Cloud computing and emerging IT platforms: Vision, hype, and reality for delivering computing as the 5th utility
Introduction
Computing is being transformed to a model consisting of services that are commoditized and delivered in a manner similar to traditional utilities such as water, electricity, gas, and telephony. In such a model, users access services based on their requirements without regard to where the services are hosted or how they are delivered. Several computing paradigms have promised to deliver this utility computing vision and these include cluster computing, Grid computing, and more recently Cloud computing. The latter term denotes the infrastructure as a “Cloud” from which businesses and users are able to access applications from anywhere in the world on demand. Thus, the computing world is rapidly transforming towards developing software for millions to consume as a service, rather than to run on their individual computers.
At present, it is common to access content across the Internet independently without reference to the underlying hosting infrastructure. This infrastructure consists of data centers that are monitored and maintained around the clock by content providers. Cloud computing is an extension of this paradigm wherein the capabilities of business applications are exposed as sophisticated services that can be accessed over a network. Cloud service providers are incentivized by the profits to be made by charging consumers for accessing these services. Consumers, such as enterprises, are attracted by the opportunity for reducing or eliminating costs associated with “in-house” provision of these services. However, since cloud applications may be crucial to the core business operations of the consumers, it is essential that the consumers have guarantees from providers on service delivery. Typically, these are provided through Service Level Agreements (SLAs) brokered between the providers and consumers.
Providers such as Amazon, Google, Salesforce, IBM, Microsoft, and Sun Microsystems have begun to establish new data centers for hosting Cloud computing applications in various locations around the world to provide redundancy and ensure reliability in case of site failures. Since user requirements for cloud services are varied, service providers have to ensure that they can be flexible in their service delivery while keeping the users isolated from the underlying infrastructure. Recent advances in microprocessor technology and software have led to the increasing ability of commodity hardware to run applications within Virtual Machines (VMs) efficiently. VMs allow both the isolation of applications from the underlying hardware and other VMs, and the customization of the platform to suit the needs of the end-user. Providers can expose applications running within VMs, or provide access to VMs themselves as a service (e.g. Amazon Elastic Compute Cloud) thereby allowing consumers to install their own applications. While convenient, the use of VMs gives rise to further challenges such as the intelligent allocation of physical resources for managing competing resource demands of the users.
In addition, enterprise service consumers with global operations require faster response time, and thus save time by distributing workload requests to multiple Clouds in various locations at the same time. This creates the need for establishing a computing atmosphere for dynamically interconnecting and provisioning Clouds from multiple domains within and across enterprises. There are many challenges involved in creating such Clouds and Cloud interconnections.
Therefore, this paper discusses the current trends in the space of Cloud computing and presents candidates for future enhancements of this technology. This paper is primarily divided into two parts. The first part examines current research issues and developments by:
- •
presenting the 21st century vision of computing and describing various computing paradigms that have promised or are promising to deliver this grand vision (Section 2),
- •
differentiating Cloud computing from two other widely explored computing paradigms: Cluster computing and Grid computing (Section 3),
- •
focusing on VM-centric Cloud services and presenting an architecture for creating market-oriented Clouds using VMs (Section 4),
- •
providing insights on market-based resource management strategies that encompass both customer-driven service management and computational risk management to sustain SLA-oriented resource allocation (Section 5),
- •
revealing our early thoughts on interconnecting Clouds for dynamically creating global Cloud exchanges and markets (Section 6), and
- •
comparing some representative Cloud platforms, especially those developed in industries along with our Aneka enterprise Cloud technology (Section 7).
The second part introduces our current work on Cloud computing which include:
- •
realizing market-oriented resource allocation of Clouds as realized in Aneka enterprise Cloud technology and highlighting the difference between High Performance Computing (HPC) workload and Internet-based services workload (Section 8),
- •
incorporating a meta-negotiation infrastructure for QoS management to establish global Cloud exchanges and markets (Section 9), and
- •
creating 3rd party cloud services based on high performance content delivery over commercial cloud storage services (Section 10).
Section snippets
The 21st century vision of computing
With the advancement of modern society, basic essential services (utilities) are commonly provided such that everyone can easily obtain access to them. Today, utility services, such as water, electricity, gas, and telephony are deemed necessary for fulfilling daily life routines. These utility services are accessed so frequently that they need to be available whenever the consumer requires them at any time. Consumers are then able to pay service providers based on their usage of these utility
Definitions, characteristics, and trends
In order to facilitate a clear understanding of what exactly is Cloud computing, we compare Cloud computing with two other recent, widely-adopted or explored computing paradigms: Cluster Computing and Grid Computing. We first examine the respective definitions of these three paradigms, then differentiate their specific characteristics, and finally highlight their recent web search trends.
Market-oriented Cloud architecture
As consumers rely on Cloud providers to supply more of their computing needs, they will require specific QoS to be maintained by their providers in order to meet their objectives and sustain their operations. Cloud providers will need to consider and meet different QoS parameters of each individual consumer as negotiated in specific SLAs. To achieve this, Cloud providers can no longer continue to deploy traditional system-centric resource management architecture that do not provide incentives
Resource management strategies for market-oriented Clouds
Since customer satisfaction is the crucial success factor to excel in the service industry [14], computing service providers have to be aware of user-centric objectives and meet them in order to achieve customer satisfaction. But, many service quality factors can influence customer satisfaction [14], [15]. Hence, we need to design SLA-oriented resource management strategies for Data Centers and Clouds that provide personalized attention to customers, such as enabling communication to keep
Global cloud exchanges and markets
Enterprises currently employ Cloud services in order to improve the scalability of their services and to deal with bursts in resource demands. However, at present, service providers have inflexible pricing, generally limited to flat rates or tariffs based on usage thresholds, and consumers are restricted to offerings from a single provider at a time. Also, many providers have proprietary interfaces to their services thus restricting the ability of consumers to swap one provider for another.
For
Emerging Cloud platforms
Industry analysts have made bullish projections on how Cloud computing will transform the entire computing industry. According to a recent Merrill Lynch research note [24], Cloud computing is expected to be a “$160-billion addressable market opportunity, including $95-billion in business and productivity applications, and another $65-billion in online advertising”. Another research study by Morgan Stanley [25] has also identified Cloud computing as one of the prominent technology trends. As the
Aneka: From enterprise Grids to market-oriented Cloud computing
We are working towards implementing a market-oriented Cloud using a .NET-based service-oriented resource management platform called Aneka [34]. Aneka is initially developed as a 3rd generation enterprise Grid technology. Recently, various new capabilities have been added to exhibit properties and potentials of the Cloud computing paradigm. An enterprise Grid [36] harnesses computing resources of desktop computers (connected over an internal network or the Internet) within an enterprise without
Meta-negotiation infrastructure between Aneka Clouds and Brokers
The Meta-Negotiation Middleware (MNM) represents the first implementation prototype for the establishment of global Cloud exchange and market infrastructure for trading services. The MNM bridges the gap between different proprietary service interfaces and diverse negotiation strategies used by service providers and consumers [46].
Before committing themselves to a SLA, the consumer and provider may enter into negotiations that determine the definition and measurement of user QoS parameters, and
Creating 3rd party Cloud services: Content delivery over Cloud storage services
Content Delivery Network (CDN) providers such as Akamai [53], [54] and Mirror Image [55] place web server clusters across the globe in order to improve the responsiveness and locality of the replicated content it hosts for end-users. However, their services are priced out of reach for all but the largest enterprise customers, and typically requiring lengthy contracts and large usage commitments [56]. We have developed an alternative approach to content delivery that leverages existing
Conclusion and future thoughts
Cloud computing is a new and promising paradigm delivering IT services as computing utilities. As Clouds are designed to provide services to external users, providers need to be compensated for sharing their resources and capabilities. In this paper, we have proposed architecture for market-oriented allocation of resources within Clouds. We have also presented a vision for the creation of global Cloud exchange for trading services. Moreover, we have discussed some representative platforms for
Acknowledgements
This work is partially supported by the Australian Department of Innovation, Industry, Science and Research (DIISR) and the Australian Research Council (ARC) through the International Science Linkage and the Discovery Projects programs respectively. Ivona Brandic was a visitor to the GRIDS Lab while performing this work. This paper is a substantially extended version of a keynote talk [70] presented at the 10th IEEE International Conference on High Performance Computing and Communications (HPCC
Rajkumar Buyya is an Associate Professor and Reader of Computer Science and Software Engineering; and Director of the Grid Computing and Distributed Systems (GRIDS) Laboratory at the University of Melbourne, Australia. He is also serving as the founding CEO of Manjrasoft Pty Ltd., a spin-off company of the University, commercialising innovations originating from the GRIDS Lab. He has pioneered Economic Paradigm for Service-Oriented Grid computing and demonstrated its utility through his
References (70)
- et al.
A computational economy for grid computing and its implementation in the Nimrod-G resource broker
Future Generation Computer Systems
(2002) - et al.
A commodity market algorithm for pricing substitutable grid resources
Future Generation Computer Systems
(2007) - et al.
Entropia: Architecture and performance of an enterprise desktop grid system
Journal of Parallel and Distributed Computing
(2003) A vision for the Internet
ST Journal of Research
(2005)Inside Track: The high-tech rebels
Financial Times
(2002)- et al.
Weaving computational grids: How analogous are they with electrical grids?
Computing in Science and Engineering
(2002) - D.S. Milojicic, V. Kalogeraki, R. Lukose, K. Nagaraja, J. Pruyne, B. Richard, S. Rollins, Z. Xu, Peer-to-peer...
Computing in the Clouds
netWorker
(2007)- Twenty experts define cloud computing....
In Search of Clusters
The grid economy
Proceedings of the IEEE
Service Quality: Research Perspectives
The Essentials of Risk Management
COSO Enterprise Risk Management: Understanding the New Integrated ERM Framework
SHARP: An architecture for secure resource peering
ACM SIGOPS Operating Systems Review
Tycoon: An implementation of a distributed, market-based resource allocation system
Multiagent and Grid Systems
‘Cloud computing’ seen as next wave for technology investors
Financial Post
Virtual workspaces: Achieving quality of service and quality of life in the grid
Scientific Programming
Cited by (0)
Rajkumar Buyya is an Associate Professor and Reader of Computer Science and Software Engineering; and Director of the Grid Computing and Distributed Systems (GRIDS) Laboratory at the University of Melbourne, Australia. He is also serving as the founding CEO of Manjrasoft Pty Ltd., a spin-off company of the University, commercialising innovations originating from the GRIDS Lab. He has pioneered Economic Paradigm for Service-Oriented Grid computing and demonstrated its utility through his contribution to conceptualisation, design and development of Cloud and Grid technologies such as Aneka, Alchemi, Nimrod-G and Gridbus that power the emerging eScience and eBusiness applications.
Chee Shin Yeo completed his Ph.D. at the Department of Computer Science and Software Engineering (CSSE), The University of Melbourne, Australia. He is currently working on economy- driven and Service Level Agreements (SLA)-based resource allocation for next-generation data centers. His research interests include distributed computing, services computing and utility computing.
Srikumar Venugopal is a Research Fellow in the Grid Computing and Distributed Systems Laboratory (GRIDS Lab) in the Dept. of Computer Science and Software Engineering, University of Melbourne, Australia. In 2006, he completed Ph.D. on scheduling of data-intensive applications. He has been part of the lab’s flagship Gridbus Project since its inception, and initially developed the Gridbus Broker.
James Broberg is an Australian Postdoctoral Fellow researching Coordinated and Cooperative Load Sharing between Content Delivery Networks (CDNs) at the University of Melbourne. He is the developer of MetaCDN, a Cloud-based CDN system. More information on his research activities can be found at : http://www.csse.unimelb.edu.au/~brobergj.
Ivona Brandic is a University Assistant Manager in the Information Systems Institute, Vienna University of Technology, Austria. In 2007, she completed Ph.D. on Grid Workflows at the University of Vienna. She is currently working on Cloud & Grid Computing; Grid, Web & Workflow Quality of Service (QoS); and Negotiation Protocols and Standards.