Minimizing Total Busy Time with Application to Energy-efficient Scheduling of Virtual Machines in IaaS clouds

TL;DR

This paper presents EMinTRE-LFT, a scheduling algorithm designed to minimize total energy consumption in IaaS cloud environments by reducing the total busy time of physical machines, considering multiple resource constraints.

Contribution

The paper introduces a novel scheduling algorithm, EMinTRE-LFT, that effectively minimizes total energy consumption by focusing on total busy time, outperforming existing algorithms.

Findings

EMinTRE-LFT achieves the lowest total energy consumption in simulations.

The algorithm outperforms state-of-the-art methods in energy efficiency.

Simulations confirm the effectiveness of the proposed approach.

Abstract

Infrastructure-as-a-Service (IaaS) clouds have become more popular enabling users to run applications under virtual machines. Energy efficiency for IaaS clouds is still challenge. This paper investigates the energy-efficient scheduling problems of virtual machines (VMs) onto physical machines (PMs) in IaaS clouds along characteristics: multiple resources, fixed intervals and non-preemption of virtual machines. The scheduling problems are NP-hard. Most of existing works on VM placement reduce the total energy consumption by using the minimum number of active physical machines. There, however, are cases using the minimum number of physical machines results in longer the total busy time of the physical machines. For the scheduling problems, minimizing the total energy consumption of all physical machines is equivalent to minimizing total busy time of all physical machines. In this paper, we propose an scheduling algorithm, denoted as EMinTRE-LFT, for minimizing the total energy consumption of physical machines in the scheduling problems. Our extensive simulations using parallel workload models in Parallel Workload Archive show that the proposed algorithm has the least total energy consumption compared to the state-of-the art algorithms.