Classification of Network Traffic on ISP Link and Analysis of Network Bandwidth during COVID-19

Researchers explore the influence of apps on network resource consumption and how to estimate whether an application would impair network performance or consume computer system resources using accessible technologies such as predictive analytics, machine learning, and business intelligence. Having a healthy computer system and network has become increasingly important for business continuity in recent years. This study focuses on assessing performance data obtained from real networks during the COVID-19 pandemic, using scripts and existing applications intended expressly for monitoring apps and networks in real time. This work is critical because monitoring real-time performance does not provide precise or clear information about the causes of network or application performance degradation. Analyzing those performance measurements over time and finding a correlation between metrics and applications, on the other hand, provides much more important information about the fundamental cause of the issue. The findings demonstrated that there is a correlation between specified performance indicators and applications, achieving the study objectives. When determining the root cause of a problem in applications or networks is challenging, the study’s benefits can be proved in analyzing intricate networks. The goal of this chapter is to develop an algorithm utilizing machine learning techniques that will dynamically classify ISP network traffic into defined categories and aid in identifying traffic flow on the network at any given time. To create forecasting models based on previous bandwidth use data to forecast bandwidth requirements for an organization or business. The network and application details will be measured through several performance metrics such as bandwidth, throughput, port numbers, source IP, destination IP, number of packets send/Recv per sec, and number of bytes send/Recv per send.