A Fuzzy Load Balancer for Adaptive Fault Tolerance Management in Cloud Platforms

Abstract

To achieve high levels of reliability, availability and performance in cloud environments, a fault tolerance approach to handle failures effectively is needed. In most existing research, the primary focus has been on explicit specification-driven solutions which requires too much effort for application developers, and leads to inflexibility. We propose a fuzzy job distributor (load balancer) for fault tolerance management to reduce levels of management complexity for the user. The proposed approach aims to reduce the possibility of fault occurrences in the system by a fair distribution of user job requests among available resources. In our self-adaptive approach, the system manages anomalous situations that might lead to failure by distributing the incoming job request based on the reliability of processing nodes, i.e., virtual machines (VMs). The reliability of VMs is a variable parameter and changes during its lifetime. Our approach is implemented and comparatively analysed using OpenStack. The experimental results show a significant reduction in the occurrence of faults in comparison with other load balancing algorithms.

A Comprehensive Study of Load Balancing Approaches in the Cloud Computing Environment and a Novel Fault Tolerance Approach

“…The failure detector unit recognizes irregularities and fault occurrences in the network during the life cycle of the system & gives signals to the defective node for retrieval mechanism. Remember that, in Figure 8, each component has its collection of functional aspects [110]. Our anomalies detector aims to avoid or identify anomalies proactively Misfits: (i) identify malicious unacceptable deterioration of the results (as a concrete anomaly) that could fail, (ii) recognize the signs and root issues of anomalous output loss to take appropriate remedial steps, utilizing Fuzzy work-sharing here, (iii) Handle the communications and connections between signs that are outward manifestations of anomalous behavior, the real issues that underlie the loss of results & Use a backup system to fine-tune future identification deficiencies and learning from authenticated outcomes to enhance effective identification of deficiencies and to keep improving implementation and implementation mechanisms through weight and primary concern modification.…”

“…Our anomalies detector aims to avoid or identify anomalies proactively Misfits: (i) identify malicious unacceptable deterioration of the results (as a concrete anomaly) that could fail, (ii) recognize the signs and root issues of anomalous output loss to take appropriate remedial steps, utilizing Fuzzy work-sharing here, (iii) Handle the communications and connections between signs that are outward manifestations of anomalous behavior, the real issues that underlie the loss of results & Use a backup system to fine-tune future identification deficiencies and learning from authenticated outcomes to enhance effective identification of deficiencies and to keep improving implementation and implementation mechanisms through weight and primary concern modification. The following activities are taken, associated with the system of the observing, inspection, organize, and implementation control loop [110].…”

“… Observing: This phase uses a ceilometer to gather information from the technique, design this information to provide a series demonstration that could be used to identify the obscured behavior in the information [110].  Inspection: The major elements of that stage are to define the dependency and the relation between defects, to determine the type of failure (failure strength degree of dispersal of an anomaly within the controlled resource), and to differentiate between defect (true diagnosis of an anomaly) and disturbance (false diagnosis) [110].  Organize & Implementation: This phase is related to the efficient fault tolerance technique Virtual machine for reassigning an ideal weight to the damaged elements to be capable of storing the verified routes by their correct weight [110].…”

“…Part of the method's result is to allow organizations to learn from errors [68]. The proposed solution focus to minimize the likelihood of fault existence in the system by making user job demands spread equally across existing resources [110].…”

“…FT in LB is one of the major challenges in cloud computing, which involves spreading workload uniformly to all nodes, detecting the fault and removing fault from the network, and sharing the workload to all nodes in order to maximize cloud network performance [68]. FT is a system's ability to perform its task accurately even though inner defects are present [110]. FT is one of the most key parameters because resource failure affects device efficiency, job performance, throughput, response time, and output.…”