International Journal of Computer Applications (0975 – 8887) Volume *– No.*, ___________ 2011
Contributing Solutions and Latest Developments in Green Cloud V.RAJ GOWTHAM
R.S GHAYATHREE
S.R.M. Arts and Science College Kattankulathur
Meenakshi College Of Engineering Chennai
[email protected]
[email protected]
ABSTRACT In recent years, most of the organizations in the world have increasingly realized the importance of the cloud platforms. The growth of mobile applications and cloud is contributing towards excessive usage of energy. Power Consumption in the data center hosting cloud is usually higher with respect to Servers, Storage and Air-Conditioners. The term combines the words green -- meaning environmentally friendly -- and cloud, the traditional symbol for the Internet and the shortened name for a type of service delivery model known as cloud computing. The savings would be primarily achieved by consolidating data centers and maximizing power usage efficiency, improving recycling efforts, lowering carbon and gas emissions and minimizing water usage in cooling the remaining centers. This paper aims to address the possible contributing solutions for Green Cloud Computing such as Consolidation, Virtualization, Reuse, Refurbishing and Energy Usage Profile (EUP) for Hardware, Applications and Operating Systems used in Cloud. The other contributing approaches and latest developments such as Wi-Fi Scheduling, Green-Web Surfing, Carbon Efficient Green Policy, Carbon Aware Green Cloud Architecture, Cloud Broker and Green Broker will be discussed in detail in this paper.
General Terms Cloud Computing, Green Computing
Keywords Cloud Computing, Green Computing, Green Cloud, Green Broker, Carbon Efficient Green Policy, Green middleware
1. INTRODUCTION Gartner Report 2007 states that IT industry contributes 2% of world's total CO2 emissions whereas U.S. EPA Report 2007 mention that 1.5% of total U.S. power consumption used by data centers which has more than doubled since 2000 and costs $4.5 billion. Amazon.com estimates that energy-related costs of its data centres amount to 42% of the total budget that include both direct power consumption and the cooling infrastructure amortized over a 15-year period. Google,
D.VENKATA SUBRAMANIAN B.S.ABDUR RAHMAN UNIVERSITY, CHENNAI
[email protected]
Microsoft, and Yahoo are building large data centres in barren desert land surrounding the Columbia River, USA to exploit cheap hydroelectric power. There is also increasing pressure from Governments worldwide to reduce carbon footprints, which have a significant impact on climate change. Computing resources and particularly servers are at the heart of a complex, evolving system. 2010 has been touted by many in the ICT sector as the „Year of the Cloud‟. While this is likely a prediction that will be repeated in subsequent years, the arrival of the iPad and growth in netbooks and other tablet computers, the launch of Microsoft‟s Azure cloud services for business, and the launch of the Google phone and the proliferation of mobile cloud applications are compelling signs of a movement towards cloud-based computing within the business sector and public consciousness in a way never seen before. More cloud-computing companies are pursuing design and strategies that can reduce the energy consumption of their data centres, primarily as a cost containment measure. For most companies, the environmental benefits of green data design are generally of secondary concern. We are living in the world of Web 2.0 and 3.0, where millions of people are connected to the Internet and social media tool like Facebook, LinkedIn, Twitter, YouTube, Flickr Blogs, etc. Due to this effect, cloud computing plays a vital role. Cloud Computing Services are primarily used by business users and social media users to help them to get connected and also to improve their communication, business operations and productivity. The Organization for Economic Co-operation and Development (OECD) has published a survey of over 90 government and industry initiatives on Green ICTs. This report concludes that initiatives tend to concentrate on the greening ICTs themselves rather than on their actual implementation to tackle global warming and environmental degradation. Green computing is nothing but the study and practice of designing, manufacturing, using, and disposing of computers, servers, and associated subsystems such as monitors, printers, storage devices, and networking and communications systems efficiently and effectively with minimal or no impact on the environment [1]. The term cloud has been used historically as a metaphor for the internet and evolved through a number of phases which include grid, utility computing, application service provision and software as a service (SaaS) as shown in the figure below.
1
International Journal of Computer Applications (0975 – 8887) Volume *– No.*, ___________ 2011 not in use, even if just for an hour or thirty minutes, turning off the monitors when not in use as opposed to running a screen saver, usage of power save mode and use of hardware and software with the energy star label, not printing documents unless necessary and using LCD‟s instead of CRTS as they are more power efficient. Apart from energy consumption of using computers, there are other negatively contributing factors such as pollution due to the chemicals and fuels being used for the manufacturing process for producing the desktop and server class computers and its components. For example, it has been estimated that between 1997 and 2004, 1.2 billion tons of lead was used in computer components and the usage of lead is growing more and more.
Figure 1: Evolution of Cloud Computing [2]
Cloud computing offers three types of responsibilities which are SaaS, PaaS and IaaS. Details of each responsibility is described below.
Software-as-a-Service (SaaS): A wide range of application services delivered via various business models normally available as public offering
Platform-as-a-Service (PaaS): Application Cloud Computing
development
platforms
provides
authoring and runtime environment
Infrastructure-as-a-Service (IaaS): Also known as elastic compute clouds, enable virtual hardware for various use
Cloud Computers and Server Class Computers consume electrical current to run the CPU(s), motherboard, memory, running the fan and spinning the disk(s). Power Consumption in the data center hosting cloud would be distributed as 50% for Server/Storage, 34% for Computer Room Air-conditioners 34%, 7% for Conversion, 7% for Network and remaining 2% for Lighting. Apart from the centralized and/or shared systems, the end user monitors (CRTs consume more power than any other computer component) and printers. Apart from this, the network landscape such as switches and controllers do consume electrical energy. Electrical energy can be reduced by taking appropriate measures such as turning off the computers and servers when
Lead can cause damage to the central and peripheral nervous systems, blood system, kidneys, endocrine system and cause negative effects on child brain development and lead accumulates in the environment and has toxic effects on plants, animals and microorganisms. Apart from this, the electronics contribute 40% of the total amount of lead found in landfills and can make its way from landfills into the water supplies. Mercury is used in batteries, switches, housing, printed circuit boards and also used in data transmission equipment, telecommunications equipment which are the key resources for cloud computing and cell phones as well. Although a small amount of mercury is used, it is used in nearly all computer construction amounting to 400,000 pounds of mercury used between 1997 and 2004. Apart from mercury there are other chemical elements such as Cadmium, Chromium and Plastics and the list continues. Considering that the average computer lifespan (2 years), disposal of the computers and computer components such as hard disks and motherboards constituted 20-50 million tons per year (about 5% of the total waste of the planet) and this waste is often called as e-Waste. The e-Waste causes the pollution by spreading Toxic chemicals in the food chain and water. The growth of mobile applications and cloud is also contributing towards excessive usage of Computer Energy in the world. So, it is important to analyze the possible contributing solutions which can reduce the power consumption and effective utilization of computers in the green cloud. Ultimately, if cloud providers want to provide a truly green and renewable cloud, they must use their power and influence to not only drive investments near renewable energy sources, but also become involved in setting the policies that will drive rapid deployment of renewable electricity generation economy-wide, and place greater R&D into storage devices that will deliver electricity from renewable sources 24/7. This paper aims to address the possible contributing solutions for Green Cloud Computing such as Consolidation, Virtualization, Reuse, Refurbishing, Recycling and Energy Usage Profile (EUP) for Hardware, Applications and Operating Systems used in Cloud. The other contributing approaches such as Power Management Policies, Procedures, Wi-Fi Scheduling, Green web surfing will be further discussed in this paper. Secondly this paper, address some of the latest developments in Green Cloud such as Carbon Efficient Green Policy, Carbon Aware Green Cloud Architecture, Cloud Broker, Green Broker and Green Middleware.
2
International Journal of Computer Applications (0975 – 8887) Volume *– No.*, ___________ 2011
2. POSSIBLE SOLUTIONS There are four possible paths identified such as Green use, green disposal, green design, and green manufacturing [1]. This paper suggests some additional solutions which can directly and indirectly contribute towards green cloud. These solutions are listed in the below sections.
2.1 Consolidation Most of the IT companies are looking at consolidation effort in order to centralize the IT operations and support. The consolidation is indeed an effective approach to not only maximize the resource utilization and also to minimize the energy consumption in a cloud computing environment. The live virtual machine migration technology is one of the preferable ways to consolidate the machines. The consolidation of servers in a data centre should not affect the performance of users and applications. The server resources such as memory cache, disk I/O should be shared among virtual machines or applications of the cloud in an optimal way. While there are multiple consolidation techniques available, heterogeneous workload consolidation methods found to be more reliable in green cloud. Some of the techniques proposed by Carrea et al[3], Zhan et al[4], Steinder et al[5[ and Goiri et al[6] seems to be appropriate and can be considered for consolidation in cloud.
2.2 Virtualization and Reuse Virtualization is a technology that allows running two or more operating systems side-by-side on just one PC or embedded controller [7]. It is rapidly adopted in the engineering world. It helps them to better utilize their available processing hardware to build more efficient systems. As multi-core processors with 4, 8, and 16 cores on a chip become a common place, many processor cores are likely to be underutilized in a typical system. Most applications will have only a finite amount of parallel tasks that can be executed at a given time, leaving many processor cores idle. Virtualization can solve this challenge by allocating group of processor cores to individual operating systems running in parallel. Thus benefits of virtualization can be enumerated as: (i) Save hardware cost and footprint. (ii)Take advantage of Operating System services. (iii)Make use of Multicore Procesors.
More importantly, the energy consumption requirement for multiple machines or operating systems is reduced due to virtualization. For example, if an application has to be tested on different type of operating systems such as Windows 7, Linux, Windows XP, instead of having three different computers and consuming energy resources, perhaps, one single machine can run three different virtual machines (VMs) with different operating systems. Most of the systems especially servers are dedicated for a particular client or application.
The dedicated servers are only used when there are active connections. The servers can be used for other purposes during their idle or inactive time. or use. The systems should not be idle and can be reused for other application execution. Though there are few drawbacks of sharing the same server for multiple applications or users, there are lot of significant advantages in terms of energy consumption and optimal resource utilization. The reuse principle can be applied to multinational companies, companies which work across cities and countries and work in different time zones.
2.3 Refurbishing The main difference between "refurbished" and "used" products is that refurbished products have been tested and verified to function properly, and are thus free of defects, while "used" products may or may not be defective. Refurbished products may be unused customer returns that are essentially "new" items, or they may be defective products that were returned under warranty, and resold by the manufacturer after repairing the defects and ensuring proper function. The refurbished servers or network equipments could cost not only less money and also utilize the refurbished servers more optimally to reduce the e-Waste.
2.4 Power Management Policies R Nathuji et al. [8] have explored the problem of power efficient resource management and pointed out the number of benefits of virtualization such as: improved fault and performance isolation between applications sharing the same resource; ability to relatively easy move of VMs between physical hosts applying live or offline migration; support for hardware and software heterogeneity. The experimental results show that reduction in power consumption up to 17% by exploiting power management heterogeneity. For every data centre utilizing cloud should consider design and development of power management policies which should be followed. It is clear that as the energy demand of the cloud grows, the supply of renewable energy must also keep pace. Additionally, because of the unique opportunities provided to the ICT sector in a carbon constrained world, the industry as a whole should be advocating for strong policies that result in economy-wide emissions reductions. Among prime concern is priority grid access for renewable sources of energy. Rules on grid access, transmission and cost sharing are very often inadequate. Legislation must be clear, especially concerning cost distribution and transmission fees. Where necessary, grid extension or reinforcement costs should be borne by the grid operators, and shared between all consumers, because the environmental benefits of renewable are a public good and system operation is a natural monopoly. Daily carbon savings is shown in the chart found in Figure 2 (source: Google).
3
International Journal of Computer Applications (0975 – 8887) Volume *– No.*, ___________ 2011 As represented in figure 3, Carbon Emission Directory is one of the important components which contain data on Power Usage Effectiveness (PUE), cooling efficiency, carbon footprint, network cost. This helps help user to select cloud services with minimum carbon footprint. Green Offer directory is incentive for the users and also choosing carbon efficient hours, lists services with their discounted prices and green hours. A typical cloud broker lease cloud services and schedule applications. Public Cloud is the most common deployment model where services are available to anyone on Internet in a pay-as-yougo manner. To support thousands of public domain users, data centres built by public Cloud providers are quite large comprising of thousands of servers with high speed network. Private Clouds are deployed within the premise of an organization to provide IT services to its internal users. The private Cloud services offer greater control over the infrastructure, improving security and service resilience because its access is restricted to one or few organizations. Figure 2: Daily Carbon Savings
3. LATEST DEVELOPMENTS This section enlist some of the latest developments which contributes towards green cloud computing. One of the latest development took place in the green cloud computing is the development of carbon aware green cloud architecture and green cloud broker. The details are discussed in the subsections below.
Hybrid Clouds is the deployment which emerged due to diffusion of both public and private Clouds advantages. In this model, organizations outsource non-critical information and processing to the public Cloud, while keeping critical services and data in their control. Therefore, organizations can utilize their existing IT infrastructure for maintaining sensitive information within the premises. The figure 4 contains the layered architecture of Green Broker used in the green cloud architecture found in figure 2. The first layer helps to analyze user requirements. The second layer aims to calculate cost and carbon foot print of services. The third layer aims to perform carbon aware scheduling.
3.1 Carbon Aware Green Cloud Architecture
Figure 4: Green Cloud Broker
Figure 3: Carbon Aware Green Cloud Architecture
Green cloud broker has similar responsibility like a cloud broker. It helps to lease cloud services on behalf of users and schedule their applications. The cloud request services first analyze the requests and their quality of service (QoS) requirements, whereas the second layer calculates the cost and
4
International Journal of Computer Applications (0975 – 8887) Volume *– No.*, ___________ 2011 carbon footprint of leasing particular cloud services. This type of information can be gathered based on information about various cloud offerings and current CO2 emission factors obtained from Green Offer Directory and Carbon Emission Directory respectively. The green policies make the decisions of leasing cloud services. If no exact match is found for a request, alternate green offers are suggested to users by cloud request services.
summarized some of the useful solutions which can contribute to optimally utilize the cloud computing environment to save the energy resources and contribute towards green ICT.
3.2 Green Wi-Fi Scheduling
[2] D. C. Wyld, Moving to the Cloud: An Introduction to Cloud Computing in Government,‖ IBM center for The Business of Government e- Government Series, 2009.
The ability to turn wifi on / off on a specific schedule is one of the great service to have with the devices. It would be better, if there are ways to set a different schedule for weekdays and weekends in order to save power. The Green Wi-Fi Routers achieve optimal energy savings when used with the Wi-Fi Scheduler, which provides a user-selectable radio shutdown option. Under the most favorable conditions with no wired links active and Wi-Fi turned off, users may achieve considerable power savings when compared to a traditional conventional router without Green technology.
REFERENCES [1] San Murugesan, “Harnessing Green IT: Principles and Practices,” IEEE IT Professional, January–February 2008, pp 24-33.
[3] D. Carrera, M. Steinder, I. Whalley, J. Torres and E. Ayguade, Managing SLAs of heterogeneous workloads using dynamic application placement,‖ HPDC '08, Proceedings of the 17th international symposium on High performance distributed computing, New York, USA, 2008. [4] J. Zhan, L. Wang, B. Tu, Y. Li, P. Wang, W. Zhou and D. Meng, Phoenix Cloud: Consolidating Different Computing Loads on Shared Cluster System for Large Organization,‖ Proceeding Workshop Cloud Computing and Its Application (CCA „08), 2008. [5]
3.3 Carbon Efficient Green Policy While there are many initiatives and quality policies were set towards green computing, there is no emphasis on carbon efficient green policies. Carbon Efficient Green Policy (CEGP) for Green Broker is aimed to periodically select the Cloud provider with the minimum carbon footprint and initiate Virtual Machines to run the jobs. Some of the possible policy statements which support the green computing efforts are listed below:
Collect resource requests from user and resource
Site information such as VMs, carbon emission rate, CPU power efficiency
Sort jobs based on deadline
Sort resource sites based on carbon footprint
Schedule greedily the most urgent deadline jobs on the most power efficient resource site
4. CONCLUSION Data centres are not only expensive to maintain, but also unfriendly to the environment due to the carbon emission. 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 centres. Green computing is nowadays a major challenge for most IT organizations that involve medium and large scale distributed infrastructures like Grids, Clouds and Clusters. Irrespective of different cloud computing and green cloud computing architecture models, there exists a need to design and develop solutions. The carbon aware green cloud architecture models are highly recommended for future cloud based solutions. Carbon Efficient Green Policy (CEGP) can save up to 10 to 25% energy while improving the carbon footprint by about 25%, so there should be some level of emphasis on establishing and maintaining the CEGP. This paper
M. Steinder, I. Whalley, D. Carrera, I. Gaweda and D. M. Chess, Server virtualization in autonomic management of heterogeneous workloads,Integrated Network Management, page no. 139-148, 2007.
[6] I. Goiri, J. Fito, F. Julia, R. Nou, J. L. Berral and J. Guitart, Multifaceted resource management for dealing with heterogeneous workloads in virtualized data centers, In 11th ACM/IEEE international conference on grid computing (Grid 2010), Brussels, Belgium, page no. 25– 32, Oct 2010. [7] Aasys, Virtualization Basics, Vol. 6, Issue 9, September 2008. [8] R. Nathuji and K. Schwan, ―Virtualpower: Coordinated power management in virtualized enterprise systems, ACM SIGOPS Operating Systems Review, Vol. 41, Issue no. 6, page no. 265–278, 2007 [9] M. Armbrust, A. Fox, R. Griffith, A. D. Joseph, R. H. Katz, A. Konwinski, G.Lee, D. A. Patterson, A. Rabkin, I. Stoica and M. Zaharia, Above the Clouds: Berkeley View of Cloud Computing‖, University of California at Berkeley, Electrical Engineering and Computer Sciences, Technical Report No.UCB/EECS-2009-28, February 10, 2009. [10] S. Zhang, S. Zhang, X. Chen and X. Huo, ―The Comparison Between Cloud Computing and Grid Computing‖, International Conference on Computer Application and System Modeling, 2010. [11] R. Buyya, C. Yeo, and S. Venugopal, ―Market-oriented cloud computing: Vision, hype, and reality for delivering it services as computing utilities‖, in Proceedings of the 10th IEEE International Conference on High Performance Computing and Communications (HPCCOB, IEEE CS Press. Los Alamitos, CA. USA), 2008. [12] P. Mell, and T. Grance, ―The NIST Definition of Cloud computing,‖ National Institute of Standards and Technology, 2009. [13] V. R. Srinivasa, N. K. R. Nageswara and E. K. Kumari, Cloud Computing: An Overview,‖ Journal of Theoretical and Applied Information Technology, 2009.
5
International Journal of Computer Applications (0975 – 8887) Volume *– No.*, ___________ 2011 [14] D. C. Wyld, ―Moving to the Cloud: An Introduction to Cloud Computing in Government,‖ IBM center for The Business of Government e- Government Series, 2009. [15] K. Jeffery, B. Neidecker-Lutz, ―The Future of Cloud Computing. Opportunities for European Cloud Computing Beyond 2010‖, Expert Group Report, European Commission, January 2010. [16] T. Dillon, C. Wu, and E. Chang, ―Cloud Computing: Issues and Challenges‖, in 24th IEEE International Conference on Advanced Information Networking and Applications, page no. 27-33, 2010. [17] V. K. Reddy, B. T. Rao, L. S. S. Reddy and P.S. Kiran, Research issues in Cloud Computing, Global Journal Computer Science & Technology Vol. 11, Issue no. 11, page no. 70-76, June 2011. [18] Q. Zhang, L. Cheng and R. Boutaba, ―Cloud Computing: state-of-the-art and research challenges‖, Journal of Internet Services and Applications, Vol. 1, Issue no. 1, page no. 7-18, Springer, 2010. [19] M. Weiser. Ubiquitous computing. IEEE Hot Topics, 26(10):71--72, 1993. [20] J. Bacon, K. Moody, J. Bates, R. Hayton, C. Ma, A. McNeil, O. Seidel, and M. Spiteri. Generic support for distributed applications, IEEE Computer, 33(3):68--76, 2000. [21] Blair, G.S., Campbell, A.J., Schmidt, D.C., "Middleware Technologies for Future Communication Networks", IEEE Network, Vol. 18, No. 1, January 2004. [22] C. Mascolo, L. Capra, Emmerich,w. "Middleware for Mobile Computing (A Survey)". In Advanced Lectures on Networking - Networking 2002 Tutorials, Pisa, Italy. Volume 2497 of LNCS, pages 20-58. [23] A. Carzaniga, D.S. Rosenblum, and A.L. Wolf "Achieving Expressiveness and Scalability in an InternetScale Event Notification Service". Nineteenth ACM Symposium on [24] Principles of Distributed Computing (PODC2000), Portland OR. July, 2000
6