As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Since computers are universalized in every aspect of modern applications, detecting malware embedded in computer systems is essential in protecting user privacy, robustness of services, and data integrity. With the rapidly growing popularity of cloud computing technologies, although the underlying system virtualization enhances the protection of service systems, it also brings new security challenges. For example, it can be less cost-effective, both to the cloud vendor and the cloud service tenants, to independently execute security services within every virtualized system based on common cloud pricing schemes. Also, malicious tenants may disrupt other tenants' service operations by means such as exhausting shared computing resources. In this paper, we focus on detecting malware running in virtual machines using the virtual machine introspection framework. The malware behaviors, in particular, the sequential system calls translated by the hypervisor of malware-infected virtual machines, are analyzed by tensor factorization techniques. By segmenting the system logs according to execution time or accumulated system call transition counts, our experiments shows that the proposed tensor-based detection approach can detect most types of malware with good accuracy using relatively short log segments.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.