Energy Efficient Load Management of Data Center for Green Cloud computing: A Survey

IJIRST –International Journal for Innovative Research in Science & Technology| Volume 3 | Issue 03 | August 2016 ISSN (online): 2349-6010

Energy Efficient Load Management of Data Center for Green Cloud Computing: A Survey Prerna Kesarwani M. Tech. Scholar Department of Computer Sciences & Engineering JSS Academy of Technical Education, Noida ,India

Mahboob Alam Assistant Professor Department of Computer Sciences & Engineering JSS Academy of Technical Education, Noida ,India

Abstract Cloud computing has changed the way in which people use the IT resources today. Now, instead of buying their own IT resources, they can use the services offered by Cloud Computing with reasonable costs based on a “pay-per-use” model. Cloud computing is an accumulation community based solutions offering you with on-demand, ubiquitous, convenient system access to a shared pool of configurable computing resources that may be rapidly provisioned and released in an easy and universal means with reduced management effort or Cloud Service Provider interaction. It offers solution that is new with additional resource that is efficient, on-demand scalability, and price decrease for organizations. Cloud computing has quickly evolved through various phases that consist of grid and utility computing and application service provisioning. These solutions are scaled along with regards to the client’s adjustable operation’s needs, ensuring maximum price efficiency. Cloud computing can enable more use that is energy-efficient of energy, especially whenever the computing tasks are of low infrequent or intensity. In this work we gave the architecture for green cloud the main advantageous asset of this architecture is emission that is co2, this directory steps the best solution that is suitable gives less carbon emission so instantly it indicates that energy will also decrease because Co2 emission and energy consumption both are straight proportionate to at least one another. Likewise, the disadvantage is that only emission that is energy that is CO2 isn't the factor become under consideration like Quality Provisioning, Security, etc.

Keywords: Green Cloud Computing, Energy Conservation, Dynamic Voltage Frequency Scaling _______________________________________________________________________________________________________ I.

INTRODUCTION

Cloud Computing Could computing is a new type of computing mode, It distributes computation task on the resource pool which consists of massive computers, so the application systems can obtain the computation power, the storage space and software service according to its demand. This kind of resource pool is called “cloud”. The Clouds are some virtual computation resources which can be maintained and managed by themselves, usually they are some large-scale server clusters, including calculating server, storage server, the broadband resources and so on. Cloud computing can concentrate all the computing resources and manage them automatically through the software without intervene. This makes application offers not to annoy for tedious details and more absorbed in his business. It will be advantageous to innovation and reduce cost. This is favorable to innovation and cost reduction. It’s the ultimate goal of cloud computing to provide calculation, services and applications as a public facility for the public, So that people can use the computer resources like using water, electricity, gas and telephone.

Fig. 1: Could Computing

Service Forms of Cloud Computing Software as a Service (SaaS) SaaS provider dispose the applied software unified on their server, the user can subscribe applied software service from the manufacturer through Internet .The Provider supply software pattern through Browser, and charge according to the quantity of

All rights reserved by www.ijirst.org

117

Energy Efficient Load Management of Data Center for Green Cloud Computing: A Survey (IJIRST/ Volume 3 / Issue 03/ 021)

software and using time. The advantage of this kind of service pattern is that the provider maintains and manages software, supplies the hardware facilities, the users can use software everywhere when they own the terminal which can log in Internet.

Platform as a Service (PaaS) PaaS takes develop environment as a service to supply. It is a kind of distribution platform server, the manufacturers supply service to the users, such as develop environment, server platform and hardware resources, and the users customize and develop their own application and transfer to other customers through their server and Internet. PaaS can provide the middleware platform, application development, database, application server and experiment for the enterprise and the individual.

Infrastructure as a Service (IaaS) IaaS takes infrastructure which is made of many servers as a measurement service to the customers. It integrates memory and I/O devices, storage and computing ability into a virtual resources pool, and provides storage resources and virtualization service for the whole industry. This is a way of hosted hardware, and the customer pays when they use the hardware. For example, Amazon Web Service and IBM Blue Cloud all rent the infrastructure as a service. The advantage of IaaS is that the user only need low cost hardware and rent computing ability and storage ability according to his need, greatly reduced cost of the hardware.

Techniques Many researchers have investigated new ways for managing the Cloud infrastructure as a means of enhancing the energy efficiency. A number of techniques have been already proposed and deployed for better resource management. For example, Data Voltage and Frequency Scaling (DVFS) and Virtual Machines (VMs) allocation have been widely studied and deployed to manage the resources more efficiently.

Dynamic voltage-frequency scaling (DVFS) is perhaps the most appealing method incorporated into many recent processors.

Energy saving with this method is based on the fact that the power consumption in CMOS circuits has a direct relationship with frequency and the square of voltage supply. Thus, the execution time and power consumption can be controlled by switching between a processor's frequencies and voltages. Although this approach was initially designed for single processor task scheduling, it has recently received much attention in multiprocessor systems as well.

DVFS technique and task scheduling can be combined in two ways:  Schedule generation.  Slack time reclamation.

Virtualization Cloud computing takes virtualization one step further:  You don’t need to own the hardware  Resources are rented as needed from a cloud  Various providers allow creating virtual servers:  Choose the OS and software each instance will have  The chosen OS will run on a large server farm  Can instantiate more virtual servers or shut down existing ones within minutes  You get billed only for what you used.

II. LITERATURE SURVEY This section, presents review of the selected literature in Energy aware profiling for cloud computing environments .The key objective is to highlight key strengths and limitations of various cloud computing approaches.

Ibrahim Alzamil et al [2015] elaborates Profiling tools that would help understand how the energy has been consumed in a system is essential in order to facilitate software developers to make energy-aware programming decisions. Schubert et al state that the developers lack the tools that indicate where the energy-hungry sections are located in their code and help them better optimize their code for enhancing energy consumption more accurately. We propose in this paper to have energy-aware profiling for the Cloud infrastructure to better understand how the energy has been consumed and assess its energy efficiency in order to help the software developers from the application layer enhance their decision-making in terms of energy-awareness when optimising their applications and services Muhammad H. Raza et al.[2015] highlight the application of cloud computing and the Internet of Things on the medical field.

The model architecture for remote monitoring cloud platform of healthcare information (RMCPHI) was established firstly. Then the RMCPHI architecture was analyzed. There is an increasing demand of new algorithm for resource management and real time scheduling in order to meet the ever increasing demand from users. So, Finally an efficient PSOSAA algorithm was proposed for the medical monitoring and managing application of cloud computing. Simulation results showed that our proposed scheme can improve the efficiency about 50%.

Andrew Carlina et al.[2015] The head of IT departments may notice that Cloud computing could be cause to reduce his functions at work. This creates major factor behind the slower growth of Cloud computing was the controversial view of the fear of job loss in the minds of IT workforce. Based on face to face interview and it surveys it is established that many in the industry did not still understand dynamism of Cloud computing and a need to educate or promote awareness of Cloud computing was identified.

All rights reserved by www.ijirst.org

118

Energy Efficient Load Management of Data Center for Green Cloud Computing: A Survey (IJIRST/ Volume 3 / Issue 03/ 021)

Dimil Josea et al.[2015] highlights Cloud computing offers user high-end and scalable infrastructure at an affordable cost.

Virtualization is the key to unlocking cloud computing. Although virtualization has great benefits to the users, the complexity in its structure, introduces unseen and forcible threats to the security of the data and to the system infrastructure. This investigates the exploitation of compromised virtual machines to execute large-scale Distributed Denial-of-Service (DDoS) attacks. A review of most recent intrusion detection and prevention systems to mitigate potential DDoS attacks is presented.

Syed Rizvi et al.[2015]A prototype model for Intelligent Vehicle Monitoring using Global Positioning System and Cloud Computing has been developed. The proposed tracking system is based on cloud computing infrastructure along with sensors useful for monitoring the fuel level, altitude, tire pressure, driver conditions, and speed of the vehicle. All the data is transferred to cloud server using GSM enabled device. All the vehicles are equipped with GPS antenna for pinpointing the location. To avoid the drunken driving, the alcohol sensors are installed to monitor the driver status. The proposed technology significantly avoids the accident in highways.

Shiliang Luo et al. [2015] employed Cognitive radio network (CRN) has become more prevalent due to the introduction of the IEEE 802.22 standard which aims to promote the sharing of the unused spectrum among primary and secondary users. Jammers still remain a threat to the realization of a CR built on a WSN. The near real-time processing requirements and external environment issues are some of the challenges of a CRCN. However, the scalability and flexibility characteristics of the cloud could mitigate the energy supply problems associated with a CRN. In addition, the resource pooling aspect of the cloud could provide the additional antennas and CR devices needed to efficiently increase the network performance. Therefore, the cloud could be the solution or one of the solutions to the logistic problems facing the realization of a true CRN. K. Djemame et al.[2014] highlights Cloud computing allows enterprises to get their applications up and running faster and enables IT to more rapidly adjust resources to meet fluctuating and unpredictable business demand. Amazon web services, Proposal of cloud computing is tightly coupled with low cost, time. Model–view–controller (MVC) and Domain driven design (DDD) are main focus of interest. MVC is a software architectural pattern mostly for implementing user interfaces. Whereas DDD is an approach to software development for complex needs by connecting the implementation to an evolving model. VM Monitoring UI is consist of views that are shown to user it also includes graph showing data points of instance which is configured by cloud service consumer in this web application.

Asgarali Bouyer [2014] states E-learning is a concept that integrates information technology in teaching and learning. Cloud computing is reducing the difference between on campus education and distance education. Fortunately cloud computing is the technology which can offer different services in three layers. Cloud computing enable students to access the knowledge by sharing distributed E-learning resources in a public, private or hybrid cloud systems. Due to using cloud computing system for deploying modern educational systems must take to account various items such as cost and accelerate delivery of learning service and privacy issue. Therefore, cloud service providers should especially attended to offering cloud-based learning for improving education status in poor countries in Asia and Africa.

Shufen Zhang [2012] Cloud computing distributes computation task on the resource pool which consists of massive computers, so the application systems can obtain the computing power, the storage space and software service according to its demand. The data storage system of cloud computing are Google File System (GFS) and Hadoop Distributed File System (HDFS) which is developed Hadoop team. Cloud adopts Map Reduce programming model, which decomposes the task into multiple subtasks. This paper introduces the definition of could computing and its main service patterns and focused on the key technologies such as the data storage, data management and programming model.

Shekhar Srikantaiah highlights Consolidation of applications in cloud computing environments presents a significant opportunity for energy optimization. The goal of energy aware consolidation is to keep servers well utilized such that the idle power cost are efficiently amortized but without taking energy penalty due to internal interventions. Main aim of designing the algorithm is to minimize energy subject to a performance constraint .It showed how the performance, resource utilization, and energy vary as multiple applications are consolidated on a common server with varying workload serviced.

Reviewing all these papers, one thing can be estimated that the cloud computing environment should be more energy efficient and help software developers understand how their applications are using the infrastructure resources and consuming energy in order to enhance their decision-making when creating and configuring new applications. investigation on energy efficiency modelling to identify new metrics and form KPIs to better understand to what extent a running application is energy efficient in relation to these KPIs and provide a meaningful feedback of these KPIs to the application developers to enhance their programming decisions with energy-awareness.

III. COMPARISON OF VARIOUS SIMULATION TECHNIQUES

Parameters Communication Network Graphical Support Availability Platform Application models

Table – 1 Comparison of various Simulation Techniques MDCSIM CLOUDSIM GREEN CLOUD Limited Limited Full NONE Limited(Cloud Analyst) Limited(Network Animator) Commercial Open Source Open source CSIM Sim Java NS2 Computation Computation and data center Computation data transfer and EXEC .deadline

All rights reserved by www.ijirst.org

119

Energy Efficient Load Management of Data Center for Green Cloud Computing: A Survey (IJIRST/ Volume 3 / Issue 03/ 021)

Simulation Time Language/ script Physical Models Energy models Support Of TCP/IP Power saving model

Seconds C++/ Java NONE Rough NONE NONE

Seconds Java NONE None NONE NONE

Tens of miles C++/ TCL Available using plug in Precise(servers + networks) FULL DVFS,DNS and both

IV. CONCLUSION Accurate vision of the upcoming workload permits the datacenter operator to locale unneeded physical servers in a low-power state to save energy. If extra arrangement capacity is needed, servers in a low-power state are transitioned back to an alert state.

In this paper, we spread our prior research by giving a new way to ascertaining the frequency of computing the forecast of the anticipated capacity. We present a Dynamic forecast quantization method to ascertain the optimal number of forecast calculation intervals. These new algorithms permit us to forecast upcoming burden inside needed Service Level Accords as minimizing the number of periods the forecast calculations have to be performed. We in the end examine this ideal by simulating the stochastic period skyline and Dynamic quantization algorithms and difference the aftermath alongside three contesting methods. The main goal of this research work was to join the Dynamic Voltage Frequency Scaling method in an effectual method to maximize the resource utilization and minimize power consumption of the data center. Minimizing Power can consequence in the less number of carbon impressions and therefore will aid extra in accomplished Green Computing. Our Aftermath display that the DVFS established Cloud Calculating we have industrialized can produce extra than 40-53% effectual aftermath above virtualized cloud arrangement.

REFERENCES [1] [2] [3] [4] [5] [6] [7]

[8] [9] [10] [11] [12] [13] [14] [15] [16]

Ibrahim Alzamil, Karim Djemame, Django Armstrong and Richard Kavanagh, “Energy-Aware Profiling for Cloud Computing Environments”,Electronic Notes in Theoretical Computer Science 318 91-108 (science Direct), 2015.

Muhammad H. Raza, Adewale Femi Adenola, Ali Nafarieh, William Robertson “The Slow Adoption of Cloud Computing and IT Workforce” Procedia Computer Science 52 1114 – 1119 Science –direct, 2015 Andrew Carlina , Mohammad Hammoudehb , Omar Aldabbas. “Defence for Distributed Denial of Service Attacks in Cloud Computing” Procedia Computer Science 73 490 – 497 Science direct,2015 Dimil Josea, Sanath Prasadb, V. G. Sridhar, “Vehicle Monitoring using global positioning system and cloud computing” Procedia Computer science 50440 – 446 Science –direct,2015 Syed Rizvi, Nathan Showan, John Mitchell “Analyzing the Integration of Cognitive Radio and Cloud Computing for Secure Networking”,Procedia Computer Science 61 206 – 212 Science –direct, 2015 Shiliang Luo, Bin Ren. “The Monitoring and Managing Application of Cloud Computing Based on Internet of Things” Computer Methods and Programs in Biomedicine, http://dx.doi.org/10.1016/j.cmpb.2016.03.024 2015.

K. Djemame, D. Armstrong, R. E. Kavanagh, A. J. Ferrer, D. G. Perez, D. R. Antona, J.-C. Deprez, C. Ponsard, D. Ortiz, M,”Energy efficiency embedded service lifecycle: Towards an energy efficient cloud computing architecture.” In the Proceedings of the Workshop on Energy Efficient Systems”(EES’2014) at ICT4S, pages 1–6, Stockholm, Sweden, August ,2014.

R. Latif, et al., "Distributed Denial of Service (DDoS) Attack in Cloud- Assisted Wireless Body Area Networks: A Systematic Literature Review," J. Med. Syst., vol. 38, pp. 1-10, 2014 H. Wang, et al., "A moving target DDoS defense mechanism," Computer Communications, vol. 46, pp. 1021, 2014.

A.S. Gulbhile, M.P. Patil, P.P. Pawar, S.V. Mahajan, S. Barve, “Secure Radio Resource Management In Cloud Computing Based Cognitive Radio Network,” International Journal of Scientific and Technology Research, vol.3, no.5, pp. 136-139, May ,2014 C. Wilke, S. Gotz, and S. Richly. JouleUnit”A Generic Framework for Software Energy Profiling and Testing”In Proceedings of the 2013 Workshop on Green in/by Software Engineering, GIBSE ’13, pages 9–14, New York, NY, USA, ACM ,2013.

W. Chengjian, L. Xiang, Y. Yang, F. Ni, and Y. Mu”System Power Model and Virtual Machine Power Metering for Cloud Computing Pricing. In Intelligent System Design and Engineering Applications (ISDEA)”, 2013 Third International Conference on, pages 1379–1382,2013 A. Beloglazov, J. Abawajy, and R. Buyya”Energy-aware resource allocation heuristics for efficient management of data centers for Cloud computing.

Future Generation Computer Systems,”28(5):755– 768, May 2013.

G. Baldini, T. Sturman, A.R. Biswas, R. Leschhorn, G. Godor, M. Street, "Security Aspects in Software Defined Radio and Cognitive Radio Networks: A Survey and A Way Ahead," IEEE Communications Surveys & Tutorials, vol.14, no.2, pp. 355-379, Second Quarter,2012 Shen, Zhiming, Sethuraman Subbiah, Xiaohui Gu, and John Wilkes. "Cloudscale: elastic resource scaling for multi-tenant cloud systems." In Proceedings of the 2nd ACM Symposium on Cloud Computing, p. 5. ACM, 2011.

Hulkury, Mohammad Naiim, and Mohammad Razvi Doomun. "Integrated green cloud computing architecture." In Advanced Computer Science Applications and Technologies (ACSAT), 2012 International Conference on, pp. 269-274. IEEE, 2012.

All rights reserved by www.ijirst.org

120