Autonomic scaling of Cloud Computing resources using BN-based prediction models

ALPINE: A Bayesian System for Cloud Performance Diagnosis and Prediction

“…Recently, BNs were applied in the area of cloud computing (e.g., [8,7,15]). Bashar [7] use BNs for autoscaling of cloud datacenter resources by balancing the desired QoS and service level agreement targets. The author using preliminary studies show the BNs can be utilised efficiently to model workloads, and QoS factors like CPU usage and response time.…”

“…Compared to the state-of-the-art research in the area [8,7,15,9,16], the main aim of this paper is to develop a system for critical diagnosis and prediction of cloud performance under uncertainty. Our system, ALPINE, considers several factors such as time-of-the-day, day-of-the-week, virtual machine-type, regions and different types of benchmarks and efficiently models complex relationships between these parameters for cloud performance diagnosis and prediction.…”

Novel Framework for Fault Tolerant in a Cloud Environment using ACO Algorithm

“…Bayesian Networks (BN) based predictive modeling framework has been proposed, implemented and evaluated to provide for an autonomic scaling of utility computing resources in the Cloud Computing systems. The BN-based model captures the historical behavior of the system involving various performance metrics (indicators) and predicts the desired unknown metric which is the SLA parameter [9]. Hybrid learners solution has been implemented which simultaneously allow several machine learning algorithms to model QoS function and dynamically select the best model for prediction.…”

“… Opnet: It is a commercial network simulator which performs application performance management in order to deliver the application performance to the users and to satisfy business demands [25]. From the point of view of Cloud computing systems, Opnet does not have the capability to implement and test Infrastructure as a Service, however, it does have the capability for application testing for Cloud systems [26].  GreenCloud: It is a simulation environment for energy-aware cloud computing data centers based on the ns2 platform.…”

Handling uncertainty in cloud resource management using fuzzy Bayesian networks

“…The recent surge in the popularity and usage of cloud computing services by both enterprise and individual consumers has necessitated the efficient and proactive management of data center resources that host services with a variety of characteristics. One of the major issues concerning both cloud service providers and consumers is real time resource management in response to highly unpredictable demands [3]. These unpredictable demands cause the workload of VMs to fluctuate dynamically, leading to imbalance in both the load and utilization of virtual and physical cloud resources.…”