Type-2 Fuzzy Logic Approach for Overloaded Hosts in Consolidation of Virtual Machines in Cloud Computing

Abstract

Dynamic consolidation of virtual machines (VMs) is an effective way to improve resource utilization and power efficiency in cloud computing. Determining when it is best to relocate VMs from an overloaded host is one aspect of dynamic VM consolidation that directly influences the resource utilization and Quality of Service (QoS) offered by the system. This paper presents a new proposal with a Type-2 Fuzzy Logic approach to address the uncertainties and inaccuracies in determining resource usage, aiming at energy savings with minimal performance degradation. Results in a simulated cloud computing environment show improvements in energy efficiency of 15.64% in the Median Absolute Deviation (MAD), 15.43% in the Inter Quartile Range (IQR), and 9.19% in the Local Regression (LR).The following research questions stand out: (i) do load balancers consider the uncertainties associated with factors such as CP, CC, and RAM? (ii) are the load balancers equipped with consistent techniques for the treatment of the uncertainties related to cloud computing environments? and (iii) do the users have the necessary knowledge about the computational and communication demands of their applications?This proposal extends the approaches in [16,15], modeling the uncertain variables generating CP, CC, and RAM in Load Balancing (LB) considering Type-2 Fuzzy Logic (T2FL) which can preserve the balance between energy efficiency and manage resources while ensuring QoS through SLAs for such cloud-based services.The paper is structured as follows. Related works are presented in section 2. Section 3 introduces basic concepts of Type-2 Fuzzy Logic. In section 4, details of the Int-FLBCC component and its conception are discussed, including database, fuzzification, rule base, inference, and defuzzification. Section 5 describes the experimental evaluation. Finally, section 6 presents conclusions and future work.

Load Balancing of Cloud Computing Service Model Empowered with Fuzzy Logic

“…The author prescribed the most optimum and acceptable version of the 'ACO' algorithm in a cloud environment [20]. It prescribed a technique along with Type-2 fuzzy logic to abide by the unchallenged and effective behavior in assessing the host to load balancing VM appropriation in the cloud architecture [21]. Values of these input parameters are used to construct a lookup table, i.e., table I, for evaluating the balanced load.…”