G-SLAM:OPTIMIZING ENERGY EFFIIENCY IN CLOUD
- 1. G-SLAM:Optimizing Energy Efficiency in Clouds Presented By : Alfiya Mahmood Research Scholar CSE Department Integral University, Lucknow
- 2. WHAT IS CLOUD COMPUTING? Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources like networks, servers, storage, applications, and services.
- 3. IaaS Amazon EC2, GoGrid, RightScale Animoto, Sales Force, Google Document User Applications User-level and infrastructure level Platform Google AppEngine, MapReduce, Aneka, Microsoft Azure Infrastructure SaaS PaaS CloudEconomy 3
- 5. Dark Side of Cloud •IT industry contributes 2% of world's total CO2 emissions • 1.5% of total U.S. power consumption used by data centers which has more than doubled since 2000 and costs $4.5 billion • In India, 14% of ICT emission is caused by data centers • Roughly50%of the emissiondue to data centers isdue to powersystem lossesand cooling loads
- 6. Server/Storage 50% Computer Rm.AC 34% Conversion 7% Network 7% Lighting 2% Source:APC Compute resources and particularly servers areat the heart of a complex, evolving system! Where Does the Power Go? Power Consumption in theDatacenter 7
- 7. Impact on the Environment 7 8/30/2013 • Data centers are not only expensive to maintain, but also unfriendly to the environment. • High energy costs and huge carbon footprints are incurred due to the massive amount of electricity needed to power and cool the numerous servers hosted in these data centers.
- 8. Green Cloud: “performance” “energy efficiency” As energy costs are increasing, there is a need to shift focus from optimising data center resource management for pure performance alone to optimising for energy efficiency while maintaining high service level performance. 8 8/30/2013
- 9. Green Service Level Agreement To fulfill the requirements of green computing, the main need is to maintain a green SLA. It is an agreement same as the basic service level agreement only with the difference of eco-efficient parameters. GSLA consists three Es of sustainability- Ethics, Ecology and Economic.
- 10. PROPOSED WORK In the proposed G-SLAM (Green Service Level Agreement-Minimization of Energy) framework, energy consumed on different levels of cloud services can be optimized with the help of software, hardware and network techniques. The classification of these approaches help us to better understand the various issues in the field of green cloud computing. Fig.4 Classification of Energy Optimization Techniques
- 11. SOFTWARE TECHNIQUES • Software approach of energy reduction can be implemented by reducing the number of active servers. • By using Multiple Correlation Coefficient method, the selection of servers is completed. The correlation coefficient is directly proportional to the risk of SLA violation. • Ant Colony Optimization is another approach for placement of virtual machines. It maximizes the resource utilization by minimizing the number of active servers. Energy reduction in homogenous data centers is achieved by minimizing the number of active servers. • The Power Aware Best Fit Decreasing algorithm allocates a new request for VM which provides the least increase in power consumption caused by allocation. VMs are sorted in decreasing order according to their current processor utilization and then allocate it to the server that provides least increase of power consumption.
- 12. HARDWARE TECHNIQUES • The optimization of consumed energy can be done by utilization of flexible hardware by controlling frequencies and voltage in the server. • Dynamic Voltage Frequency Scaling (DVFS) technique is applied with a number of processors that can operate at different voltage and frequency levels. • Dynamic Voltage Scaling (DVS) is also applied to save energy while testing the ability of task to meet the deadlines. DVS provides a balance between the task deadline and power consumption. • A heuristics workflow algorithm namely Energy Aware Resource Efficient Workflow Scheduling under Deadline constraint (EARES-D) utilizes DVFS technique.
- 13. NETWORK TECHNIQUES The communication of virtual machines consumes energy in data center. Reduction of network traffic reduces the consumption of energy.The algorithm helps in identification of under-loaded servers and heavily-loaded servers. After that it identifies the physical servers with sufficient capacity and arranges them in ascending order according to their load. Data Center Energy Efficient Network Aware Scheduling (DENS) is used to reduce energy consumption by optimization of tradeoff between task consolidation and distribution of traffic pattern. It selects best fit server based on weighted computational function to execute the job. Disjoint Edge Node Divided Spanning Tree with traffic-aware Flow Migration (DENDIST-FM) is applied. It generates various disjoint spanning trees to avoid path overlapping and chooses the least utilized path for routing. Energy and Topology aware VM Migration (ETA-VMM) detects and migrates VMs from under or over utilized servers to the nearest machine. The nearest machine is based on the nearest network distance.
- 14. CONCLUSION In recent years, growing needs of cloud services led to uneconomical energy consumption. Computing resources results in high operational cost as well as huge amount of carbon di oxide emission to the environment. Therefore, cloud services required to be eco-friendly. Researchers are continuously struggling with optimizing energy aspects. This paper class classifies different energy reduction techniques for making green cloud services. We introduced cloud simulators focusing on CloudSim. Our future work consists of implementation and evaluation of energy efficient framework by simulating it on CloudSim toolkit. 14
- 15. REFERENCES 1. NIST, 2011, The NIST Definition of Cloud Computing Recommendations of the National Institute of Standards and Technology. NIST Special Publication, 145, p.7. Available at: http://www.mendeley.com/research/the-nist-definition-about-cloud-computing/.K.Elissa, “Title of paper if known,” unpublished 2. L. Wu, and R. Buyya, “Service Level Agreement (SLA) in Utility Computing Systems, ” Performance and Dependability in Service Computing: Concepts, Techniques and Research Directions, V. Cardellini et. al. (eds), ISBN: 978-1-60-960794-4, IGI Global, Hershey, PA, USA, July 2011, pp.1-25. 3. G. S. Akula and A. Potluri, “Heuristics for migration with consolidation of ensembles of virtual machines,” Proc. Communication Systems and Networks (COMSNETS), 2014 6th Int. Conf., pp. 1, 4, 6–10. 4. S. K. Garg and R. Buyya, “Green Cloud Computing and Environmental Sustainability” . 5. A. Beloglazov, J. Abawajy, and R. Buyya, “Energy-aware resource allocation heuristics for efficient management of data centers for Cloud computing,” Futur. Gener. Comput. Syst., vol. 28, no. 5, pp. 755–768, 2012. 6. Z. Cao, “Energy-aware framework for virtual machine consolidation in Cloud computing,” Int. Conf. High Perform. Comput. Commun., p. 429, 2013. 7. Habibullah, Khan Mohammad, "Developing strategies to mitigate the energy consumed by network infrastructures." (2016).
- 16. 16