Earticle

With the rapid development of the Internet, the antivirus software of the network is always emerging and constantly changing. Traditional detection methods can’t effectively kill the new viruses and malicious software, the complexity of which also makes itself easy to be attacked by malicious software. The emergence of cloud computing has changed the status quo. Therefore, the architecture model of virus malware detection based on cloud computing is proposed in this paper. Based on the combination of the method for detecting malicious software virus based on cloud computing and the algorithm analysis theory in machine learning, a new type of distributed CFO algorithm is proposed, and the closed environment of cloud computing virtual machine nodes is used to realize dynamic behavior monitoring to the virus malware, then the distributed fluctuations PIF algorithm is used to describe the process of dynamic analysis and analysis reporting, besides, the wave algorithm is carried out corresponding improvement based on the analysis of the environment. Experimental results show that the model can detect the conditional trigger behavior of virus malware, so as to find the conditions for triggering malicious behavior and the input data that satisfy these conditions and the performance of this monitoring system is greatly improved compared with the common single machine system.