CN114374617A - Fault-tolerant prefabricating method for deterministic network - Google Patents
Fault-tolerant prefabricating method for deterministic network Download PDFInfo
- Publication number
- CN114374617A CN114374617A CN202111514275.3A CN202111514275A CN114374617A CN 114374617 A CN114374617 A CN 114374617A CN 202111514275 A CN202111514275 A CN 202111514275A CN 114374617 A CN114374617 A CN 114374617A
- Authority
- CN
- China
- Prior art keywords
- data
- network
- detnet
- fault
- tolerant
- 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.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/34—Source routing
-
- 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
- H04L47/2483—Traffic characterised by specific attributes, e.g. priority or QoS involving identification of individual flows
Abstract
The invention discloses a fault-tolerant prefabricating method for a deterministic network, which comprises the following steps of 1: the fault-tolerant prefabrication module carries out DetNet flow identification detection on a control plane of a deterministic network and carries out historical data statistics; step 2: according to the statistical historical data, a prediction model based on an artificial intelligence algorithm is combined to predict whether future data need to be subjected to the DetNet flow copying operation, if so, the DetNet flow copying operation is configured and issued, and a path and measures for data packet loss are guided; and step 3: and the path planning module guides the global network flow path by utilizing centralized path setting by means of a control plane of the deterministic network. The invention realizes reliability, ensures time delay, saves network bandwidth resources, provides more link resources for non-DetNet flow, and provides a more reliable and rapid network guarantee.
Description
Technical Field
The invention belongs to the technical field of network reliable transmission, and particularly relates to a fault-tolerant prefabricating method for a deterministic network.
Background
The development and application of informatization and network have penetrated all walks of life and brings convenience to people's life. At present, the emerging industries such as Internet of vehicles, unmanned driving, intelligent medical treatment, intelligent factories and the like develop rapidly, and higher requirements are put forward on the time delay and the reliability of the network, for example, the time delay requirement of remote control is within 5ms, the reliability needs to reach 99.999 percent, while the discrete automatic motion control needs to realize the reaction between 1us and 1ms, and the reliability needs to reach 99.9999 percent.
The deterministic network technology can provide guarantee for the development of emerging industries. Currently, deterministic networks achieve the goal of not losing data by packet duplication and elimination to achieve reliability. In the current reliability measures of deterministic networks, how to perform packet replication and effective fault tolerance are not involved. If the packet replication is performed all the time, link resources are greatly wasted, and link congestion is easily caused.
Disclosure of Invention
The invention aims to solve the technical problem of providing a fault-tolerant prefabricating method for a deterministic network, which aims at overcoming the defects of the prior art, and provides a more reliable and faster network guarantee by opening an IP network and a non-IP network through the deterministic network and providing an effective DetNet flow fault-tolerant prefabricating mechanism according to historical statistical data and an artificial intelligence prediction technology to ensure the integrity of network information, directly and selectively send redundancy and reduce the secondary retransmission time delay of the traditional Ethernet, thereby ensuring the time delay of network data transmission, realizing the reliability, ensuring the time delay, saving network bandwidth resources, providing more link resources for the non-DetNet flow.
In order to achieve the technical purpose, the technical scheme adopted by the invention is as follows:
a fault tolerant prefabrication method usable with deterministic networks, comprising:
step 1: the fault-tolerant prefabrication module carries out DetNet flow identification detection on a control plane of a deterministic network and carries out historical data statistics;
step 2: according to the statistical historical data, a prediction model based on an artificial intelligence algorithm is combined to predict whether future data need to be subjected to the DetNet flow copying operation, if so, the DetNet flow copying operation is configured and issued, and a path and measures for data packet loss are guided;
and step 3: and the path planning module guides the global network flow path by utilizing centralized path setting by means of a control plane of the deterministic network.
In order to optimize the technical scheme, the specific measures adopted further comprise:
in the control plane of the deterministic network in the above step 1, the DetNet traffic information of the data layer is periodically collected in seconds, and the information is stored in the database server;
the DetNet traffic information includes, but is not limited to, information of time stamp, packet coding, stream copy label, stream erasure label, each node through which the DetNet stream passes, whether the packet is used label, stream source address, stream destination address.
And 2, providing a user interface by the fault-tolerant prefabricating module in the step 2 for defining the data volume of the training and verifying set, inputting and outputting the predicted data and selecting a prediction model.
The prediction model in the step 2 comprises a prediction model based on a decision tree, a random forest and a support vector machine.
The step 2 is to input data in the database server as training verification data through a prediction model based on an artificial intelligence algorithm in the control plane, and output whether a link needs to perform stream replication operation and the number of stream replication in the next time period, specifically:
with DetNet flow data as a training verification set, sequentially carrying out preprocessing, flow replication behavior segmentation, data set feature extraction and feature set construction on the training verification set, then carrying out behavior identification by using data at the front end of a time sequence in the set, training a prediction model, carrying out identification verification by using subsequent residual data on the time sequence, and then predicting the DetNet flow information of a link in the next time period, wherein the specific prediction result comprises the following steps: and if the stream replication operation is needed, the number of output stream replication is determined.
The step 3 is implemented by pce (path Computation element), and directly guides the display route of the data plane of the deterministic network, specifically:
the method comprises the steps of adopting a software defined network architecture, using a controller as a PCE (personal computer equipment), calculating a constraint path of a global network by the PCE in a centralized mode, performing whole network routing guidance by means of an SR/SRv6 technology, and issuing the constraint path to a data plane, wherein the data plane issues the path to PCC network equipment through PCEP (policy and charging rules).
The fault-tolerant prefabrication module and the path planning module are used as independent program component modules and are in butt joint with an existing SDN controller and a network management platform, and management functions are improved.
The invention has the following beneficial effects:
(1) compared with a method for ensuring reliability through a retransmission technology, the method can pre-arrange the time and the times of the packet fault-tolerant replication of the DetNet flow according to historical statistical data, avoid unnecessary network data retransmission, save network resources, meet the requirements of a receiving node on reliability, stability and time delay, and is suitable for industrial control scenes.
(2) The result of the fault-tolerant prefabrication scheme of the invention directly guides the path planning scheme, and is combined with SRv6 technology for use, so that centralized network path planning is completed from a control plane, the network management flexibility and the operation and maintenance efficiency are improved, the dependence on operation and maintenance personnel is reduced, and the network management automation is realized.
Drawings
FIG. 1 is a schematic diagram of the implementation of the method of the present invention.
Detailed Description
Embodiments of the present invention are described in further detail below with reference to the accompanying drawings.
Referring to fig. 1, the invention relates to a fault-tolerant prefabrication method for deterministic networks, comprising:
step 1: the fault-tolerant prefabrication module carries out DetNet flow identification detection on a control plane of a deterministic network and carries out historical data statistics;
step 2: according to the statistical historical data, a prediction model based on an artificial intelligence algorithm is combined to predict whether future data need to be subjected to the DetNet flow copying operation, if so, the DetNet flow copying operation is configured and issued, and a path and measures for data packet loss are guided, such as secondary backup or tertiary backup;
step 2, outputting whether the DetNet flow copying operation needs to be issued, if yes, determining the number of issued flows; otherwise no instruction is given to step 3.
And step 3: the path planning module guides the global network flow path by using centralized path setting, namely calculates the path and executes a issued command to the network equipment by means of a control plane of the deterministic network.
In the embodiment, a DetNet flow project of the deterministic network adopts a framework of a software defined network, and a DetNet flow is globally planned on a control plane.
Step 1, periodically collecting DetNet flow information of a data layer on a control plane of a deterministic network by taking seconds as a unit, and storing the information in a database server;
the DetNet traffic information includes, but is not limited to, information of time stamp, packet coding, stream copy label, stream erasure label, each node through which the DetNet stream passes, whether the packet is used label, stream source address, stream destination address.
For example, in the control plane of a deterministic network, the DetNet traffic information for the data layer is collected every 1 second, stored in a database server in close physical proximity, and a data storage entry may be provided.
In an embodiment, the fault-tolerant prefabrication module in step 2 provides a user interface for defining the data volume of the training verification set, the input and output of the prediction data and selecting the prediction model.
In an embodiment, the prediction model in step 2 includes a prediction model based on a decision tree, a random forest, and a support vector machine.
In an embodiment, in step 2, data in the database server is input as training verification data through a prediction model based on an artificial intelligence algorithm in the control plane, and whether a link needs to perform a stream replication operation and the number of stream replications in a next time period are output, specifically:
the method comprises the steps of taking DetNet flow data as a training verification set, sequentially carrying out preprocessing, flow copying behavior segmentation, data set feature extraction and feature set construction on the training verification set, then carrying out behavior recognition by using data at the front end of a time sequence in the set, training a prediction model, carrying out recognition verification by using subsequent residual data on the time sequence, and then predicting DetNet flow information of a link in the next time period.
The prediction result comprises the following steps: and if the stream replication operation is needed, the number of output stream replication is determined.
The control plane includes a prediction module that uses artificial intelligence, illustrated by way of example as a random forest.
Taking data in a database server as input of a prediction model based on a random forest, taking DetNet flow data of a working day as a training verification set, sequentially preprocessing the training verification set, performing flow replication behavior segmentation, extracting features of a data set, constructing a feature set, then performing behavior recognition by using a data set of the first four days of the working day, training the prediction model, performing recognition verification by using data of the fifth day of the working day, and then predicting DetNet flow information of the working day of the next week to obtain the number of times and time of flow replication, for example, 10 in Tuesday: 50-11: 10 a triple backup stream copy operation is performed.
The above preprocessing operations include, but are not limited to: deleting redundant data, and carrying out numerical normalization processing on the data.
Stream replication split behavior: the data may be segmented based on time windows, with fixed time windows and sliding time windows. Wherein the fixed window is divided into a fixed time length window and a fixed event length window.
Feature extraction: the original data is subjected to scaling operations, such as calculating time differences, number of stream copies, and the like.
Constructing a feature set: the method is a set formed by characteristics obtained by extracting data in a certain time period and converting the data.
In an embodiment, step 3 is implemented by a PCE to directly instruct a display route of a data layer of a deterministic network, specifically:
the method comprises the steps of adopting a software defined network architecture, using a controller as a PCE (personal computer equipment), calculating a constraint path of a global network by the PCE in a centralized mode, performing whole network routing guidance by means of an SR/SRv6 technology, and issuing the constraint path to a data plane, wherein the data plane issues the path to PCC network equipment through PCEP (policy and charging rules).
In the embodiment, the fault-tolerant prefabrication module and the path planning module are used as independent program component modules and are in butt joint with an existing SDN controller and a network management platform, so that the management function is improved.
The invention is suitable for the requirement of industrial control on the network, and the IP network and the non-IP network are communicated to carry out data information interaction, thereby meeting the requirements of network stability, reliability and time delay, being applied to the infrastructure construction of scenes such as Internet of vehicles, field industrial control, telemedicine and the like, and being combined with SRv6 technology to complete centralized network path planning. The method can be directly applied to the deterministic network of industrial control, improves the industrial automation level and the production efficiency, and directly generates economic benefits. In the construction management and operation and maintenance work of the deterministic network, the work of related personnel is assisted, the effective utilization rate of the network bandwidth is improved, and unnecessary network data retransmission is avoided.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.
Claims (7)
1. A fault-tolerant prefabrication method usable with a deterministic network, comprising:
step 1: the fault-tolerant prefabrication module carries out DetNet flow identification detection on a control plane of a deterministic network and carries out historical data statistics;
step 2: according to the statistical historical data, a prediction model based on an artificial intelligence algorithm is combined to predict whether future data need to be subjected to the DetNet flow copying operation, if so, the DetNet flow copying operation is configured and issued, and a path and measures for data packet loss are guided;
and step 3: and the path planning module guides the global network flow path by utilizing centralized path setting by means of a control plane of the deterministic network.
2. The fault-tolerant prefabrication method for the deterministic network according to claim 1, characterized in that step 1, in the control plane of the deterministic network, the DetNet traffic information of the data plane is periodically collected in seconds and stored in the database server;
the DetNet traffic information includes, but is not limited to, information of time stamp, packet coding, stream copy label, stream erasure label, each node through which the DetNet stream passes, whether the packet is used label, stream source address, stream destination address.
3. The fault-tolerant prefabrication method for deterministic networks according to claim 1, characterized in that step 2 said fault-tolerant prefabrication module provides a user interface for defining the data volume of the training verification set, the prediction data input output and the selection of the prediction model.
4. The fault-tolerant prefabrication method for deterministic networks according to claim 1, characterized in that the prediction model of step 2 comprises a prediction model based on decision trees, random forests, support vector machines.
5. The fault-tolerant prefabrication method for deterministic networks according to claim 1, wherein the step 2 takes data in a database server as training verification data input through a prediction model based on an artificial intelligence algorithm in a control layer, and outputs whether a link needs to perform a stream replication operation and the number of stream replications in the next time period, specifically:
with DetNet flow data as a training verification set, sequentially carrying out preprocessing, flow replication behavior segmentation, data set feature extraction and feature set construction on the training verification set, then carrying out behavior identification by using data at the front end of a time sequence in the set, training a prediction model, carrying out identification verification by using subsequent residual data on the time sequence, and then predicting the DetNet flow information of a link in the next time period, wherein the specific prediction result comprises the following steps: and if the stream replication operation is needed, the number of output stream replication is determined.
6. The fault-tolerant prefabrication method for deterministic networks according to claim 1, characterized in that said step 3 is implemented by PCE, directly guiding the explicit routing of the data layer of the deterministic network, specifically:
the PCE calculates the constraint path of the global network in a centralized mode, performs whole network routing guidance by means of SR/SRv6 technology, and issues the path to the data plane, and the data plane issues the path to the PCC network equipment through PCEP.
7. The fault-tolerant prefabrication method for the deterministic network as claimed in claim 1, wherein the fault-tolerant prefabrication module and the path planning module are both used as separate program component modules, and interface with an existing SDN controller and a network management platform to complete management functions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111514275.3A CN114374617A (en) | 2021-12-13 | 2021-12-13 | Fault-tolerant prefabricating method for deterministic network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111514275.3A CN114374617A (en) | 2021-12-13 | 2021-12-13 | Fault-tolerant prefabricating method for deterministic network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114374617A true CN114374617A (en) | 2022-04-19 |
Family
ID=81139239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111514275.3A Pending CN114374617A (en) | 2021-12-13 | 2021-12-13 | Fault-tolerant prefabricating method for deterministic network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114374617A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114900509A (en) * | 2022-05-18 | 2022-08-12 | 华中科技大学 | Method for generating flow demand description protocol suitable for deterministic network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180123887A1 (en) * | 2016-10-31 | 2018-05-03 | Cisco Technology, Inc. | Full path diversity for virtual acess point (vap) enabled networks |
| CN110417707A (en) * | 2018-04-27 | 2019-11-05 | 中兴通讯股份有限公司 | Data send guard method, device, system and computer readable storage medium |
| CN112511995A (en) * | 2020-03-30 | 2021-03-16 | 中兴通讯股份有限公司 | Message interaction method, device, equipment and storage medium |
| CN112910690A (en) * | 2021-01-18 | 2021-06-04 | 武汉烽火技术服务有限公司 | Network traffic prediction method, device and equipment based on neural network model |
| CN113596146A (en) * | 2021-07-27 | 2021-11-02 | 彭亮 | Resource scheduling method and device based on big data |
| US20210385148A1 (en) * | 2020-06-03 | 2021-12-09 | Cisco Technology, Inc. | Determining node behavior in deterministic networks |
-
2021
- 2021-12-13 CN CN202111514275.3A patent/CN114374617A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180123887A1 (en) * | 2016-10-31 | 2018-05-03 | Cisco Technology, Inc. | Full path diversity for virtual acess point (vap) enabled networks |
| CN110417707A (en) * | 2018-04-27 | 2019-11-05 | 中兴通讯股份有限公司 | Data send guard method, device, system and computer readable storage medium |
| CN112511995A (en) * | 2020-03-30 | 2021-03-16 | 中兴通讯股份有限公司 | Message interaction method, device, equipment and storage medium |
| US20210385148A1 (en) * | 2020-06-03 | 2021-12-09 | Cisco Technology, Inc. | Determining node behavior in deterministic networks |
| CN112910690A (en) * | 2021-01-18 | 2021-06-04 | 武汉烽火技术服务有限公司 | Network traffic prediction method, device and equipment based on neural network model |
| CN113596146A (en) * | 2021-07-27 | 2021-11-02 | 彭亮 | Resource scheduling method and device based on big data |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114900509A (en) * | 2022-05-18 | 2022-08-12 | 华中科技大学 | Method for generating flow demand description protocol suitable for deterministic network |
| CN114900509B (en) * | 2022-05-18 | 2023-06-13 | 华中科技大学 | Method for generating flow demand description protocol applicable to deterministic network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Junges et al. | Evaluating the performance of DCOP algorithms in a real world, dynamic problem | |
| CN109995546B (en) | Intelligent factory automation system with edge computing and cloud computing cooperating | |
| Vyas | Optimizing Data Ingestion and Streaming for AI Workloads: A Kafka-Centric Approach | |
| Lu et al. | Dynamit2. 0: Architecture design and preliminary results on real-time data fusion for traffic prediction and crisis management | |
| CN110224427B (en) | Information physical system modeling method based on micro-grid energy control strategy | |
| CN114374617A (en) | Fault-tolerant prefabricating method for deterministic network | |
| CN115384586B (en) | Railway parallel scheduling system, method and application thereof | |
| WO2023241423A1 (en) | Routing decision method and system based on traffic prediction | |
| US10747956B2 (en) | Artificial intelligence process automation for enterprise business communication | |
| CN114785666A (en) | Network fault troubleshooting method and system | |
| CN116468287A (en) | Intelligent park control system based on digital twinning | |
| Liu et al. | Modeling and efficient passenger-oriented control for urban rail transit networks | |
| Huber et al. | Optimization of charging strategies for battery electric vehicles under uncertainty | |
| Shen et al. | Modeling transportation networks during disruptions and emergency evacuations | |
| CN117251825B (en) | Multi-sensor data fusion platform for new energy power station | |
| CN116974217B (en) | Factory production simulation prediction system and factory simulation prediction method | |
| Heckel | Stochastic analysis of graph transformation systems: A case study in P2P networks | |
| Liu et al. | Dynamic constraint and objective generation approach for real-time train rescheduling model under human-computer interaction | |
| CN116611813A (en) | Intelligent operation and maintenance management method and system based on knowledge graph | |
| CN112529419B (en) | Power grid data transparent application method and system based on correlation analysis | |
| Cao | A machine learning-based approach to railway logistics transport path optimization | |
| CN112231811B (en) | BIM-based urban medium-voltage distribution network transition net rack transformation method | |
| Dai et al. | Cooperative Path Planning of Multi-Agent Based on Graph Neural Network | |
| CN105571604A (en) | Co-evolved route optimizing method in dynamic road network environment | |
| Pu et al. | Short-Term Interval Prediction of Inbound Passenger Flow of Subway Station under Failure Events |
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 |