Future Generation Computer Systems ( ) – Contents lists available at SciVerse ScienceDirect Future Generation Computer Systems journal homepage: www.elsevier.com/locate/fgcs A service framework for energy-aware monitoring and VM management in Clouds Gregory Katsaros a,* , Josep Subirats b , J. Oriol Fitó b , Jordi Guitart b , Pierre Gilet c , Daniel Espling d a National Technical University of Athens (NTUA), Greece b Universitat Politecnica de Catalunya (UPC) and Barcelona Supercomputing Center (BSC), Spain c High Performance Computing Center Stuttgart (HLRS), Germany d Umeå University and High Performance Computing Center North (HPC2N), Sweden article info Article history: Received 3 April 2012 Received in revised form 18 September 2012 Accepted 7 December 2012 Available online xxxx Keywords: Monitoring Cloud Energy efficiency Energy consumption abstract The monitoring of QoS parameters in Services Computing as well as in Clouds has been a functionality provided by all contemporary systems. As the optimization of energy consumption becomes a major concern for system designers and administrators, it can be considered as another QoS metric to be monitored. In this paper, we present a service framework that allows us to monitor the energy consumption of a Cloud infrastructure, calculate its energy efficiency, and evaluate the gathered data in order to put in place an effective virtual machine (VM) management. In that context, a simulation scenario of an eco-driven VM placement policy resulted in a 14% improvement of the infrastructure’s energy efficiency. In total, the proposed approaches and implementations have been validated against a testbed, producing very promising results regarding the prospect of energy efficiency as an important quality factor in Clouds. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Datacenters supporting Cloud computing consume an enor- mous amount of power [1], representing a financial burden for their operating organizations, an infrastructure burden on power utilities, and an environmental burden on society. Cloud comput- ing is an emerging paradigm for business computing and as its pop- ularity grows, its potential energy impact grows in significance. This increased usage of Cloud computing, together with the in- creasing energy costs and the need to reduce carbon emissions call for energy-efficient technologies to sustain Cloud datacenters. Whereas large Internet companies (e.g., Google and Microsoft) have significantly improved the energy efficiency of their multi- megawatt datacenters, they have so far focused mostly on hardware aspects. There are still great energy saving potentialities waiting to be turned into reality with respect to system operation. In addition to large datacenters, efficient operation would be extremely helpful for small and medium-sized datacenters, which account for the majority of the energy consumed by datacenters [1] and in general cannot afford expensive hardware improvements to save energy. * Corresponding author. E-mail addresses: gregkats@mail.ntua.gr (G. Katsaros), josep.subirats@bsc.es (J. Subirats), josep.oriol@bsc.es (J.O. Fitó), jordi.guitart@bsc.es (J. Guitart), gilet@hlrs.de (P. Gilet), espling@cs.umu.se (D. Espling). This paper contributes a service framework for Cloud providers to enable energy efficiency awareness. By using the framework, providers can address the energy-efficient management of their resources and running services. The framework offers methods for measuring, analyzing, and evaluating energy use during service de- ployment and operation. It operates a monitoring infrastructure to provide real-time and foreseen status information about services, physical resources, virtual resources, energy consumption, carbon generation, etc. Using these energy consumption measurements, together with the mechanisms for predicting the energy impact based on run-time state, historical usage patterns, and predictions of future demands, the energy and ecological efficiency of a Cloud provider can be assessed. This assessment is used to drive self- management policies in order to fulfill the provider’s energy ef- ficiency requirements. In particular, this paper proposes the usage of energy assessments to optimize the placement of virtual ma- chines (VMs) in a Cloud provider in order to optimize its energy efficiency. The rest of the paper is structured as follows: Section 2 discusses other works related to this approach. Section 3 presents an extended background and motivation behind this work. Section 4 describes the suggested approach for monitoring in detail and Section 5 elaborates on how the collected data is evaluated. Practical use of energy data is presented in Section 6 before an in-depth validation and evaluation of the approach is presented in Section 7. Finally, the paper is concluded in Section 8. 0167-739X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.future.2012.12.006