DNS Server Caching and Forwarding with Load Balance
| Autor: | Hassan Dao, Nur Zaimah Ahmad, Megat F. Zuhairi, Eiad Yafi |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
business.industry
Network packet Computer science Domain Name System 020206 networking & telecommunications 02 engineering and technology Network Load Balancing Network planning and design Network congestion Wide area network Server 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Network performance business Computer network |
| Zdroj: | IMCOM |
| DOI: | 10.1109/imcom48794.2020.9001768 |
| Popis: | The Domain Name System (DNS) is an essential building block or component infrastructure of the Internet. Nearly every network applications depend on it to operate properly and the role of DNS. The purpose of this research project is to alleviate the network issue at Universiti Kuala Lumpur, Malaysian Institute of Information Technology (UniKL MIIT) student residences. A preliminary study of the network performance are indicative of unbalanced traffic, which lead to sub-optimal network performance. In principal, an enhanced network design is proposed, which accounts for network load balancing method. Such technique may alleviate network stress on a particular link i.e. Wide Area Network (WAN) interface. In the existing network, the system forwarded each traffic packet to the "best" WAN interface link and route to the DNS with standalone server on a particular WAN interface link. Such mechanism results in suboptimal network performance. The improved performance is measured using DNSPerf tool operated on linux platform, i.e. Ubuntu 16.04.2 LTS Operating System (OS). The tool is capable for benchmarking the authoritative servers to gather the result and measure the performance of DNS by emitting a configured number of queries. The research project fine-tune the MikroTik, used as a load balancer mechanism. Upon completion of the experiments, the results are analysed. The outcome of the study is twofold. Firstly an improved design of load-balanced network is proposed, which alleviates the network congestion. Data is analysed based on the real-time traffic that is continuously collected for a period of 1 month. Second, the improved network performance increases the quality of experience for network users in the residential area. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|