Online virtual machine migration for renewable energy usage maximization in geographically distributed cloud data centers

Abstract

Summary Energy consumption and its associated costs represent a huge part of cloud providers' operational costs. In this study, we explore how much energy cost savings can be made knowing the future level of renewable energy (solar/wind) available in data centers. Since renewable energy sources have intermittent nature, we take advantage of migrating virtual machines to the nearby data centers with excess renewable energy. In particular, we first devise an optimal offline algorithm with full future knowledge of renewable level in the system. Since in practice, accessing long‐term and exact future knowledge of renewable energy level is not feasible, we propose two online deterministic algorithms, one with no future knowledge called deterministic and one with limited knowledge of the future renewable availability called future‐aware. We show that the deterministic and future‐aware algorithms are 1+1/s and 1+1/s−ω/s.Tm competitive in comparison to the optimal offline algorithm, respectively, where s is the network to the brown energy cost, ω is the look‐ahead window‐size, and Tm is the migration time. The effectiveness of the proposed algorithms is analyzed through extensive simulation studies using real‐world traces of meteorological data and Google cluster workload.

A Taxonomy and Future Directions for Sustainable Cloud Computing

“…Kernel-based Virtual Machine (KVM) is used to test the proposed technique to prove its effectiveness to calculate the migration time more accurately. Khosravi et al [106] developed a technique to improve the usage of renewable energy for Online Virtual Machine Migration (OVMM), from one physical server to another, to enable sustainable cloud computing. In this research work, an optimal offline algorithm and an online algorithm are proposed for VM migration [141].…”

“…Furthermore, the energy consumption depends on the size of the VM, which is migrated. Remarkably, for the similar set of VMs, different orders of migrations lead to different migration-time based on VM size [102] [106]. So, there is a requirement to investigate the trade-off between energy cost and migration time of VMs.…”

Green IT scheduling for data center powered with renewable energy

“…In the scope of our work, we explore the on-site renewable energy sources case and try to take advantage of the temporal diversity and flexibility of the workload. Several works focus on geographically distributed data center around the world [7][8][9], each having either on-site or offsite renewable energy sources. This perspective allows to exploit the spatial aspects of the workload.…”

“…One of these perspectives is to take advantage of the geographical distribution of such data centers [7][8][9]. Indeed, the weather conditions between distant places are little correlated.…”

Network-Aware Energy-Efficient Virtual Machine Management in Distributed Cloud Infrastructures with On-Site Photovoltaic Production

“…The remaining execution time of the VMs and the energy cost of the live migration are also not considered when migrating a VM. • OOD-MARE [18] is another approach from literature consisting in allocating the incoming and running VMs according to the current local green power productions. With this approach, a Most Available Renewable Energy (MARE) algorithm deploys the VMs on the DC that has the highest amount of available green energy.…”