CN117499325B - Switch service message distribution method and system based on artificial intelligence - Google Patents
Switch service message distribution method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN117499325B CN117499325B CN202311849864.6A CN202311849864A CN117499325B CN 117499325 B CN117499325 B CN 117499325B CN 202311849864 A CN202311849864 A CN 202311849864A CN 117499325 B CN117499325 B CN 117499325B
- Authority
- CN
- China
- Prior art keywords
- message
- cluster
- classification model
- standard deviation
- parameters
- 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 15
- 238000013145 classification model Methods 0.000 claims abstract description 51
- 238000012544 monitoring process Methods 0.000 claims abstract description 19
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 239000013598 vector Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 10
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/12—Network monitoring probes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/22—Parsing or analysis of headers
-
- 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
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a method and a system for distributing service messages of an exchanger based on artificial intelligence, wherein the method comprises the following steps: data acquisition, message feature extraction, message classification model establishment, load balancing and real-time monitoring. The invention belongs to the technical field of data processing, in particular to a service message distribution method and a service message distribution system based on artificial intelligence, wherein the scheme adopts parameters defining a message classification model, obtains estimated parameters by minimizing an objective function, calculates experience risks, introduces a hierarchical structure, obtains a message classification model based on a multi-level structure, and realizes classification processing of service message data; and calculating the standard deviation of the average cluster energy and the standard deviation of the average cluster size to obtain the proximity degree of the nodes in the cluster, dynamically adjusting the load distribution in the cluster, and improving the shunt capacity of the system.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to an artificial intelligence-based switch service message distribution method and system.
Background
The switch service message distribution method is a method for intelligently distributing and managing service messages received by a network switch according to specific rules, and aims to realize intelligent management and scheduling of network traffic, improve network safety and usability and improve user experience. However, the existing service message distribution method of the switch has the technical problems that service message data contains multi-level information, classification processing is difficult to accurately perform, and distribution is difficult; the method has the technical problems that the processing pressure is high and effective management is difficult because the service message data of large-scale flow management is high-speed gushed into the system.
Disclosure of Invention
Aiming at the technical problems that the existing service message data contains multi-level information, is difficult to accurately classify and process, and causes difficult splitting, parameters defining a message classification model are adopted, estimated parameters are obtained by minimizing an objective function, experience risks are calculated, a hierarchical structure is introduced, and a message classification model based on the multi-level structure is obtained; aiming at the technical problems that service message data with large-scale flow management is high in speed and is large in processing pressure and difficult to effectively manage due to the fact that the service message data is gushed into a system, the standard deviation of average cluster energy and the standard deviation of average cluster size are calculated to obtain the proximity degree of nodes in a cluster, load distribution in the cluster is dynamically adjusted, and the shunting capacity of the system is improved.
The technical scheme adopted by the invention is as follows: the invention provides a service message distribution method of an exchanger based on artificial intelligence, which comprises the following steps:
step S1: collecting data;
step S2: extracting message characteristics;
step S3: establishing a message classification model, specifically defining parameters of the message classification model, obtaining estimated parameters by minimizing an objective function, calculating experience risks, and introducing a hierarchical structure to obtain the message classification model based on the multi-level structure;
step S4: load balancing, namely calculating the standard deviation of the energy of an average cluster and the standard deviation of the size of the average cluster to obtain the proximity degree of nodes in the cluster, dynamically adjusting the load distribution in the cluster, and improving the shunt capacity of the system;
step S5: and (5) monitoring in real time.
Further, in step S1, the data acquisition, specifically, the connection to a network traffic analyzer, captures the traffic of the message passing through the switch, and extracts the service message data.
Further, in step S2, the extracting the message features includes the following steps:
step S21: basic feature extraction, including message length, message time stamp, message source and message protocol;
step S22: statistical feature extraction, including total byte count, average packet length, number of packets, and packet inter-arrival time;
step S23: extracting protocol features, including TCP features, HTTP features and DNS features;
step S24: and extracting flow characteristics, including flow direction characteristics and flow size characteristics.
Further, in step S3, the establishing a packet classification model includes the following steps:
step S31: inputting message characteristics, carrying out data preprocessing on the extracted message characteristics, and inputting a message classification model;
step S32: parameters of the message classification model are defined, and the following formula is used:
；
wherein W is a set of parameter vectors, e is a hierarchical structure, N is a set of hierarchical structures, W e Is the parameter vector of the e-th hierarchical structure;
step S33: estimating parameters, obtaining the estimated parameters by minimizing an objective function, wherein the formula is as follows:
；
in the method, in the process of the invention,is an estimated value of the parameter vector, argmin representsDetermining a parameter value that minimizes an objective function, the objective function being [ lambda (w) +alpha x R emp ]Lambda (w) is a regularization term that helps control model complexity, prevents overfitting, R emp Representing the empirical risk on the training dataset, being the sum of the model's losses on the training dataset, α being the adjustment parameter for balancing the weights between the regularization term and the empirical risk;
step S34: the empirical risk of the packet classification model is calculated as the loss generated by the instance at the leaf node in the hierarchy, using the formula:
；
wherein L is a loss function, n is a leaf node, the leaf node is an end node in the hierarchical structure and represents the final classification result of the message, T is the sum of all leaf nodes in the hierarchical structure, M is all examples in each leaf node, i is an index of an example and represents the position of a specific message in service message data, y i Is the true value of instance i, E i Is the feature set of instance i, w n Is a parameter of leaf node n, used to determine a threshold for a particular classification;
step S35: introducing a hierarchical structure, and combining a recursive structure into a regularization term, wherein the formula is as follows:
；
wherein w is τ (n) is the parent node parameter of leaf node n,the square of the Euclidean distance between two parameter vectors is represented and used for measuring the difference of parameters of a message classification model.
Further, in step S4, the load balancing includes the following steps:
step S41: the standard deviation of the average cluster energy is calculated using the following formula:
；
where k is the number of clusters, i is the index of the clusters, σ CE Is the standard deviation of the average cluster energy, sigma CE The lower the value of (c), the higher the value of fitness and ζ i Is the average cluster energy of the i-th cluster;
step S42: the standard deviation of the average cluster size was calculated using the following formula:
；
in sigma CS Is the standard deviation of the average cluster size, sigma CS The lower the value of (2), the higher the value of fitness, θ being the expected value of the average cluster size, θ i Is the cluster size of the i-th cluster;
step S43: the proximity of the nodes in the cluster is calculated using the following formula:
；
where μ is the proximity of the node in the cluster, K is the proportionality constant, d m (a, b) is the distance between node a and node b in the mth cluster;
step S44: load distribution is dynamically adjusted by comprehensively considering average cluster energy, average cluster size and the proximity degree of nodes in the clusters, so that more balanced load distribution is realized, and the shunting capacity is improved.
Further, in step S5, the real-time monitoring is specifically monitoring the real-time performance and accuracy of the shunting result, and is timely fed back to the system administrator, so as to continuously optimize the shunting strategy and the parameters of the message classification model.
The invention provides an artificial intelligence based switch service message distribution system, which comprises a data acquisition module, a message characteristic extraction module, a message classification model building module, a load balancing module and a real-time monitoring module;
the data acquisition module is connected with the network flow analyzer, captures the message flow passing through the switch and extracts the service message data;
the message feature extraction module is used for extracting basic features, statistical features, protocol features and flow features;
the message classification model building module is used for defining parameters of a message classification model, obtaining estimated parameters by minimizing an objective function, calculating experience risks, and introducing a hierarchical structure to obtain the message classification model based on the multi-level structure;
the load balancing module is used for calculating the standard deviation of the energy of the average cluster and the standard deviation of the size of the average cluster to obtain the proximity degree of the nodes in the cluster, dynamically adjusting the load distribution in the cluster and improving the shunt capacity of the system;
the real-time monitoring module is used for monitoring the real-time performance and accuracy of the shunting result and feeding the shunting result back to a system administrator in time, so that the shunting strategy and the parameters of the message classification model are optimized continuously.
The beneficial results obtained by adopting the scheme of the invention are as follows:
(1) Aiming at the technical problems that the existing service message data contains multi-level information, classification processing is difficult to accurately perform, and flow distribution is difficult to perform, parameters defining a message classification model are adopted, estimated parameters are obtained by minimizing an objective function, experience risks are calculated, a hierarchical structure is introduced, and a message classification model based on the multi-level structure is obtained;
(2) Aiming at the technical problems that service message data with large-scale flow management is high in speed and is large in processing pressure and difficult to effectively manage due to the fact that the service message data is gushed into a system, the standard deviation of average cluster energy and the standard deviation of average cluster size are calculated to obtain the proximity degree of nodes in a cluster, load distribution in the cluster is dynamically adjusted, and the shunting capacity of the system is improved.
Drawings
Fig. 1 is a schematic flow chart of a service message splitting method of an exchange based on artificial intelligence;
FIG. 2 is a schematic diagram of an artificial intelligence based switch service message distribution system according to the present invention;
FIG. 3 is a flow chart of step S3;
fig. 4 is a flow chart of step S4.
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention; all other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," "outer," and the like indicate orientation or positional relationships based on those shown in the drawings, merely to facilitate description of the invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the invention.
Referring to fig. 1, the present invention provides a service message splitting method for an artificial intelligence based switch, which includes the following steps:
step S1: collecting data;
step S2: extracting message characteristics;
step S3: establishing a message classification model, specifically defining parameters of the message classification model, obtaining estimated parameters by minimizing an objective function, calculating experience risks, and introducing a hierarchical structure to obtain the message classification model based on the multi-level structure;
step S4: load balancing, namely calculating the standard deviation of the energy of an average cluster and the standard deviation of the size of the average cluster to obtain the proximity degree of nodes in the cluster, dynamically adjusting the load distribution in the cluster, and improving the shunt capacity of the system;
step S5: and (5) monitoring in real time.
In step S1, the data acquisition, specifically, the connection network traffic analyzer, captures the traffic of the message passing through the switch, and extracts the service message data, referring to fig. 1.
In a third embodiment, referring to fig. 1, the method is based on the above embodiment, and in step S2, the extracting of the message features includes the following steps:
step S21: basic feature extraction, including message length, message time stamp, message source and message protocol;
step S22: statistical feature extraction, including total byte count, average packet length, number of packets, and packet inter-arrival time;
step S23: extracting protocol features, including TCP features, HTTP features and DNS features;
step S24: and extracting flow characteristics, including flow direction characteristics and flow size characteristics.
In a fourth embodiment, referring to fig. 1 and 3, the method is based on the above embodiment, and in step S3, the creating a packet classification model includes the following steps:
step S31: inputting message characteristics, carrying out data preprocessing on the extracted message characteristics, and inputting a message classification model;
step S32: parameters of the message classification model are defined, and the following formula is used:
；
wherein W is a set of parameter vectors, e is a hierarchical structure, N is a set of hierarchical structures, W e Is the parameter vector of the e-th hierarchical structure;
step S33: estimating parameters, obtaining the estimated parameters by minimizing an objective function, wherein the formula is as follows:
；
in the method, in the process of the invention,is an estimate of the parameter vector, argmin represents the parameter value that is calculated to minimize the objective function, which is [ lambda (w) +alpha x R emp ]Lambda (w) is a regularization term that helps control model complexity, prevents overfitting, R emp Representing the empirical risk on the training dataset, being the sum of the model's losses on the training dataset, α being the adjustment parameter for balancing the weights between the regularization term and the empirical risk;
step S34: the empirical risk of the packet classification model is calculated as the loss generated by the instance at the leaf node in the hierarchy, using the formula:
；
wherein L is a loss function, n is a leaf node, the leaf node is an end node in the hierarchical structure and represents the final classification result of the message, T is the sum of all leaf nodes in the hierarchical structure, M is all examples in each leaf node, i is an index of an example and represents the position of a specific message in service message data, y i Is the true value of instance i, E i Is the feature set of instance i, w n Is a parameter of leaf node n, used to determine a threshold for a particular classification;
step S35: introducing a hierarchical structure, and combining a recursive structure into a regularization term, wherein the formula is as follows:
；
wherein w is τ (n) parent node parameter for leaf node nThe number of the product is the number,the square of the Euclidean distance between two parameter vectors is represented and used for measuring the difference of parameters of a message classification model.
By executing the steps, the method adopts the parameters for defining the message classification model, obtains estimated parameters by minimizing the objective function, calculates experience risks, introduces a hierarchical structure, and obtains the message classification model based on the multi-level structure, thereby solving the technical problems that service message data contains multi-level information, classification processing is difficult to accurately perform, and shunting is difficult.
Embodiment five, referring to fig. 1 and 4, the load balancing in step S4, based on the above embodiment, includes the following steps:
step S41: the standard deviation of the average cluster energy is calculated using the following formula:
；
where k is the number of clusters, i is the index of the clusters, σ CE Is the standard deviation of the average cluster energy, sigma CE The lower the value of (c), the higher the value of fitness and ζ i Is the average cluster energy of the i-th cluster;
step S42: the standard deviation of the average cluster size was calculated using the following formula:
；
in sigma CS Is the standard deviation of the average cluster size, sigma CS The lower the value of (2), the higher the value of fitness, θ being the expected value of the average cluster size, θ i Is the cluster size of the i-th cluster;
step S43: the proximity of the nodes in the cluster is calculated using the following formula:
；
where μ is the proximity of the node in the cluster, K is the proportionality constant, d m (a, b) is the distance between node a and node b in the mth cluster;
step S44: load distribution is dynamically adjusted by comprehensively considering average cluster energy, average cluster size and the proximity degree of nodes in the clusters, so that more balanced load distribution is realized, and the shunting capacity is improved.
By executing the steps, the method adopts the standard deviation of the average cluster energy and the standard deviation of the average cluster size to obtain the proximity degree of the nodes in the clusters, dynamically adjusts the load distribution in the clusters, improves the system distribution capacity, and solves the technical problems that the service message data of large-scale flow management is gushed into the system at a high speed, the processing pressure is high, and the effective management is difficult to carry out.
In a sixth embodiment, referring to fig. 1, the embodiment is based on the foregoing embodiment, and in step S5, the real-time monitoring is specifically monitoring the real-time performance and accuracy of the splitting result, and is timely fed back to the system administrator, so as to continuously optimize the splitting strategy and the parameters of the message classification model.
An embodiment seven, referring to fig. 2, based on the above embodiment, the invention provides an artificial intelligence based switch service message distribution system, which includes a data acquisition module, a message feature extraction module, a message classification model establishment module, a load balancing module and a real-time monitoring module;
the data acquisition module is connected with the network flow analyzer, captures the message flow passing through the switch and extracts the service message data;
the message feature extraction module is used for extracting basic features, statistical features, protocol features and flow features;
the message classification model building module is used for defining parameters of a message classification model, obtaining estimated parameters by minimizing an objective function, calculating experience risks, and introducing a hierarchical structure to obtain the message classification model based on the multi-level structure;
the load balancing module is used for calculating the standard deviation of the energy of the average cluster and the standard deviation of the size of the average cluster to obtain the proximity degree of the nodes in the cluster, dynamically adjusting the load distribution in the cluster and improving the shunt capacity of the system;
the real-time monitoring module is used for monitoring the real-time performance and accuracy of the shunting result and feeding the shunting result back to a system administrator in time, so that the shunting strategy and the parameters of the message classification model are optimized continuously.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown in the drawings. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.
Claims (3)
1. A service message distribution method of an exchanger based on artificial intelligence is characterized in that: the method comprises the following steps:
step S1: collecting data;
step S2: extracting message characteristics;
step S3: establishing a message classification model, specifically defining parameters of the message classification model, obtaining estimated parameters by minimizing an objective function, calculating experience risks, and introducing a hierarchical structure to obtain the message classification model based on the multi-level structure;
step S4: load balancing, namely calculating the standard deviation of the energy of an average cluster and the standard deviation of the size of the average cluster to obtain the proximity degree of nodes in the cluster, and dynamically adjusting the load distribution in the cluster;
step S5: monitoring in real time;
in step S3, the establishing a packet classification model includes the following steps:
step S31: inputting message characteristics, carrying out data preprocessing on the extracted message characteristics, and inputting a message classification model;
step S32: parameters of the message classification model are defined, and the following formula is used:
；
wherein W is a set of parameter vectors, e is a hierarchical structure, N is a set of hierarchical structures, W e Is the parameter vector of the e-th hierarchical structure;
step S33: estimating parameters, obtaining the estimated parameters by minimizing an objective function, wherein the formula is as follows:
；
in the method, in the process of the invention,is an estimate of the parameter vector, argmin represents the parameter value that is calculated to minimize the objective function, which is [ lambda (w) +alpha x R emp ]Lambda (w) is a regularization term that helps control model complexityDegree, prevent overfitting, R emp Representing the empirical risk on the training dataset, being the sum of the model's losses on the training dataset, α being the adjustment parameter for balancing the weights between the regularization term and the empirical risk;
step S34: the empirical risk of the packet classification model is calculated as the loss generated by the instance at the leaf node in the hierarchy, using the formula:
；
wherein L is a loss function, n is a leaf node, the leaf node is an end node in the hierarchical structure and represents the final classification result of the message, T is the sum of all leaf nodes in the hierarchical structure, M is all examples in each leaf node, i is an index of an example and represents the position of a specific message in service message data, y i Is the true value of instance i, E i Is the feature set of instance i, w n Is a parameter of leaf node n, used to determine a threshold for a particular classification;
step S35: introducing a hierarchical structure, and combining a recursive structure into a regularization term, wherein the formula is as follows:
；
in the method, in the process of the invention,is the parent node parameter of leaf node n, +.>The square of the Euclidean distance between two parameter vectors is represented and used for measuring the difference of parameters of a message classification model;
in step S4, the load balancing includes the following steps:
step S41: the standard deviation of the average cluster energy is calculated using the following formula:
；
where k is the number of clusters, i is the index of the clusters, σ CE Is the standard deviation of the average cluster energy, sigma CE The lower the value of (c), the higher the value of fitness and ζ i Is the average cluster energy of the i-th cluster;
step S42: the standard deviation of the average cluster size was calculated using the following formula:
；
in sigma CS Is the standard deviation of the average cluster size, θ is the expected value of the average cluster size, θ i Is the cluster size of the i-th cluster;
step S43: the proximity of the nodes in the cluster is calculated using the following formula:
；
where μ is the proximity of the node in the cluster, K is the proportionality constant, d m (a, b) is the distance between node a and node b in the mth cluster;
step S44: load distribution, namely dynamically adjusting the load distribution in the cluster by comprehensively considering the average cluster energy, the average cluster size and the proximity degree of the nodes in the cluster;
in step S1, the data acquisition is specifically connected to a network traffic analyzer, capturing the traffic of a message passing through a switch, and extracting service message data;
in step S2, the extracting the message features includes the following steps:
step S21: basic feature extraction, including message length, message time stamp, message source and message protocol;
step S22: statistical feature extraction, including total byte count, average packet length, number of packets, and packet inter-arrival time;
step S23: extracting protocol features, including TCP features, HTTP features and DNS features;
step S24: extracting flow characteristics, including flow direction characteristics and flow size characteristics;
in step S5, the real-time monitoring is specifically monitoring the real-time performance and accuracy of the shunting result, and is timely fed back to the system administrator, so as to continuously optimize the shunting strategy and the parameters of the message classification model.
2. An artificial intelligence based switch service message distribution system for implementing an artificial intelligence based switch service message distribution method as described in claim 1, wherein: the system comprises a data acquisition module, a message characteristic extraction module, a message classification model establishment module, a load balancing module and a real-time monitoring module.
3. The system for distributing service messages of an artificial intelligence based switch according to claim 2, wherein: the data acquisition module is connected with the network flow analyzer, captures the message flow passing through the switch and extracts the service message data;
the message feature extraction module is used for extracting basic features, statistical features, protocol features and flow features;
the message classification model building module is used for defining parameters of a message classification model, obtaining estimated parameters by minimizing an objective function, calculating experience risks, and introducing a hierarchical structure to obtain the message classification model based on the multi-level structure;
the load balancing module is used for calculating the standard deviation of the average cluster energy and the standard deviation of the average cluster size to obtain the proximity degree of the nodes in the cluster, dynamically adjusting the load distribution in the cluster and improving the model shunting capability;
the real-time monitoring module is used for monitoring the real-time performance and accuracy of the shunting result and feeding the shunting result back to a system administrator in time, so that the shunting strategy and the parameters of the message classification model are optimized continuously.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311849864.6A CN117499325B (en) | 2023-12-29 | 2023-12-29 | Switch service message distribution method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311849864.6A CN117499325B (en) | 2023-12-29 | 2023-12-29 | Switch service message distribution method and system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117499325A CN117499325A (en) | 2024-02-02 |
| CN117499325B true CN117499325B (en) | 2024-03-15 |
Family
ID=89685357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311849864.6A Active CN117499325B (en) | 2023-12-29 | 2023-12-29 | Switch service message distribution method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117499325B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111211994A (en) * | 2019-11-28 | 2020-05-29 | 南京邮电大学 | Network traffic classification method based on SOM and K-means fusion algorithm |
| WO2021068489A1 (en) * | 2019-10-12 | 2021-04-15 | 深圳壹账通智能科技有限公司 | Routing path intelligent selection method and apparatus, device, and readable storage medium |
| WO2021114844A1 (en) * | 2019-12-10 | 2021-06-17 | 华为技术有限公司 | Traffic classification method and traffic management device |
| CN113191434A (en) * | 2021-05-07 | 2021-07-30 | 支付宝(杭州)信息技术有限公司 | Method and device for training risk recognition model |
| CN113590819A (en) * | 2021-06-30 | 2021-11-02 | 中山大学 | Large-scale category-level text classification method |
| CN114582107A (en) * | 2022-03-01 | 2022-06-03 | 中用科技有限公司 | Intelligent early warning system for limited space operation safety |
| CN115277587A (en) * | 2022-07-29 | 2022-11-01 | 中国电信股份有限公司 | Network traffic identification method, device, equipment and medium |
| WO2023179014A1 (en) * | 2022-03-23 | 2023-09-28 | 中兴通讯股份有限公司 | Traffic identification method and apparatus, electronic device, and storage medium |
-
2023
- 2023-12-29 CN CN202311849864.6A patent/CN117499325B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021068489A1 (en) * | 2019-10-12 | 2021-04-15 | 深圳壹账通智能科技有限公司 | Routing path intelligent selection method and apparatus, device, and readable storage medium |
| CN111211994A (en) * | 2019-11-28 | 2020-05-29 | 南京邮电大学 | Network traffic classification method based on SOM and K-means fusion algorithm |
| WO2021114844A1 (en) * | 2019-12-10 | 2021-06-17 | 华为技术有限公司 | Traffic classification method and traffic management device |
| CN113191434A (en) * | 2021-05-07 | 2021-07-30 | 支付宝(杭州)信息技术有限公司 | Method and device for training risk recognition model |
| CN113590819A (en) * | 2021-06-30 | 2021-11-02 | 中山大学 | Large-scale category-level text classification method |
| CN114582107A (en) * | 2022-03-01 | 2022-06-03 | 中用科技有限公司 | Intelligent early warning system for limited space operation safety |
| WO2023179014A1 (en) * | 2022-03-23 | 2023-09-28 | 中兴通讯股份有限公司 | Traffic identification method and apparatus, electronic device, and storage medium |
| CN115277587A (en) * | 2022-07-29 | 2022-11-01 | 中国电信股份有限公司 | Network traffic identification method, device, equipment and medium |
Non-Patent Citations (1)
| Title |
|---|
| RHS-CNN:一种基于正则化层次Softmax的CNN文本分类模型;王勇;何养明;陈荟西;黎春;;重庆理工大学学报(自然科学);20200515(05);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117499325A (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107634911B (en) | Adaptive congestion control method based on deep learning in information center network | |
| WO2022057268A1 (en) | Base station energy-saving policy determining method and apparatus, and storage medium | |
| CN104580222B (en) | Ddos attack Distributed Detection and response method based on comentropy | |
| CN104579823A (en) | Large-data-flow-based network traffic abnormality detection system and method | |
| CN109768981B (en) | Network attack defense method and system based on machine learning under SDN architecture | |
| CN116545954B (en) | Communication gateway data transmission method and system based on internet of things | |
| CN104618377A (en) | NetFlow based botnet network detection system and detection method | |
| CN110324260B (en) | Network function virtualization intelligent scheduling method based on flow identification | |
| CN106131918A (en) | The associating Path selection of energy acquisition node and power distribution method in wireless sense network | |
| CN110149207A (en) | New type of continuous variable quantum key delivering method based on machine learning | |
| CN113518007A (en) | Multi-internet-of-things equipment heterogeneous model efficient mutual learning method based on federal learning | |
| CN103596191A (en) | Intelligent configuration system and intelligent configuration method for wireless sensor network | |
| CN109831808A (en) | A kind of resource allocation methods of the hybrid power supply C-RAN based on machine learning | |
| CN103747003A (en) | Peer-to-peer botnet core node detection method and detection device | |
| CN107659505B (en) | SDN network routing method and SDN controller | |
| CN115314399A (en) | Data center flow scheduling method based on inverse reinforcement learning | |
| CN117499325B (en) | Switch service message distribution method and system based on artificial intelligence | |
| CN110601916A (en) | Flow sampling and application sensing system based on machine learning | |
| CN106789429B (en) | A kind of adaptive low-cost SDN network link utilization measurement method and system | |
| Ma et al. | Asynchronous federated learning for elephant flow detection in software defined networking systems | |
| CN114979017B (en) | Deep learning protocol identification method and system based on original flow of industrial control system | |
| CN108234202B (en) | Method for discovering network topology based on life span | |
| WO2022121979A1 (en) | Inner loop value adjustment method and device, storage medium, and electronic device | |
| CN105792258B (en) | The cross-layer optimizing method of wireless sense network medium-rate and reliability collaboration | |
| Liaojun et al. | Fair data collection scheme in wireless sensor 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 |