An energy aware framework for virtual machine placement in cloud federated data centres

Abstract

International audienceData centres are powerful ICT facilities which constantly evolve in size, complexity, and power consumption. At the same time users' and operators' requirements become more and more complex. However, existing data centre frameworks do not typically take energy consumption into account as a key parameter of the data centre's configuration. To lower the power consumption while fulfilling performance requirements we propose a flexible and energy-aware framework for the (re)allocation of virtual machines in a data centre. The framework, being independent from the data centre management system, computes and enacts the best possible placement of virtual machines based on constraints expressed through service level agreements. The framework's flexibility is achieved by decoupling the expressed constraints from the algorithms using the Constraint Programming (CP) paradigm and programming language, basing ourselves on a cluster management library called Entropy. Finally, the experimental and simulation results demonstrate the effectiveness of this approach in achieving the pursued energy optimization goals

Constraint Programming Based Large Neighbourhood Search for Energy Minimisation in Data Centres

“…All the instances are available online. 6 We observed on this benchmark that the cpu constraint is the tight one, as opposed to the memory constraint which is always satisfiable. Based on the original instances we also generated larger instances by just duplicating each virtual machine and each server.…”

“…Constraint programming based approaches have also been used previously to solve some related problems [6,3,8]. In [3] a data centre is viewed as a dynamic bin packing system where servers host applications with varying resource requirements and varying relative placement constraints.…”

“…Although in [6] an energy-aware framework is proposed for the reallocation of virtual machines in a data centre to reduce the power consumption, the goal is to find the best possible placement of virtual machines for a given time-period subject to service level agreements.…”

Planning Live-Migrations to Prepare Servers for Maintenance

“…BtrPlace is an extensible VM manager that can be customized to augment its inferring capabilities. It provides a composable VM placement algorithm that has already been used to address energy-efficiency [8], or scheduling concerns such as the continuous respect of server licensing policies [6]. The use of BtrPlace might also be beneficial to support side constraints that have to be expressed by the operators.…”

A Catalogue of Green Architectural Tactics for the Cloud

“…Example. Dupont et al [24] provide a sample implementation of the Consolidation tactic, depicted in Figure 2b. The Power Calculator, of type Energy Model, provides a power consumption estimation to the CP Engine, of type SLA Violation Checker.…”