Design And Implementation Of DNS Performance Prediction And Attack Detection System Based On Cloud Platform

In the rapid development of the network today, DNS as the core infrastructure of network services, its normal operation is the protection of Internet services. However, in the DNS security-related aspects, DNS in the design of the beginning of the existence of some imperfections in the DNS server data communication is not necessary to encrypt the two sides did not carry out an effective communication authentication mechanism and data integrity is not guaranteed, Which determines it is fragile, coupled with its important role in the network services, it is vulnerable to hackers and other criminals attack damage. Which is also necessary to carry out the corresponding measurement and forecasting of the monitoring.This paper first describes the DNS protocol and DNS design at the beginning of the shortcomings, the lack of DNS at the beginning of the design of the necessary security mechanisms, coupled with the increasing size of the system that the configuration frequency of failures and other reasons, resulting in DNS there are many Of the security risks. DNSSEC is a security extension of DNS, and it is a security mechanism designed to solve DNS spoofing and cache pollution. Although the data is not encrypted, but it provides a source of data authentication and data integrity verification. Then this article also describes several common DNS attack: DNS spoofing, DNS cache poisoning, DDOS attacks, DNS redirects and the use of the machine's hosts file hijacking and so on.Finally, we introduce several methods of DNS attack detection, and information entropy and time series analysis are the most commonly used and effective methods in DNS attack detection.The main work of this paper is as follows:1?Firstly, the regression prediction algorithm is introduced, and several regression prediction algorithms are introduced, such as linear regression algorithm, local weighted regression, ridge regression and forward stepwise regression. The linear regression prediction algorithm was selected by comparison. Finally, the analytical time of improved dimensionality reduction (fixed province, operator) is selected as the eigenvalue of the forecasting algorithm, and the design of the regression forecasting scheme is given by combining the square error and the correlation coefficient.2?Design and implement a DNS based performance prediction and attack detection system (DNS). The overall system architecture is divided into: application layer, cloud platform, data layer, data analysis presentation layer. The design and implementation of the four functions of DNS measurement and forecasting system: (1) measurement sites DNS resolution time (2) compared with the DNS analytical performance of different operators designated provinces (3) compared with the DNS parsing performance designated operators in different provinces (4)analytic prediction of time trend website DNS. The detailed design and implementation of the DNS measurement and prediction system are completed3?Build and deploy the DNS test measurement and forecasting system, but also on the design of four functional system ((1) measurement sites DNS resolution time (2) compared with the DNS analytical performance of different operators designated provinces (3) compared with the DNS parsing performance designated operators in different provinces (4) analytic prediction of time trend website DNS) for functional testing, all functions are to meet the design expectations, fully illustrated the prediction system design and realization scheme are feasible. Finally, the paper gives some suggestions on DNS attack detection.Through the above work, we provide a measurement and prediction of DNS service performance measurement system for enterprises with the user to select the DNS server provides a practical theoretical basis. To improve the domestic DNS server quality of service provides a real and effective theoretical basis. Finally, some suggestions on DNS attack detection are given.