This paper proposes a hierarchical management method of DNS (Domain Name S

ystem) blocked domain names based on big data analysis.

This method collects logs through the traffic mirror technology

which realizes real-time and efficient collection and analysis of DNS log data

avoids network congestion and inefficient analysis

and provides a strong guarantee for the security and stability of the Internet ecosystem. Compared with traditional DNS blocking strategies

this method not only reduces computational costs and resolution latency

but also improves system stability and security. The number of effective domain name blocking measures has decreased by 97.5%

resolution latency has decreased by 98%

and memory usage and CPU (Central Processing Unit) utilization have also decreased by 5%.

The innovation lies in the use of traffic mirroring technology for log collection

which enables real-time and efficient collection and analysis of DNS log data

By introducing machine learning algorithms and real-time data analysis techniques

intelligent classification and popularity evaluation of malicious domain names have been achieved

enabling precise management and optimization of domain name blocking. A hierarchical management strategy has been proposed

which formulates corresponding deletion rules based on the classification results and popularity scores of domain names

achieving refined management of blocked domain names.