Adaptive Resource Utilization Prediction System for Infrastructure as a Service Cloud

Abstract

Infrastructure as a Service (IaaS) cloud provides resources as a service from a pool of compute, network, and storage resources. Cloud providers can manage their resource usage by knowing future usage demand from the current and past usage patterns of resources. Resource usage prediction is of great importance for dynamic scaling of cloud resources to achieve efficiency in terms of cost and energy consumption while keeping quality of service. The purpose of this paper is to present a real-time resource usage prediction system. The system takes real-time utilization of resources and feeds utilization values into several buffers based on the type of resources and time span size. Buffers are read by R language based statistical system. These buffers' data are checked to determine whether their data follows Gaussian distribution or not. In case of following Gaussian distribution, Autoregressive Integrated Moving Average (ARIMA) is applied; otherwise Autoregressive Neural Network (AR-NN) is applied. In ARIMA process, a model is selected based on minimum Akaike Information Criterion (AIC) values. Similarly, in AR-NN process, a network with the lowest Network Information Criterion (NIC) value is selected. We have evaluated our system with real traces of CPU utilization of an IaaS cloud of one hundred and twenty servers.

Autonomous Virtual Machine Sizing and Resource Usage Prediction for Efficient Resource Utilization in Multi-Tenant Public Cloud

“…Datacenter resource prediction using neural networks ANN is a suitable method of prediction in our case because we have shown that it data has a non-Gaussian distribution. It has also shown good performance than other prediction models [29,38]. Moreover, ANN has a great ability to model a non-linear function thus able to handle complex time series [39,48,49].…”

“…Efficient management of power usage in data centres through prediction and VM sizing is on the rise [11,28,29]. However, the approaches used have not attempted to combine VM sizing and resource usage prediction.…”

“…ARIMA models have also been used in [33] to predict workload for achieving QoS. Finally, in [29] the authors proposes a data centre resource usage prediction that is based on either Autoregressive Integrated Moving Average (ARIMA) or Autoregressive Neural Network (AR-NN). Real-time resources usage in a server is monitored at intervals and is used to forecast future resource demands.…”

VM placement with effective energy management in cloud using optimal VM allocation framework (OVAF)

“…Load balancing at the IaaS level can be effectively done by predicting the future resource requirements of the new tasks and allocating the resources at the physical machine accordingly [24]. Based on the predicted load the set of physical machines is sorted in descending order.…”

Predictive Probabilistic Resource Availability based Cloud Workflow Scheduling (PPRA)

“…There could also be delay introduced due to VM provisioning and deprovisioning. IaaS cloud resources can be effectively utilized by predicting the future workload or the future resource utilization pattern [8].…”