculating the compact weight of each flow cluster to obtain a cluster center node of each flow cluster;
and determining the flow center node according to the cluster center node of each flow cluster.
Optionally, the clustering of the nodes according to the node labels of the nodes and the request processing number of the nodes to obtain a plurality of traffic clusters includes:
counting the request processing quantity of adjacent nodes of each node according to the node label of each node;
determining the weight of a target node according to the request processing quantity of the adjacent nodes of each node;
updating the node labels of all the nodes according to the weight of the target node to obtain update labels;
and clustering the nodes according to the update labels of the nodes to obtain a plurality of traffic clusters.
Optionally, after the node identification is completed according to the traffic center node and the topology center node, the method further includes:
determining a traffic migration node according to the traffic center node and the center label of the topology center node;
and when receiving the traffic migration task, completing the traffic migration task according to the traffic migration node.
Optionally, when receiving a traffic migration task, the completing the traffic migration task according to the traffic migration node includes:
when receiving a traffic migration task, acquiring the request quantity of a topology center node;
determining a request consumption parameter according to the traffic migration task;
and completing the traffic migration task according to the request consumption parameter, the request quantity, the traffic migration node and a preset evaluation mode.
In addition, in order to achieve the above object, the present invention further provides a traffic migration-oriented node identification device, where the traffic migration-oriented node identification device includes:
the acquisition module is used for determining the request processing capacity of each node according to a preset evaluation mode and acquiring the request processing quantity of each node;
the determining module is used for determining a topology center node according to the request processing capacity of each node and the user service requirement;
The determining module is further used for determining a flow center node according to the request processing quantity of each node;
and the completion module is used for completing node identification according to the traffic center node and the topology center node.
In addition, in order to achieve the above object, the present invention further provides a traffic migration-oriented node identification device, where the traffic migration-oriented node identification device includes: the system comprises a memory, a processor and a traffic migration oriented node identification program stored on the memory and capable of running on the processor, wherein the traffic migration oriented node identification program is configured to realize the traffic migration oriented node identification method.
In addition, in order to achieve the above object, the present invention further proposes a storage medium, on which a traffic migration-oriented node identification program is stored, which when executed by a processor implements the traffic migration-oriented node identification method as described above.
The invention determines the request processing capacity of each node according to a preset evaluation mode and acquires the request processing quantity of each node; determining a topology center node according to the request processing capacity of each node and the service requirements of users; determining a flow center node according to the request processing quantity of each node; and completing node identification according to the traffic center node and the topology center node. Through the method, the request processing capacity of each node is determined according to the preset evaluation mode, the topology center node is determined based on the request processing capacity and the user service demands, the flow center node is determined based on the request processing quantity of each node, and finally, the identification of the key nodes for flow migration in the edge X heterogeneous node network is realized according to the flow center node and the topology center node, so that the user can adjust the flow subsequently according to the flow center node and the topology center node, and the edge X system can complete the flow identification and the flow migration tasks in real time in the service operation process.
Drawings
FIG. 1 is a schematic structural diagram of a node identification device for traffic migration in a hardware running environment according to an embodiment of the present invention;
FIG. 2 is a flow chart of a first embodiment of a method for identifying a node for traffic migration according to the present invention;
FIG. 3 is a schematic overall flow chart of an embodiment of a method for identifying a node for traffic migration according to the present invention;
FIG. 4 is a flowchart illustrating a second embodiment of a method for identifying a node for traffic migration according to the present invention;
fig. 5 is a block diagram of a first embodiment of a traffic migration oriented node identification device according to the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a node identification device for traffic migration in a hardware running environment according to an embodiment of the present invention.
As shown in fig. 1, the traffic migration-oriented node identification device may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the structure shown in fig. 1 does not constitute a limitation of the node identification device facing traffic migration, and may include more or fewer components than shown, or may combine certain components, or may be a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a node identification program for traffic migration may be included in the memory 1005 as one type of storage medium.
In the traffic migration-oriented node identification device shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the traffic migration-oriented node identification device of the present invention may be disposed in the traffic migration-oriented node identification device, where the traffic migration-oriented node identification device invokes the traffic migration-oriented node identification program stored in the memory 1005 through the processor 1001, and executes the traffic migration-oriented node identification method provided by the embodiment of the present invention.
The embodiment of the invention provides a node identification method facing traffic migration, and referring to fig. 2, fig. 2 is a flow diagram of a first embodiment of the node identification method facing traffic migration.
The node identification method facing to the flow migration comprises the following steps:
step S10: and determining the request processing capacity of each node according to a preset evaluation mode, and acquiring the request processing quantity of each node.
It should be noted that, the execution body of the embodiment is a terminal device, and the terminal device may be an intelligent terminal such as a computer, a tablet, etc., which is not limited in this embodiment. The terminal equipment determines the request processing capacity of each node according to a preset evaluation method, acquires the request processing quantity of each node, determines the topology center node according to the request processing capacity of each node and the user service requirement, determines the flow center node according to the request processing quantity of each node, and completes node identification according to the flow center node and the topology center node.
It can be understood that for the problems of traffic migration, load balancing and the like, the traditional research method mainly adopts methods of traffic detour forwarding, bandwidth lease, dynamic auction and the like according to the bandwidth conditions among nodes, and realizes adjustment and redesign in aspects of network topology, network protocol, resource allocation mechanism and the like. Load balancing in the internet scenario is mainly oriented to uniformly distributing traffic to a plurality of different available paths to avoid network congestion, and provides users with an information network infrastructure with better network quality of service (Quality of Service, qoS) and network quality of experience (Quality of Experience, qoE). In addition to the traditional bandwidth factor, the EdgeX heterogeneous node network (i.e., edgeX Foundation heterogeneous node network) needs to consider a series of other factors such as node computing capability, node storage capability, node topology, etc., so that the traffic migration-oriented node identification method of the embodiment is provided. The node identification method facing to flow migration in this embodiment may be deployed in a system management service of EdgeX (EdgeX Foundry), where the service may communicate with other systems and services to obtain configuration and status of the service and send management requests such as start and stop to the service, so that nodes of the entire system in the EdgeX may be managed. The service provides an access point of an external management system, so that the node identification method facing traffic migration in this embodiment can be accessed into the external management system or the identification method can be embedded into other management systems and then accessed into the edge x, and the edge x referred to in the subsequent embodiments is specifically expressed as edge x foundation.
In a specific implementation, the preset evaluation mode refers to designing a request processing capability evaluation function applicable to a node in the edge X network according to a specific service requirement scene of the edge X, and the preset evaluation mode can adopt a simple linear weighting function or can adopt prediction technologies such as machine learning to complete the node in the heterogeneous node networkFitting of the request capability function. In this embodiment, the predetermined evaluation mode may be +.>Wherein->Representing the number of node factors involved in the node processing request, < +.>Representing different node factors involved in processing a request by a node, e.g.>Representing node->A certain number of CPU clocks requesting the number of processing completions, etc.>Representing node->Can accommodate the number of request processes, +.>Representing node->The number of requests that can be processed in parallel; />Representing node factor->The corresponding weights, all weights should satisfy。
It should be noted that, the request processing capability of each node in the edge x network is calculated according to a preset evaluation mode, and the number of request processing of each node refers to the number of request processing in the node counted in real time by each node in the edge x network.
Step S20: and determining a topology center node according to the request processing capacity of each node and the service requirements of the users.
It should be noted that, a suitable cluster hierarchy is selected according to the user service requirement, and a corresponding node is selected from the selected cluster hierarchies according to the request processing capability of each node, so that a topology center node of the corresponding cluster hierarchy in the edge x network can be obtained.
Step S30: and determining a flow center node according to the request processing quantity of each node.
It should be noted that, the traffic center node refers to a request traffic center in the EdgeX network. The traffic center node in the edge x network may be determined based on the number of request processing for each node in the edge x network.
It may be appreciated that, to determine an accurate traffic center node according to the number of request processing of each node, further, the determining a traffic center node according to the number of request processing of each node includes: initializing labels of all nodes to obtain node labels of all nodes; clustering nodes according to the node labels of the nodes and the request processing quantity of the nodes to obtain a plurality of traffic clusters; calculating the compact weight of each flow cluster to obtain a cluster center node of each flow cluster; and determining the flow center node according to the cluster center node of each flow cluster.
In a specific implementation, initializing labels of all nodes to obtain node labels of all nodes, wherein the specific process is as follows: each node in edge X networkInitializing node tag with a unique tag +.>And->。
It should be noted that, the number of requests of each node is the number of requests processed in each node counted in real time, and the nodes are recordedTo node->The number of requests propagated->. And clustering the nodes according to the node labels of the nodes and the request processing quantity of the nodes, so as to obtain clustering results of a plurality of request flows in the edge X network, wherein the clustering results of the plurality of request flows are a plurality of flow clusters.
It will be appreciated that the tight centrality weight value for each node is calculated in each traffic clusterThe node with the highest tight centrality weight value in each flow cluster is the cluster center node of each flow cluster, and the cluster center node is the flow center node. Wherein (1)>,/>Representing the number of nodes in the traffic cluster to which the nodes belong; />Representing the total number of nodes in the EdgeX network; />Representing node->And node->Is used for controlling the flow distance of the air conditioner,the greater the traffic transmission between nodes, the smaller the distance that a node depends on the same traffic cluster.
In a specific implementation, in order to perform accurate traffic clustering according to the node labels and the request processing numbers of the nodes, further, the performing node clustering according to the node labels and the request processing numbers of the nodes to obtain a plurality of traffic clusters includes: counting the request processing quantity of adjacent nodes of each node according to the node label of each node; determining the weight of a target node according to the request processing quantity of the adjacent nodes of each node; updating the node labels of all the nodes according to the weight of the target node to obtain update labels; and clustering the nodes according to the update labels of the nodes to obtain a plurality of traffic clusters.
The method is characterized in that the request processing quantity of the adjacent nodes of each node is counted according to the node label of each node, and the specific process is as follows: each node in the edge X networkTo node->Propagation node tag->And request quantity->Node->Receiving adjacent node->Node tag of->Corresponding request quantity ∈ ->Statistics and node->Other nodes with the same tag->Request quantity +.>Each node->Other nodes with the same tag->Request quantity +.>I.e. the number of request processing of the neighboring node of each node, and +. >。
It can be understood that the target node weight is determined according to the number of the request processing of the adjacent nodes of each node, and the node label of each node is updated according to the target node weight to obtain the update label, which comprises the following specific processes: after the request processing quantity of the adjacent nodes of each node is obtained, the weight of each adjacent node label can be determined, wherein the weight of each adjacent node label is the node weight, and each nodeSelect the node with the greatest weight +.>The adjacent node label corresponding to the maximum node weight is the update label, and a specific calculation formula can be: />The steps are repeated repeatedly: counting the request processing quantity of adjacent nodes of each node according to the node labels of each node, and determining the weight of a target node according to the request processing quantity of the adjacent nodes of each node; updating the node labels of all nodes according to the weight of the target node to obtain updated labels until the densely connected node groups reach consensus on one unique label, only a small number of labels are reserved in the edge X network at the moment, and nodes with the same label are +.>And node->And in the same request traffic cluster, obtaining a plurality of traffic clusters.
Step S40: and completing node identification according to the traffic center node and the topology center node.
It should be noted that, after determining the traffic center node and the topology center node in the EdgeX network, the identification of the key node facing the traffic migration in the EdgeX network is completed.
It can be understood that, in order to facilitate subsequent adjustment of the traffic, improve overall performance of the system, and propose a migration policy responsive to the traffic migration task, further, after completing node identification according to the traffic center node and the topology center node, the method further includes: determining a traffic migration node according to the traffic center node and the center label of the topology center node; and when receiving the traffic migration task, completing the traffic migration task according to the traffic migration node.
In a specific implementation, for a scene that a traffic center node and a topology center node in an edge X network are not coincident, traffic transmission nodes which cause the two centers to be not coincident are identified, and traffic migration problems are modeled to guide the nodes to complete traffic migration. The main function of the part is to obtain the flow clustering conditions of a flow center and a topology center in an edge X network by using a semi-supervised label propagation clustering method based on label pushing; and identifying key nodes for traffic migration according to the clustered edge nodes and modeling and guiding the nodes to complete traffic migration tasks according to traffic migration problems.
It should be noted that, according to the center labels of the traffic center node and the topology center node, searching the nodes adjacent to the topology center node cluster in the traffic center node cluster, where the nodes are used as the key nodes of traffic migration in the edge x network, and the key nodes of traffic migration are the traffic migration nodes.
It can be understood that each node in the edge x network counts the number of request processing in the node in real time, and records the nodeTo node->The number of requests transmitted->. Meanwhile, each traffic center node and topology center node in the edge X network uses unique label initialization node label ++>And->Nodes with labels in edge network +.>Finding dependent adjacency node in adjacency nodes +.>To the adjacent node->Propagation node tag->And request quantity->Node->Receiving adjacent node->Node tag of->Corresponding request quantity ∈ ->Statistics and node->Other nodes with the same tag->Request quantity +.>I.e. the number of request processing of adjacent nodes of each node, andeach node->Select the node with the greatest weight +.>The adjacent node labels of (a) are matched and updated, and a specific calculation formula can be as follows: />And repeating the steps repeatedly until each node in the edge X network obtains the node label of the traffic center node or the topology center node, wherein the node label of the traffic center node or the topology center node is the center label.
In a specific implementation, when a traffic migration task is received, a linear programming tool, a genetic algorithm, a particle swarm algorithm and the like are used in a system management service of the edge X to solve a traffic migration problem, and a traffic migration node is guided to complete the traffic migration task.
It should be noted that, in order to accurately complete a traffic migration task according to a traffic migration node, further, when the traffic migration task is received, the completing the traffic migration task according to the traffic migration node includes: when receiving a traffic migration task, acquiring the request quantity of a topology center node; determining a request consumption parameter according to the traffic migration task; and completing the traffic migration task according to the request consumption parameter, the request quantity, the traffic migration node and a preset evaluation mode.
It can be appreciated that traffic migration tasks in the EdgeX network are modeled as a multi-constraint multi-knapsack problem. Wherein the request tasks related to different traffic center nodes can be regarded as different articles and usedRepresenting the total number of requests; the need to consume CPU, bandwidth, memory and other resources in the process of completing the request can be regarded as various constraints in knapsack problem, and +. >Representing the number of resource constraints; the requests of the traffic center can be migrated to different topology center nodes, all of which constitute a set +.>One can consider the problem of backpacks to involve multiple backpacks for selective placement of items. The value of the cost function of the traffic migration can be evaluated using the load product on the topology centre,/>. For the traffic migration task, its mathematical model is described as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,topology center node after traffic migration +.>The number of requests to be processed, +.>Is a topology center node->Is/are/is/are evaluated in a preset evaluation mode>For request->Whether to place topology center node->Binary variable of processing->For request->The amount of j-th resource to be consumed, +.>Is a topology center node->Can be used for processing requestsThe j-th resource number. The limiting resources of the node in specific use include: />Is a topology center node->Handling requests->Occupied CPU utilization, +.>Is a topology center node->Handling requests->Occupied memory, in->Is a topology center node->Handling requests->Occupied network bandwidth, < > on->Is a topology center node->Handling requests->Occupied storage space.
In particular implementations, the request consumption parameters include, but are not limited to 、/>、/>、And modeling and solving the flow migration problem by using a linear programming tool, a genetic algorithm, a particle swarm algorithm and the like in the system management service of the edge X, and guiding the flow migration node to complete the flow migration task.
It should be noted that, as shown in fig. 3, a request processing capability evaluation function is designed to determine a preset evaluation mode, request processing capability evaluation is performed on each node, each node determines a topology center node of each request capability cluster based on greedy expansion cluster optimization modularity, initializes each node label in the edge x network, defines a request flow distance based on the request processing number of each node, calculates a tight center weight of each flow cluster, determines a flow center node of each flow cluster, forms node clusters of the topology center node and the flow center node according to flow migration conditions, identifies key nodes of flow migration according to cluster edges, and guides the key nodes to complete flow migration work.
In the embodiment, software evolution information of the edge X is obtained; constructing a dynamic evolution graph according to the software evolution information; performing weight calculation according to the dynamic evolution graph, and determining key nodes; and determining the key software of the edge X according to the key node. By the method, the dynamic evolution graph is constructed based on the software evolution information, the weight calculation is carried out according to the dynamic evolution graph, so that the key node is determined, the key software of the edge X in the evolution process can be determined according to the key node, the identification of the key software of the edge X in the evolution process is realized, the software module mainly surrounded by developers and maintainers in the edge X is determined, and the module key reference is provided for the development and maintenance of the follow-up software of the edge X.
Referring to fig. 4, fig. 4 is a flow chart of a second embodiment of a node identification method for traffic migration according to the present invention.
Based on the above-mentioned first embodiment, the step S20 in the node identification method for traffic migration according to this embodiment includes:
step S21: and calculating the modularity according to the request processing capacity of each node, and determining the community modularity of each node.
It should be noted that the community modularity measures the degree of compactness of the node clusters formed by clustering different nodes. After the request processing capacity of each node is determined according to a preset evaluation mode, module calculation is performed based on the request processing capacity of each node, and community module of each node is determined, wherein the specific process is as follows: each node takes the node as own community label, traverses all neighbor nodes of the node, tries to update the own community label into the community label of the neighbor node, and selects the label with the largest module increment as the own community label based on the greedy idea, wherein the calculation formula of the community module Q is as follows:wherein->For node->And node->Can be determined by the node +.>And node->Is obtained by summing the request processing capacities of (a), m is the sum of all relation weights in the edge X network,/o >And->For node->And node->Sum of all relation weights of +.>For node->And node->The identification function of whether the same community is 1 or not, and the identification function of whether the same community is not 0.
Step S22: and determining a plurality of request capability clusters according to the community modularity of each node.
It should be noted that, in the EdgeX heterogeneous node network, the nodes with strong adjacent request processing capability will preferentially form a cluster, and finally form a plurality of preliminary clustering results with strong request processing capability in the EdgeX network, each community in the EdgeX network is combined into a new super node, the edge weight of the super node is the sum of the edge weights of all the nodes in the original community, so as to form a new EdgeX network with coarser granularity, and the steps of calculating the community modularity of each node and forming the new EdgeX network with coarser granularity are iterated continuously until the EdgeX network reaches global optimum and cannot be further modularized, so as to obtain a request processing capability clustering result of the multi-level EdgeX heterogeneous node network, and the request processing capability clustering result of the multi-level EdgeX heterogeneous node network is a plurality of request capability clusters in the EdgeX network.
Step S23: and calculating the compact weight of each request capacity cluster to obtain a cluster center node of each request capacity cluster.
The tight centrality weight value of each node in each request capacity cluster is calculated, and the cluster center node of each request capacity cluster is obtained according to the tight centrality weight value of each node.
It may be appreciated that, to accurately obtain the cluster center node of each request capability cluster, further, the calculating the tight weight of each request capability cluster to obtain the cluster center node of each request capability cluster includes: acquiring the total number of nodes and the number of clustered nodes clustered by each request capability; determining the compact weight of each node in each request capacity cluster according to the node cluster distance, the total number of the nodes and the cluster node number of each request capacity cluster; sorting the compact weights of the nodes in the request capacity clusters to obtain a sorting result; and determining a clustering center node of each request capacity cluster according to the sequencing result.
In a specific implementation, the node clustering distance of each request capability cluster is the shortest distance between two nodes in the edge X networkNodes in the edge X network obtain the node +.>And other nodes->Is>Wherein node- >And node->In the case of adjacency, the distance between two nodes is calculated using a formula。
It should be noted that, the total number of nodes refers to the number N of all nodes in the EdgeX network, and the number of clustered nodes refers to the number N of nodes in the request capability cluster to which the nodes belong. The tight weight is the tight centrality weight value of each node,/>。
It can be understood that after the tight weights of the nodes in the request capability clusters are obtained, the tight weights of the nodes are ordered, so that a forward ordering result of the tight weights of the nodes is obtained. And selecting the node with the highest tight weight from each request capability cluster according to the sequencing result as a cluster center node of the request capability cluster.
Step S24: and determining the topological center node according to the clustering center nodes of the request capacity clusters and the user service requirements.
It should be noted that, a suitable request capability cluster is selected according to the user service requirement, and the topology center node can be determined according to the cluster center node of the request capability cluster.
In the embodiment, the community modularity of each node is determined by carrying out module calculation according to the request processing capacity of each node; determining a plurality of request capability clusters according to community modularity of each node; calculating the compact weight of each request capacity cluster to obtain a cluster center node of each request capacity cluster; and determining the topological center node according to the clustering center nodes of the request capacity clusters and the user service requirements. By the method, a plurality of request capacity clusters are determined according to the community modularity of each node, the clustering center node is determined based on the tight weight of each request capacity cluster, and finally the topology center node is determined based on the clustering center node of each request capacity cluster and the user service requirement, so that the accuracy and pertinence of the determination of the topology center node are ensured.
In addition, referring to fig. 5, an embodiment of the present invention further provides a traffic migration-oriented node identification device, where the traffic migration-oriented node identification device includes:
the obtaining module 10 is configured to determine a request processing capability of each node according to a preset evaluation mode, and obtain a request processing number of each node.
A determining module 20, configured to determine a topology center node according to the request processing capability and the user service requirement of each node.
The determining module 20 is further configured to determine a traffic center node according to the number of request processing of each node.
A completion module 30, configured to complete node identification according to the traffic center node and the topology center node.
According to the method, the request processing capacity of each node is determined according to a preset evaluation mode, and the request processing quantity of each node is obtained; determining a topology center node according to the request processing capacity of each node and the service requirements of users; determining a flow center node according to the request processing quantity of each node; and completing node identification according to the traffic center node and the topology center node. Through the method, the request processing capacity of each node is determined according to the preset evaluation mode, the topology center node is determined based on the request processing capacity and the user service demands, the flow center node is determined based on the request processing quantity of each node, and finally, the identification of the key nodes for flow migration in the edge X heterogeneous node network is realized according to the flow center node and the topology center node, so that the user can adjust the flow subsequently according to the flow center node and the topology center node, and the edge X system can complete the flow identification and the flow migration tasks in real time in the service operation process.
In an embodiment, the determining module 20 is further configured to perform module degree calculation according to the request processing capability of each node, and determine a community module degree of each node;
determining a plurality of request capability clusters according to community modularity of each node;
calculating the compact weight of each request capacity cluster to obtain a cluster center node of each request capacity cluster;
and determining the topological center node according to the clustering center nodes of the request capacity clusters and the user service requirements.
In an embodiment, the determining module 20 is further configured to obtain a total number of nodes and a number of clustered nodes clustered by each request capability;
determining the compact weight of each node in each request capacity cluster according to the node cluster distance, the total number of the nodes and the cluster node number of each request capacity cluster;
sorting the compact weights of the nodes in the request capacity clusters to obtain a sorting result;
and determining a clustering center node of each request capacity cluster according to the sequencing result.
In an embodiment, the determining module 20 is further configured to initialize a label of each node to obtain a node label of each node;
clustering nodes according to the node labels of the nodes and the request processing quantity of the nodes to obtain a plurality of traffic clusters;
Calculating the compact weight of each flow cluster to obtain a cluster center node of each flow cluster;
and determining the flow center node according to the cluster center node of each flow cluster.
In an embodiment, the determining module 20 is further configured to count the number of request processing of the neighboring nodes of each node according to the node label of each node;
determining the weight of a target node according to the request processing quantity of the adjacent nodes of each node;
updating the node labels of all the nodes according to the weight of the target node to obtain update labels;
and clustering the nodes according to the update labels of the nodes to obtain a plurality of traffic clusters.
In an embodiment, the completion module 30 is further configured to determine a traffic migration node according to the traffic hub node and the hub label of the topology hub node;
and when receiving the traffic migration task, completing the traffic migration task according to the traffic migration node.
In an embodiment, the completion module 30 is further configured to obtain the number of requests of the topology center node when receiving the traffic migration task;
determining a request consumption parameter according to the traffic migration task;
and completing the traffic migration task according to the request consumption parameter, the request quantity, the traffic migration node and a preset evaluation mode.
Because the device adopts all the technical schemes of all the embodiments, the device at least has all the beneficial effects brought by the technical schemes of the embodiments, and the description is omitted here.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium stores a traffic migration oriented node identification program, and the traffic migration oriented node identification program realizes the steps of the traffic migration oriented node identification method when being executed by a processor.
Because the storage medium adopts all the technical schemes of all the embodiments, the storage medium has at least all the beneficial effects brought by the technical schemes of the embodiments, and the description is omitted here.
It should be noted that the above-described working procedure is merely illustrative, and does not limit the scope of the present invention, and in practical application, a person skilled in the art may select part or all of them according to actual needs to achieve the purpose of the embodiment, which is not limited herein.
In addition, technical details not described in detail in the present embodiment may refer to the node identification method for traffic migration provided in any embodiment of the present invention, which is not described herein.
Furthermore, it should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. Read Only Memory)/RAM, magnetic disk, optical disk) and including several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (8)
1. The traffic migration-oriented node identification method is characterized by comprising the following steps of:
determining the request processing capacity of each node according to a preset evaluation mode, and acquiring the request processing quantity of each node;
determining a topology center node according to the request processing capacity of each node and the service requirements of users;
determining a flow center node according to the request processing quantity of each node;
completing node identification according to the traffic center node and the topology center node;
wherein, the determining the topology center node according to the request processing capability of each node and the user service requirement includes:
calculating the modularity according to the request processing capacity of each node, and determining the community modularity of each node;
determining a plurality of request capability clusters according to community modularity of each node;
calculating the compact weight of each request capacity cluster to obtain a cluster center node of each request capacity cluster;
Determining a topology center node according to the clustering center nodes of each request capacity cluster and the user service requirement;
the determining the traffic center node according to the request processing quantity of each node comprises the following steps:
initializing labels of all nodes to obtain node labels of all nodes;
clustering nodes according to the node labels of the nodes and the request processing quantity of the nodes to obtain a plurality of traffic clusters;
calculating the compact weight of each flow cluster to obtain a cluster center node of each flow cluster;
and determining the flow center node according to the cluster center node of each flow cluster.
2. The traffic migration-oriented node identification method according to claim 1, wherein the calculating the tight weight of each request capability cluster to obtain a cluster center node of each request capability cluster includes:
acquiring the total number of nodes and the number of clustered nodes clustered by each request capability;
determining the compact weight of each node in each request capacity cluster according to the node cluster distance, the total number of the nodes and the cluster node number of each request capacity cluster;
sorting the compact weights of the nodes in the request capacity clusters to obtain a sorting result;
And determining a clustering center node of each request capacity cluster according to the sequencing result.
3. The traffic migration-oriented node identification method according to claim 1, wherein the clustering of the nodes according to the node labels of the nodes and the request processing number of the nodes to obtain a plurality of traffic clusters comprises:
counting the request processing quantity of adjacent nodes of each node according to the node label of each node;
determining the weight of a target node according to the request processing quantity of the adjacent nodes of each node;
updating the node labels of all the nodes according to the weight of the target node to obtain update labels;
and clustering the nodes according to the update labels of the nodes to obtain a plurality of traffic clusters.
4. The traffic migration-oriented node identification method according to claim 1, wherein after the node identification is completed according to the traffic center node and the topology center node, further comprising:
determining a traffic migration node according to the traffic center node and the center label of the topology center node;
and when receiving the traffic migration task, completing the traffic migration task according to the traffic migration node.
5. The traffic migration-oriented node identification method according to claim 4, wherein said completing the traffic migration task according to the traffic migration node when the traffic migration task is received comprises:
When receiving a traffic migration task, acquiring the request quantity of a topology center node;
determining a request consumption parameter according to the traffic migration task;
and completing the traffic migration task according to the request consumption parameter, the request quantity, the traffic migration node and a preset evaluation mode.
6. A traffic migration-oriented node identification device, characterized in that the traffic migration-oriented node identification device comprises:
the acquisition module is used for determining the request processing capacity of each node according to a preset evaluation mode and acquiring the request processing quantity of each node;
the determining module is used for determining a topology center node according to the request processing capacity of each node and the user service requirement;
the determining module is further used for determining a flow center node according to the request processing quantity of each node;
the completion module is used for completing node identification according to the traffic center node and the topology center node;
the determining module is further used for calculating the modularity according to the request processing capacity of each node, and determining the community modularity of each node; determining a plurality of request capability clusters according to community modularity of each node; calculating the compact weight of each request capacity cluster to obtain a cluster center node of each request capacity cluster; determining a topology center node according to the clustering center nodes of each request capacity cluster and the user service requirement;
The determining module is further used for initializing the labels of the nodes to obtain the node labels of the nodes; clustering nodes according to the node labels of the nodes and the request processing quantity of the nodes to obtain a plurality of traffic clusters; calculating the compact weight of each flow cluster to obtain a cluster center node of each flow cluster; and determining the flow center node according to the cluster center node of each flow cluster.
7. A traffic migration oriented node identification device, the device comprising: a memory, a processor, and a traffic migration oriented node identification program stored on the memory and executable on the processor, the traffic migration oriented node identification program configured to implement the traffic migration oriented node identification method of any one of claims 1 to 5.
8. A storage medium, wherein a traffic migration oriented node identification program is stored on the storage medium, and when the traffic migration oriented node identification program is executed by a processor, the traffic migration oriented node identification method according to any one of claims 1 to 5 is implemented.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310093567.7A CN115776445B (en) | 2023-02-10 | 2023-02-10 | Traffic migration-oriented node identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310093567.7A CN115776445B (en) | 2023-02-10 | 2023-02-10 | Traffic migration-oriented node identification method, device, equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115776445A CN115776445A (en) | 2023-03-10 |
CN115776445B true CN115776445B (en) | 2023-06-30 |
Family
ID=85393483
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310093567.7A Active CN115776445B (en) | 2023-02-10 | 2023-02-10 | Traffic migration-oriented node identification method, device, equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115776445B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111741082A (en) * | 2020-06-06 | 2020-10-02 | 宋倩云 | Network processing method based on 5G and edge computing and cloud communication network server |
CN114374655A (en) * | 2021-12-22 | 2022-04-19 | 中国电信股份有限公司 | Network flow characteristic extraction method, device, equipment and storage medium |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8976710B2 (en) * | 2011-12-27 | 2015-03-10 | Infosys Limited | Methods for discovering and analyzing network topologies and devices thereof |
CN109213566B (en) * | 2017-06-29 | 2022-05-13 | 华为技术有限公司 | Virtual machine migration method, device and equipment |
CN108268603A (en) * | 2017-12-22 | 2018-07-10 | 中国电子科技集团公司第三十研究所 | A kind of community discovery method based on core member's identification |
CN108683448B (en) * | 2018-04-24 | 2020-10-09 | 中国民航大学 | Influence node identification method and system suitable for aviation network |
CN110730138A (en) * | 2019-10-21 | 2020-01-24 | 中国科学院空间应用工程与技术中心 | Dynamic resource allocation method, system and storage medium for space-based cloud computing architecture |
CN113393250A (en) * | 2021-06-09 | 2021-09-14 | 北京沃东天骏信息技术有限公司 | Information processing method and device and storage medium |
CN115473686A (en) * | 2022-08-11 | 2022-12-13 | 西北工业大学 | Botnet key node identification method and identification system |
-
2023
- 2023-02-10 CN CN202310093567.7A patent/CN115776445B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111741082A (en) * | 2020-06-06 | 2020-10-02 | 宋倩云 | Network processing method based on 5G and edge computing and cloud communication network server |
CN114374655A (en) * | 2021-12-22 | 2022-04-19 | 中国电信股份有限公司 | Network flow characteristic extraction method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN115776445A (en) | 2023-03-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Sun et al. | Multi-objective optimization of resource scheduling in fog computing using an improved NSGA-II | |
CN104995870B (en) | Multiple target server arrangement determines method and apparatus | |
Zheng et al. | Deep reinforcement learning-based workload scheduling for edge computing | |
CN104038540B (en) | Method and system for automatically selecting application proxy server | |
Patel et al. | Survey of load balancing techniques for grid | |
CN114126066B (en) | MEC-oriented server resource allocation and address selection joint optimization decision method | |
Li et al. | Method of resource estimation based on QoS in edge computing | |
Jazayeri et al. | A latency-aware and energy-efficient computation offloading in mobile fog computing: a hidden Markov model-based approach | |
Liu et al. | An adaptive approach to better load balancing in a consumer-centric cloud environment | |
Zhao et al. | Optimize the placement of edge server between workload balancing and system delay in smart city | |
Li et al. | Computation offloading and service allocation in mobile edge computing | |
Li et al. | DQN-enabled content caching and quantum ant colony-based computation offloading in MEC | |
Soula et al. | Intelligent tasks allocation at the edge based on machine learning and bio-inspired algorithms | |
Qayyum et al. | Mobility-aware hierarchical fog computing framework for Industrial Internet of Things (IIoT) | |
Sarkar et al. | Deep learning-based energy-efficient computational offloading strategy in heterogeneous fog computing networks | |
Asghari et al. | Server placement in mobile cloud computing: a comprehensive survey for edge computing, fog computing and cloudlet | |
Huang et al. | Computation offloading for multimedia workflows with deadline constraints in cloudlet-based mobile cloud | |
Subrahmanyam et al. | Optimizing horizontal scalability in cloud computing using simulated annealing for Internet of Things | |
Faraji-Mehmandar et al. | A self-learning approach for proactive resource and service provisioning in fog environment | |
CN115776445B (en) | Traffic migration-oriented node identification method, device, equipment and storage medium | |
Zhai et al. | Optimal task partition with delay requirement in mobile crowdsourcing | |
Cao et al. | Performance and stability of application placement in mobile edge computing system | |
Rodríguez et al. | Improvement of edge computing workload placement using multi objective particle swarm optimization | |
Khansoltani et al. | A request redirection algorithm in content delivery network: Using promethee approach | |
Park et al. | ActiveDBC: learning Knowledge-based Information propagation in mobile social networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |