A Meta-Modeling Service Paradigm for Cloud Computing and Its Implementation*

Jia Hangsheng,Cheng Fei,Tang Xiaoting

(1.School of Management,Hefei University of Technology,Hefei 230009,China;2.School of Management,University of Science and Technology of China,Hefei 230026,China)

A Meta-Modeling Service Paradigm for Cloud Computing and Its Implementation*

Jia Hangsheng1,Cheng Fei1,Tang Xiaoting2

(1.School of Management,Hefei University of Technology,Hefei 230009,China;2.School of Management,University of Science and Technology of China,Hefei 230026,China)

It is argued in this article to shed light on service supply chain issues associated with cloud computing by examining several interrelated questions:service supply chain architecture from service perspective;basic clouds of service supply chain and development of managerial insights into these clouds.In particular,to demonstrate how those services can be utilized and the processes involved in their utilization,a hypothetical meta-modeling service of cloud computing can be given.Moreover,the paper defines the architecture of cloud managed for SV or ISP in infrastructure of cloud computing in service supply chain:IT services,business services,business processes,which creates atomic and composite software services that are used to perform business processes with business service choreographies.

service supply chain,service vendor,integration service provider,cloud computing,coordination

*Acknowledgement:This research was funded by the National Natural Science Foundation of China(No.70631003,No.70801024)

1 Introduction

Supply chain integration,as a general concept,has been under a lot of criticism in the last years.A myriad constructs and measurements can be found in the literature.A supply chain can be viewed as an organization plus its supply network,its distribution network,its alliance network,and its end consumers involved in procuring,producing and delivering products and services to customers(Deshpande,2000).Organizations have traditionally taken a uniform approach to service network design in organizing their service activities to meet a single service standard.

There exists a need for strategies and methodologies to adapt supply chains to serve differentiated customers in a cost effective way.

At present,the computing world is rapidly transforming towards developing software for millions to consume as a service,rather than to run on their individual computers.Cloud computing is an extension of a 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(Buyya et al.,2009).Since consumers’requirements for cloud services are varied,service providers have to ensure that they can be flexible in their service delivery while keeping the consumers isolated from the underlying infrastructure.The goal of this paper is to help decision makers in service supply chain to understand where cloud computing may be appropriate and to describe clouds for service supply chain.

2 Service supply chain architecture from service perspective

To highlight the interdisciplinary nature of the effort,IBM started an initiative in this area that it has called“Service Science,Management and Engineering” (SSME)as we mentioned earlier (Demirkan et al.,2008).For most services,elements of both product and service oriented operations need to be included (Brady and Fellenz,2007).Much of technology within supply chains has been aligned with the automation of the process.One key element of managing the service supply chain is to have an explicit knowledge and understanding of how the service network structure is configured.Cloud computing promises to deliver on this objective:customers are able to rent infrastructure as needed,deploy applications and store data on the infrastructure,and access the application and data via web protocols on a pay-per-use basis(Klems et al.,2009).For the enterprises,Cloud computing is primarily a new business paradigm,as opposed to a new technical paradigm in service supply chain management at present.

To demonstrate how those services can be utilized and the processes involved in their utilization,a hypothetical meta-modeling service of cloud computing can be given.As illustrated in Fig.1.Many activities use software services as theirbusiness basis in service supply chain.These integration service providers(ISP)make services accessible to the customers or market through internet-based interfaces.Clouds aim to outsource the provision of the computing infrastructure required to customer service request.This infrastructure is offered as a service to customer by the integration service provider,moving computing resources from the service vendor(SV)to the customer,so the ISP and the customer can gain in flexibility and reduce costs.

As can see from logical relationship,the service supply chain consists of service vendors (SVs),integration service providers(ISPs),and customers.The service vendors supply original service productions or activities to the SIPs that in turn sells the value-add integration service to customers.For meta-modeling services of cloud computing in service supply chain,cloud vendors(service vendors)provider hardware or part application as a service to integration service provider.The XaaS provider (Integration service provider)allows customers to gain the capabilities of a simple server-albeit a virtual one-in which the processing,network,and storage resources are controlled dynamically.More sophisticated clouds in XaaS clouds also provide useful management capabilities, programming environments, web service platforms,or access to particular applications.Customers can acquire or relinquish processing power and storage,often in minutes,merely by sending a service request to the XaaS provider.The communication channel is virtual in the sense that the XaaS provider capacity as demanded from its infrastructure platform.

3 Clouds for service supply chain

Cloud computing is a new paradigm shift following the mainframeand client-servershiftsthatpreceded it.It describes a new supplement,consumption and delivery model for IT services based on the internet,and it typically involves the provision of dynamically scalable and often virtualized resources as a service over the internet.There seems to be many definitionsofcloud computing allowing forthe extraction of a consensus definition as well as a minimum definition containing the essential characteristics,but they mainly focus on just certain aspects of the technology.A study by Luis M.Vaquero et al.discussed the concept of cloud computing to achieve a complete definition of what a cloud is,using the main characteristics typically associated with this paradigm in the literature(Vaquero et al.,2008).In fact,no common standard or definition for cloud computing seems to exist about more than 20 definitions (Voas and Zhang,2009).A more commonly used definition describes it as clusters of distributed computers which provide on-demand resources and services over a networked medium(Sultan,2009).At a cursory glance,the majority of cloud computing infrastructure consists of reliable services delivered through data centers and built on servers.Clouds often appear as single points of access for all customers’computing needs to meet a specific service-level agreement via web service technologies.

3.1 Cloud features

For service supply chain,the following features are commonly associated with XaaS clouds,which showing the cloud features identified from many of definitions in the literature.A XaaS user can be a consortium or participant(Rosenthal et al.,2009).

Resource outsourcing:instead of a consumer providing their own hardware,the cloud vendor assumes responsibility for hardware acquisition and maintenance.

Large numbers of service workloads:clouds are typically constructed using large numbers ofdifferentservice workloads.As a result,the cloud vendor can more easily add capacity and can more rapidly align workloads as needed,compared with having service workloads in multiple circumstances.Generally speaking these service workloads are as homogeneous as possible both in terms of configuration and location in service supply chain.

Utilitycomputing:thecustomerrequestsadditional resources as needed,and similarly releases these resources when they are not needed according to their self service demands.Different clouds offer different sorts of resources.

Virtualization:virtualization is a key enabler of architecture as wellas cloud applications.Hardware resources in clouds are usually virtual resources,which are shared by multiple services request to improve efficiency.That is,several lightly-utilized logical resources can be supported by the same physical resource.

High level services:clouds suffer a certain lack of high level services,which is probably related to the lower level of maturity of the paradigm.Clouds let these issues to be treated at the application level,although federated Clouds will likely require several mechanisms to deal with these topics.

Scalability and self-management: clouds free programmers of dealing with scalability issues.Clouds offer the automatic resizing of virtualized hardware resource.Scalability and self-management is simpler in a single administrative domain,but many problems can be found across organizational frontiers.

3.2 Cloud infrastructure for service supply chain

In the service supply chain,the integration service providers (XaaS providers)effectively sell computation and storage resources as service commodities,providing customers with the illusion of a single virtual machine or cluster.For example,the customers control computational and storage capacity by sending a service request to add or subtract resources as needed based on the Google Apps application platform.The release capacity is typically measured in minutes.To help explain how cloud computing is used in service supply chain;itis valuable to consideran architectural view of the building blocks that typically exist in service supply chain.Basically,this framework is based on using software components to create atomic and composite software service that are used to perform business processes with business service choreographies.As shown in Fig.2

We define the architecture of cloud managed for SV or ISP in infrastructure of cloud computing in service supply chain:IT services,business services,business processes.The major functions (supporting or core)of a business such as human resource,accounting,etc.are defined as coarse grained services,referred to as business services in the cloud management.It can be divided into common,industryspecific,and company-specific.These descriptions can be used to find services from the marketplace that may fulfill the requirements.The IT services represent the services that are offered in the cloud.Innovative approaches in service search technology are required to combine techniques as well as highly scalable to index many services that will be available in the cloud and allow service seekers to pose potentially diverse constraints,transfer and protection.Indeed,there is a need for approaches to tag directly the data with security and privacy policies so that the proper technical controls can be employed by the various cloudsprovides.Finally,the business processes is the representation of selection,design,integration and composition ofservicesthatfulfillthe requirements of outlined business environment.Emerging cloud services for integration such as Bungee Connect has taken the first step by providing pre-defined connectors.However,it is inadequate for unsupported application or complex.To enable scaling such approaches to the diversity ofapplicationson the internet,a service integration development platform and marketplace has to be built.

Cloud computing refers to both the applications delivered as services over the internet and the hardware and systems software in the datacenters that providethose services.The services themselves have long been referred to as “X as a service (XaaS)”;values of X we have seen in print include infrastructure,software,database,and platform.However,by understanding the type of service offered by cloud computing,one begins to understand what this new approach is all about.Depending on the type of provided capability,we summarize the four main types of services that can be offered by the cloud in the following(see Fig.2).

Infrastructure as a Service (IaaS):the ISP manages a large set of computing resources,such as storing and processing capacity.Based on service request from customers,the ISP is able to split,assign and dynamically resize these resources to build ad-hoc systems,which including the remote delivery through the internet of a full computer infrastructure(e.g.virtual computers,servers,storage devices,etc.).At present,some companies are billed for their usage following a utility computing model,where usage of resources is metered.For example,HP flexible computing services provides computing and storage infrastructure as services for businesses.Amazon often offers EC2 for computing power for small businesses and individual consumers where they can deploy and run their programs.

Software as a Service(SaaS):Under this cloud,software applications are delivered through the medium of the internet as a service,and offered as services rather than as software packages to be purchased by individual customers.This is an alternative to locally run applications.There are services of potential interest to the ISPs and the SVs.This type of cloud service offers complete application functionality applications to programs instead of installing and maintaining software.Example of commercial application in this category is Salesforce.com offering its CRM application as a service,which can be pay per use or monthly subscription per use.

Database as a service (DaaS):Daas is a viable model and has a good chance of emerging as a successful commercial offering for some applications.DaaS often adopts amulti-tenantarchitecture,wherethedataofmany customers is kept in the same physical table on the cloud.Example of service provider is Amazon SimpleDB,an internet-based database service built on the top of which providesuserswith toolsforapplication development,creating and loading tables,and performing queries and transactions.Also,in most cases,the database structure is not relational.Database service provider provides seamless mechanisms for organizations to create,store and access their databases.i.e.customer wishing to access data will now access it using the hardware and software at the service providerinstead oftheirown organization’scomputing infrastructure.Moreover,this would alleviate the problem of purchasing,installing,maintaining and updating the software and administrating the system.

Platform as a Service (PaaS):PaaS is a paradigm for delivering operating systems and associated services over the internet without downloads or installation,which providing facilitiestosupporttheentireapplication development lifecycle including design, implementation, debugging,testing,deployment,operation and support of rich web applications and services on the internet.Example of platform in this category is Google Apps Engine.Services can be obtained from diverse sourcesthatcross international boundaries.Meanwhile,initial and ongoing costs can be reduced by the use of infrastructure services from a single vendor rather than maintaining multiple hardware facilities that often perform duplicate functions from incompatibility problems.However,limited platform interoperability and limitations of programming platforms in supporting capabilities are majorconcernsofusingthesecurrent platforms.Nevertheless,overallexpenses can also be minimized by unification of programming development efforts.3.3 Cloud concerns questions Cloud computing,as we noted earlier,is an emerging computing service paradigm.There exist uncertainties and concerns about the maturity of technology,mainly including vendor lock,performance,latency,transparency,security,privacy,reliability,etc.

In some cases,a service vendor may wish to build in the right but not the obligation to stop the delivery of services,when the cost of doing such business becomes unprofitable.Option pricing methods typically emphasize the variance of the underlying cost of service delivery,and the extent to which they may become unfavorable to the vendor(Demirkan et al.,2008).Application of value-at-risk thinking in the service management context comes with a desire on the vendor’s part to identify the value-optimizing duration of a service contract (Kauffman and Sougstad,2008).If customers are dissatisfied with one cloud computing service or the ISP goes out of business,then customers cannot necessarilyeasilyand inexpensivelytransferserviceto another ISP or bring its clouds back in house.Instead,if customers or firm bring service clouds in house,it is pressure to hire employees with the skill to work with the technology,and reformat or transfer service applications to a new ISP,a potentially complex process.

IT managers are likely to be wary of surrendering control of their resource to outside providers who can change the underlying technology without customers’consent.Issues relating to performance and latency are also cited as problematic (Sultan,2009).Meanwhile,there are no cloud computing standards for processes such as APIs processing data import and export.There are also hampering adoption by limiting the probability of data and applications in different system.

Service vendors are trying to address their capabilities concern by have their service workload audited in advance by the ISP and documenting procedures designed to address customers’demands.The ISP can demonstrate who has access to data and how keep unauthorized personnel from retrieving service information.For cloud users,reliability may be an important problem.Cloud computing hasn’t always provided reliability.Forexamples,Salesforce.com left customers without service for six hours on 12 February 2008.And Amazon's S3(Simple Storage Service)and EC2(Elastic Compute Cloud)services suffered a three-hour outage three days later (Leavitt,2009).When most service vendors provide service credits for outages,this will result in miss and executive cut off from business information so as to discomfort service opportunities.

In addition,internet connections for cloud computing can also be a serious problem for users.For example,broadband internet connections aren’t fast or reliable enough for cloud computing to be dependable.If there aren’t available in every place,then those are far from service central office at some times and in some areas.Also,there aren't enough major cloud services providers yet,which reduces the level of customer choice and marketplace competition.

4 Conclusion

Cloud computing is a disruptive technology that is set to changehow IT systemsaredeployed because ofits apparently cheap, simple and scalable nature(Khajeh-Hosseini et al.,2010).The main contribution of this paper is our unique approach to configuration meta-modeling of service supply chain based on cloud computing.To demonstrate how those services can be utilized and the processes involved in theirutilization,a hypothetical meta-modeling service of cloud computing can be given.Moreover,toperform businessprocesseswithbusiness service choreographies,we define the architecture of cloud managed for SV or ISP in infrastructure of cloud computing in service supply chain.

2010－07－05)