Affiliation of Author(s):计算机科学与技术学院/人工智能学院/软件学院

Journal:Proc. - IEEE Int. Conf. Big Data Secur. Cloud, BigDataSecurity, IEEE Int. Conf. High Perform. Smart Comput., HPSC IEEE Int. Conf. Intell. Data Secur., IDS

Abstract:Hadoop is a very popular big data processing framework, however, due to its distributed and large-scale characteristics, its security problems have not been solved very well. Existing research does not systematically analyze attacks in big data platforms. This paper proposes four innovative hadoop attack models. Through adjusting heartbeat time, tampering intermediate data, blocking network, attackers prolong the execution time of jobs, and damage the correctness of job result. We implemented these attacks in hadoop and evaluate the effects of them through experiments. The experimental results show that our attacks are effective and harmful. © 2019 IEEE.

Translation or Not:no

Date of Publication:2019-05-01

Co-author:Li, Ningwei,Liu Liang,Zhang, Fan,zfh,Wang, Wenxuan

Correspondence Author:gh