Energy Conserving Secure VM Allocation in Untrusted Cloud Computing Environment

Abstract

Cloud computing is the latest buzz in most of the IT organizations which are witnessing a a trend of migration from traditional computing to cloud computing, thereby reducing their infrastructure cost and improving efficiency and performance. Cloud computing provides services through virtualization layer, which helps to execute more than one operating systems and applications on a single machine. Being a crucial part of cloud computing, virtualization layer faces major security threats, most challenging being an insider threat wherein attacker can either compromise existing virtual machines (VMs) or create rogue VMs. The objective of this work is to propose virtual machine (VM) allocation algorithm which operates in an untrusted cloud computing environment with non-trustworthy VMs. Our approach is based on the notion of trust. Lack of trust is modeled by either introducing faults or monitoring SLAs per host on which VMs are hosted. Detailed experiments considering varying cloud infrastructure and varying workloads are conducted using CloudSim. Results show that proposed algorithm works well in untrusted environment while at the same time is energy efficient and reduces the computational costs by decreasing the number of migrations and SLA violations.