| Summary: | Nowadays, computer networks are used in education, sports, business, transportation, manufacturing, and more. Thus, it makes the network more complex and harder to monitor, especially in terms of performance. The implementation of network monitoring software presents a viable solution for resolving this issue. It is because network monitoring software can monitor behaviours, provide alertness for any failing component of a computer network, and measure performance such as CPU usage, bandwidth, throughput, response time, and packet loss. However, network monitoring software has common problems, such as being hard to set up and making it hard to find the best network monitoring that fits the requirements. Thus, this study aims to identify network monitoring software that is easy to configure and analyse the performance of network monitoring software in terms of response time and packet loss. The study comprises the implementation of open-source network monitoring software, including PRTG, OpManager, Zabbix, and LibreNMS. The two experiments were conducted by launching TCP attacks for response time and UDP attacks for packet loss. Both attacks consisted of four scenarios: 15 threads, 30 threads, 45 threads, and 60 threads. As a result, PRTG offers a simpler installation procedure and easier configuration options, with a straightforward and attractive Graphical User Interface (GUI) that allows for all configurations to be conveniently performed directly through the interface. For monitoring websites, users need to add the device that was targeted by the website and choose an HTTP sensor for response time and PING for packet loss. The evaluation was based on the lowest average response time and consistency in the detection of packet loss. Consequently, Zabbix demonstrates the highest level of responsiveness about web services, which have the lowest average response time and consistently exhibit the lowest occurrence of packet loss.
|
|---|