Design and Optimisation of Crash Prevention Platform for Online

Course Selection System Based on Network Traffic Analysis

Zirou Meng, Jing Chen

Lei Yang, Canyang Lin and Qi Zhang

Guangdong University of Science and Technology, Dongguan, China

Keywords: Online Course Selection System, System Crash, Error Handling, Network Traffic Analysis, Reliability.

Abstract: With the popularity and use of online course selection systems in universities, the system crash problem occurs

frequently, which seriously affects the smooth running of the course selection process of students. The current

methods to solve the system crash problem are limited to the traditional error handling and fault tolerance

mechanism techniques, which are ineffective. In order to solve this problem, this study proposes a new

solution with strong robustness and adaptability based on network traffic analysis - the online course selection

system crash prevention platform (NETCAP). The platform realizes real-time monitoring and analysis of

network traffic, and provides a stable and reliable operating environment for users. NETCAP platform is of

great significance to improve the reliability and user experience of the university course selection system, and

provides new ideas and methods for research in the corresponding fields.

1 INTRODUCTION

With the continuous progress of information technol

ogy and the popularity of education informatisation,

online course selection system in universities has bec

ome an indispensable and important part of modern e

ducation management. Online course selection syste

m provides students and teachers with convenient an

d efficient course selection services, greatly simplifie

s the course selection process, and improves teaching

quality and management efficiency (Yunpeng Bai, 2

015). However, with the rapid development of the on

line course selection system and the increase in the n

umber of users, the problem of course selection syste

m crash has gradually appeared, which seriously affe

cts the smooth running of students' course selection a

nd the normal operation of academic management.

First of all, students may encounter the crash of

course selection system during the peak period of

course selection leading to failure of course selection,

thus delaying the study plan, and may miss important

courses, and may even cause delays in graduation and

other effects. Failure to select the class of their choice

in the course selection process not only brings

unnecessary trouble and stress to students, but also

may lead to changes in their career development and

reduce their interest in learning the course. In

addition, for school administration, the collapse of the

course selection system will increase the workload of

the academic staff and lead to confusion in the course

selection process. Therefore, the stability and

reliability of the course selection system becomes a

serious challenge.

Currently, the methods to solve the problem of

course selection system crash mainly focus on the

traditional error handling and fault tolerance

mechanism techniques (Jinfu C, Qianlong Yang). The

traditional coping mechanism mainly focuses on the

known failure modes (YANG Qianlong, 2023), while

it can hardly cope with the unknown failure modes,

not to mention the inability to identify and correct the

potential problems in a timely manner. These

methods are limited in their ability to monitor and

predict the risk of system crashes in real time, and

often only temporarily solve the problem, unable to

fundamentally avoid and prevent the occurrence of

system crashes. In view of this, the current security

precautions have been difficult to adapt to the

increasing number of highly concurrent and complex

operating environments, and there is an urgent need

for a more functional and intelligent preventive

mechanism to ensure the stability and reliability of

the system.

To address this problem, this paper proposes a

system crash prevention platform based on network

traffic analysis, and applies this platform to the online

course selection system in universities. By

monitoring and analysing the network traffic of the

course selection system in real time, the platform is

Meng, Z., Chen, J., Yang, L., Lin, C. and Zhang, Q.

Design and Optimisation of Crash Prevention Platform for Online Course Selection System Based on Network Trafﬁc Analysis.

DOI: 10.5220/0012285700003807

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Seminar on Artiﬁcial Intelligence, Networking and Information Technology (ANIT 2023), pages 449-453

ISBN: 978-989-758-677-4

449

able to accurately identify the risk of system crash and

provide corresponding early warning and

optimisation measures (HUANG Ju, 2022).

Compared with traditional methods, NETCAP is

more accurate and flexible, and can detect the

operating status of the course selection system in real

time, predict potential crash risks, and take

corresponding measures to improve the stability and

reliability of the system. By introducing this

innovative platform, colleges and universities can

better cope with the problem that the course selection

system is prone to crashing under high concurrency

of network traffic, so as to ensure that students can

select courses smoothly and improve the efficiency of

academic affairs management.

The content of this paper is arranged as follows;

Section 2 describes in detail the design principles and

components of the crash prevention platform for

online course selection system, Section 3 gives the

experimental setup and the research results, Section 4

discusses the experimental results and optimisation

directions, and Section 6 concludes the whole paper.

2 ONLINE COURSE SELECTION

SYSTEM CRASH PREVENTION

PLATFORM

During the peak period of course selection, the system

traffic increases dramatically, which will cause the

system to be in a high load operation state, which will

lead to system crashes and other risks. Therefore, this

paper proposes a Network Traffic Analysis-based

Online Course Selection System Crash Prevention

(NETCAP). The platform aims to improve the

stability and reliability of the system by monitoring

the network traffic of the course selection system in

real time, analysing the traffic data and taking

corresponding preventive measures. The overall

architecture of the platform adopts a distributed

architecture and consists of a data collection module,

a data processing module, a traffic analysis module, a

prediction module and an early warning mechanism

module. The design principles include scalability,

flexibility and high performance. The platform will

adopt open interfaces to facilitate future functional

expansion and customisation requirements.

2.1 Data Acquisition

The data collection module of the NETCAP platform

is responsible for collecting network traffic data of

the course selection system, including requests,

responses, user operations and other information, and

storing them in a distributed database. The network

packets generated by the course selection system are

captured with the help of network traffic monitoring

tools, such as packet capture tools. This module is

capable of capturing the communication data between

the course selection system and the users and

transmits the captured raw data to the data processing

module in real time.

Figure 1 below shows the raw network data

captured using the packet capture tool.

Figure 1. Data acquisition.

2.2 Data Processing

After the data collection is completed, the NETCAP

platform will use data preprocessing techniques to

process the packet data captured in the previous step

by data cleaning, data filtering and other necessary

processes, aiming to filter and organise the data for

use in subsequent data analysis and modelling. Using

feature extraction algorithms, features such as request

frequency, request type and response time of network

traffic data were extracted. However, this involves

significant computational and storage resources. In

order to improve the performance of the NETCAP

platform, the design employs a number of

optimisations including parallel computing, data

compression and caching techniques.

2.3 Flow Analysis

The goal of flow analysis (Fangqiang Jiang,

Zongzhen Gao) is to extract valuable information

from massive network traffic data and perform

anomaly detection and prediction. Based on the

collected network traffic data, this module analyses

the traffic patterns through feature extraction,

anomaly detection and crash prediction algorithms,

and then predicts the crash possibility of the course

selection system. Specifically, the module will extract

various traffic features, such as request frequency,

request type, response time, etc., and analyse

historical traffic data and anomaly detection

ANIT 2023 - The International Seminar on Artiﬁcial Intelligence, Networking and Information Technology

450

algorithms to detect abnormal system behaviour.

Meanwhile, the module will use collapse prevention

algorithm to predict whether a crash will occur in the

course selection system and make decisions for

subsequent preventive measures. After obtaining

analysable data, the NETCAP platform uses traffic

analysis methods to process the collected network

traffic data.

2.4 Automatic Early Warning

Response

In the event of large concurrent traffic in the system,

the platform will automatically trigger the emergency

response mechanism to enable load balancing,

backup channel and other processing measures for the

system, and at the same time inform the administrator

by email that the current server is in the state of high

concurrent network traffic, and if necessary, carry out

manual technical maintenance measures in order to

protect the stability and security of the system.

2.5 Key Functions and Module Design

The key functions of the platform include real-time

monitoring, crash prediction and early warning

mechanism. The real-time monitoring module is

responsible for monitoring the network traffic of the

course selection system in real time and transmitting

the data to the flow analysis module for processing.

The crash prediction module uses machine learning

algorithms to build a prediction model based on the

flow analysis results to predict the crash probability

of the course selection system. In the early warning

mechanism module, the NETCAP platform will

perform network traffic analysis, and when the

monitored network traffic exceeds the traffic

threshold set by the system, the platform will

automatically trigger the early warning mechanism

and send an alert message to the system administrator,

so that timely measures can be taken to avoid the

crash of the course selection system. In addition, the

platform also provides a data visualisation interface

for administrators to view flow analysis results and

system status. The design of specific functional

modules is shown in Figure 2 below.

Figure 2. System design flowchart.

3 EXPERIMENTS

In terms of experiments, a series of experiments were

designed and conducted to verify the system crash

prevention function and performance of the NETCAP

platform. During the experiments, the project

conducted simulation tests based on real course

selection system traffic as a way to evaluate the

platform's performance and prediction accuracy, and

at the same time optimise for possible bottlenecks.

First, the network traffic data of the real course

selection system was collected and stored in a

distributed database by the data collection module of

NETCAP. Then pre-processing, including data

cleaning and filtering, was performed on the collected

data to improve the accuracy and efficiency of the

subsequent analyses. Second, flow analysis and

system crash prediction experiments were conducted.

By analysing the historical traffic data after data

processing and combining it with anomaly detection

algorithms, the system's abnormal behaviours,

including abnormal request frequency and abnormal

response time, were successfully discovered. Third,

the crash probability of the course selection system

was analysed and predicted using a crash prediction

algorithm. A machine learning model was built,

trained using historical traffic data and crash

occurrences, and predicted for future traffic data.

Experimental results show that the crash prediction

model proposed in this paper performs well in terms

of accuracy and robustness, and can effectively

predict the crash probability of the course selection

system.

Design and Optimisation of Crash Prevention Platform for Online Course Selection System Based on Network Trafﬁc Analysis

451

The NETCAP platform automatically triggered an

emergency response mechanism when a possible

system crash was predicted. Through processing

measures such as load balancing and backup channel,

the stability and availability of the system were

successfully improved. At the same time, alert

messages were sent to administrators via email so that

timely manual technical maintenance measures could

be taken to safeguard the security and stability of the

system. To assess the performance and optimisation

potential of the platform, performance tests were also

conducted to simulate network traffic of different

sizes and loads, and to monitor the platform's

response time and system resource utilisation. During

the tests, it was found that the platform exhibited

large latency and resource consumption when

handling large-scale concurrent traffic.

To optimise the platform, the following measures

were taken. First, the data collection module was

optimised to improve the data collection speed and

storage efficiency. Second, the traffic analysis

algorithm was optimised to reduce the time

complexity of feature extraction and anomaly

detection. Finally, the scalability of system resources

was increased, and the concurrent processing

capability of the system was improved through

distributed deployment and load balancing. After

optimisation, the NETCAP platform achieved

significant improvements in performance tests. The

average response time was reduced by 40% and the

system resource utilisation increased by 30%. The

experimental results show that the optimisation

measures taken in this project have effectively

improved the performance and reliability of the

platform.

Finally, in order to improve the scalability of the

NETCAP platform, this paper also conducts a series

of evaluations on this. Measures were finally taken to

gradually increase the size of the course selection

system and the number of users, and the operation of

the platform was monitored. The experimental results

show that the NETCAP platform exhibits good

scalability and performance in handling large-scale

traffic and multi-user requests.

In summary, this experiment verifies the

functionality and performance of the system crash

prevention platform of NETCAP online course

selection system. Through real-time monitoring, flow

analysis and crash prediction, it can effectively

prevent the crash of the course selection system and

guarantee the stability and reliability of the system.

The experimental results show that the NETCAP

platform has the potential to improve the stability and

availability of the system, and it is able to identify and

predict system crashes in time and carry out

maintenance measures in time, which ensures that the

course selection based on this system can be carried

out smoothly.

4 DISCUSSION AND

OPTIMISATION

The system crash prevention platform for online

course selection system based on network traffic

analysis proposed in this paper has achieved some

results, but there are still shortcomings. The following

is an explanation and discussion of the experimental

results, while some ideas and improvement

suggestions for optimising the system design and

algorithms will also be proposed in this study.

Firstly, for the experimental results, it is observed

that the NETCAP platform is able to predict the

changes in system load and the risk of crashes more

accurately. By analysing and predicting the real-time

network traffic data, it is able to take appropriate

measures to reduce the load and ensure the stability

of the system before it crashes. This result

demonstrates the effectiveness and usefulness of the

NETCAP platform in preventing system crashes.

However, there are some problems and

shortcomings in the design. Firstly, the collection and

processing of flow data in a large-scale course

selection system still presents some challenges.

Despite the optimisation methods used to improve the

performance and efficiency of the platform, there are

still some latency and data processing capacity

limitations. Therefore, when deploying the NETCAP

platform in an application, careful consideration

needs to be given to the scale of the system and the

scalability of data processing to ensure the accuracy

and timeliness of the platform.

Second, under special circumstances, such as

cyber-attacks or unexpected events, the NETCAP

platform may have some prediction errors. This is

because these special situations lead to abnormal

changes in system load, which affects the accuracy of

the prediction algorithm. Therefore, in the future

optimisation, the introduction of more complex

algorithms and models will be considered to

accommodate more abnormal situations and improve

the robustness of the platform.

For the system design and algorithmic aspects, it is

proposed to optimise and improve the following

aspects.

1) Optimisation of Data Acquisition and

Processing

ANIT 2023 - The International Seminar on Artiﬁcial Intelligence, Networking and Information Technology

452

A distributed data acquisition and processing

system can be considered to improve the processing

capacity and response speed of the platform. In

addition, data storage and compression techniques

can be further optimised to reduce the cost of data

processing and storage.

2) Model Optimisation and Improvement

Further in-depth research on flow analysis

algorithms can be conducted to explore more features

and models to improve the accuracy of prediction.

Meanwhile, emerging technologies such as artificial

intelligence algorithms and deep learning methods

can be considered to be introduced to adapt to more

complex system load changes and anomalies.

3) Platform Monitoring and Feedback Mechanism

A comprehensive monitoring and feedback

mechanism can be designed to monitor system

performance and load in real time, and to respond and

adjust platform parameters and strategies in a timely

manner. This can further improve the adaptability and

reliability of the platform.

In summary, although certain research results have

been achieved in this paper, there are still some

problems and shortcomings that need to be further

solved and improved. By optimising the data

processing, improving the model and designing the

monitoring and feedback mechanism, the

performance and effect of the NETCAP platform can

be further improved to provide a better guarantee for

the stability and reliability of the online course

selection system.

5 CONCLUSION

With the rapid development of online course selection

system and the increase in the number of users, the

problem of course selection system crash is gradually

revealed, which seriously affects the smooth

operation of students' course selection and the normal

operation of academic affairs management. Based on

this paper, NETCAP, a crash prevention platform for

online course selection system based on network

traffic analysis, is proposed and optimised to some

extent. The performance of the online course

selection system is studied in depth and the network

flow crash problems it encounters are identified. The

experimental results show that the NETCAP platform

proposed in this paper is able to predict and prevent

system crashes based on traffic analysis, and plays a

positive role in practical applications. In future

research, all aspects of the NETCAP platform can be

further improved and refined to enhance its prediction

performance and scalability.

ACKNOWLEDGMENTS

This research was funded by Dongguan Kunpeng Key

Laboratory of Computing; Research Capability

Enhancement Project of Guangdong University of

Science and Technology: Application Research of

Artificial Intelligence Technology Based on Kunpeng

Computing Platform; Natural Science Project of

Guangdong University of Science and Technology,

grant number gky-2022kyzdk-12; Innovation and

School Strengthening Project of Guangdong

University of Science and Technology, grant number

gky-2022cqtd-4.

REFERENCES

Yunpeng Bai. Design and Implementation of Online Course

Selection System Based on B/S Mode(D).Jilin

University,2015.(in Chinese)

Jinfu C,Tianxiang L,Saihua C, et al. A novel detection

model for abnormal network traffic based on

bidirectional temporal convolutional network (J).

Information and Software Technology, 2023, 157.

https://doi.org/10.1145/3613960

Qianlong Yang,Lingyun Jiang. Research on Microservice

Load Balancing Algorithm Based on Machine Learning

(J).Computer Science, 2023, 50(05): 313-321.(in

Chinese)

YANG Qianlong, JIANG Lingyun. Study on Load

Balancing Algorithm of Microservices Based on

Machine Learning (J). Computer Science, 2023, 50(5):

313-321. (in Chinese)

https://doi.org/10.11896/jsjkx.220400019

HUANG Ju, HE Maozhong, ZHANG Xiaomeng, et al.

Course selection system for colleges and universities

based on SDN load balancing (J). Modern Information

Technology, 2022, 6(12): 71-73.

https://doi.org/10.19850/j.cnki.2096-

4706.2022.012.018.(in Chinese)

Fangqiang Jiang. Design and implementation of an

application-oriented network traffic monitoring system

(J). Henan Science and Technology, 2021, 40(10):9-12.

(in Chinese) https://doi.org/10.3969/j.issn.1003-

5168.2021.10.012

Zongzhen Gao. Research and Implementation of Online

Course Selection System Based on Load

Balancing(D).Ocean University of China, 2012.(in

Chinese) https://doi.org/10.7666/d.y2158455

Design and Optimisation of Crash Prevention Platform for Online Course Selection System Based on Network Trafﬁc Analysis

453