A Workload-Aware Energy Model for Virtual Machine Migration

Abstract

Abstract-Energy consumption has become a significant issue for data centres. Assessing their consumption requires precise and detailed models. In the latter years, many models have been proposed, but most of them either do not consider energy consumption related to virtual machine migration or do not consider the variation of the workload on (1) the virtual machines (VM) and (2) the physical machines hosting the VMs. In this paper, we show that omitting migration and workload variation from the models could lead to misleading consumption estimates. Then, we propose a new model for data centre energy consumption that takes into account the previously omitted model parameters and provides accurate energy consumption predictions for paravirtualised virtual machines running on homogeneous hosts. The new model's accuracy is evaluated with a comprehensive set of operational scenarios. With the use of these scenarios we present a comparative analysis of our model with similar state-of-the-art models for energy consumption of VM Migration, showing an improvement up to 24% in accuracy of prediction.

Modelling resource heterogeneities in cloud simulations and quantifying their accuracy

“…Most importantly, the accuracy of the modelled energy behaviour is shown to be in line with the accuracy of the real CPU model in use [SPECpower benchmarks]. Moreover, DISSECT-CF simulator also models the energy consumption of a VM migration as described in [30], [35]. Note that similar energy models and benchmarks, except the VM migration, are also used in CloudSim and several other cloud simulators such as DCSim [5].…”

“…However, the energy consumption of the source host goes down with increase in CPU utilization of the migrated VM. Another energy consumption model for live migration is presented in [30]. This model increases the accuracy of the model presented in [29] by ∼3.9% (11.8% vs. 15.7% error) for the source host and by ∼7.9% (5% vs 12.9% error) for the target host.…”

“…This model increases the accuracy of the model presented in [29] by ∼3.9% (11.8% vs. 15.7% error) for the source host and by ∼7.9% (5% vs 12.9% error) for the target host. The authors [30] have suggested an accuracy improvement of ∼24% over the model in [29]. With these contradictory results, it is reasonable to assume the VM size (data copied) as a major factor to estimate the energy consumption of a VM migration, as suggested in [12].…”

An experiment-driven energy consumption model for virtual machine management systems

“…In [41], the authors present a model depending on the CPU utilization (considering both hosts and the migrated VM), the VM memory utilization, the available bandwidth and the memory dirtying rate. The evaluation is conducted through live-migrating a VM in a host where there are already 8 running VMs.…”

Exact and Hybrid Solutions for the Multi-Objective VM Reassignment Problem

“…We do not claim to be exhaustive or even totally accurate and we refer the readers to 39,40,41 for more details on migration costs. Here we only assume that migration has a non-negligible cost in data centres and we use the model and values given by the challenge mentioned above.…”