A.Peter soosai anandaraj and G. Indumathi. Energy Aware Based Optimal Load Distribution and Attack Detection in Cloud Computing Environment. Dynamic Systems and Applications 29 (2020) No. 6, 2325 – 2346

https://doi.org/10.46719/dsa202029611

ABSTRACT:
Cloud providers and users are easily prone to a dreadful attack called distributed denial of service (DDos). Intrusion detection systems are employed to detect these attacks of which fuzzy extreme learning machine classifier (FELM) is the contemporary technique. Rendering good quality services with low energy consumption is a challenging task for cloud providers. Nonetheless, it is resolved using Bernoulli distribution function that is introduced to each and every host to find the energy requirement of them depending upon their availability; whether they are busy or idle in this research a trust model is framed using improved graph clustering algorithm and dynamic support vector machine classifier (DSVM) has been applied to detect attacks in a cloud computing environment. Kernel function and hypervisors are used in the DSVM classifier to increase the attack detection rate and aggregate the monitoring results respectively, at last optimal load distribution is detected among virtual machines by Integrated Bee Cuckoo Search (IBCS) algorithm. The proposed method is compared with other existing methods using metrics such as attack detection, false-positive rate, false-negative rate, etc. For different counts of a virtual machine, results show that the proposed method is energy efficient and can detect attacks and distribute the load optimally.

Keywords: Energy Consumption, Bernoulli distribution, Hypervisor, Distributed Denial of Service (DDoS) attacks, trust relationships, Integrated Bee Cuckoo Search (IBCS), Dynamic Support Vector Machine (DSVM).