· · · Institutional Repository

Infrastructure as a Service (IaaS) is a delivery model of cloud computing, which provides the ability to users to acquire and release resources according to their demand and pay according to their usage. Resources are provisioned from the cloud as Virtual Machines (VMs), many of which can be deployed on a single computing node, realizing a multi-tenancy model. While virtualization and multi-tenancy are two sources of workload-execution overhead that have been studied in the past, we still need a thorough, empirical investigation of the joint impact of these overheads, on workload execution. Additionally, commercial and private IaaS providers offer mechanisms that facilitate the lease and use of single infrastructure resources, but to execute multi-job workloads IaaS users still need to select adequate provisioning and allocation policies to instantiate resources and map computational jobs to them. Even though some studies on the policies employed in cloud environments already exist, current and potential IaaS users need deeper insight on the achieved performance and incurred cost of the used policies, derived through empirical investigation. In this work, we address these problems with the use of SkyMark, a performance analysis framework for IaaS clouds. SkyMark has three key features: first, it is designed to analyze IaaS deployments through a sequence of automated tests and the subsequent automated analysis of results. Second, it can analyze the impact of individual provisioning and allocation policies to the performance of the workload execution. Lastly, it is able to generate complex workloads, stressing any of the compute, memory and disk components. With the use of SkyMark, we first study the overheads that the cloud software stack imposes to the workload execution. Subsequently, we analyze the performance and cost of six provisioning and three allocation policies through experimentation in three IaaS environments, including Amazon EC2.

Cloud computing has led to large pools of interconnected computer resources that can be acquired or released in a matter of minutes. No long-term contracts or large upfront investment required; resources are available on a pay-per-use basis. Consequently, a thousand servers used for an hour are as expensive as using one server for a thousand hours. It also enables resources to be provisioned according to demand, reducing unnecessary costs, while simultaneously making it resilient against significant and unexpected increases in demand. This model greatly reduces the risk and startup costs to innovate and disrupt. The rise of cloud computing has placed existing application owners in a challenging position. Their application is often not designed with this level of flexibility in mind. Because of the time and money invested in an existing application it is likely preferable to use this investment rather than abandoning it. Starting anew would also require a larger and relatively long-term investment that carries more risk. For this reason a hybrid approach to cloud computing could be an attractive means to leverage past investments as well as benefit from cloud computing. Through the use of an industrial test case we have explored some of the opportunities and challenges of cloud computing from the perspective of an application that was not designed with cloud computing in mind. These insights as well as the system's current limitations to scale led us to create a working prototype to store a subset of data to Amazon's Simple Storage Service (S3). Based on our results, storage costs for this subset of data are reduced twenty fold, the administrative burden to scale is completely transferred to Amazon and performance remains acceptable despite an overhead of 88 milliseconds due to internet latency. The prototype serves as a concrete proof of concept that demonstrates hybrid cloud computing can be a viable means to improve an existing system.

Information technology is the common denominator for all the industries in the 21st century, therefore any important change in this area is prone to have an impact on small and large businesses alike. The latest shock wave storming through IT is Cloud computing. Due to the importance and sensitive nature of applications used by financial institutions, the main goal of this research is to investigate how Cloud computing could change the way services are provided to customers and what is the emerging role of IT consultancy companies for this specific market segment. In order to do so, a proposed reference architecture has been created based on existing models and services in combination with the opinion of various experts from Capgemini and financial institutions. The impact of the proposed model, "Capgemini Immediate for Financial Institutions", has been expressed from a business and an IT perspective. Also, its functionality has been showcased in a scenario meant to underline the impact of the proposed changes on the boundaries of the system and the interaction of the financial institution with other entities. This process has been evaluated and supervised by experts from Capgemini in order to meet the standards used in the industry.

The increasing ubiquity and capabilities of smartphones provide opportunities to combine their processing power, storage, connectivity and sensors into a shared resource pool for end-users and service providers. This is analogous to the concept of cloud computing, in which cloud computing datacenters provide a shared resource pool. Several technological architectures and prototypes exist in which smartphones themselves serve as such a mobile cloud computing resource pool. However, little guidance is available regarding the possibilities for commercialization of mobile cloud computing. This study identifies factors influencing the feasibility and viability of the mobile cloud concept in order to provide input for future mobile cloud business models. To this end the opportunities for mobile cloud to deliver multi-sided platforms are explored. Multi-sided platforms serve as matchmakers between supply and demand side customers, while the platform owner can benefit from mediating between different types of customers. Mobile cloud platforms can mediate between end-users of mobile cloud services, individuals providing smartphone resources and service providers. To evaluate the feasibility and viability of mobile cloud platforms, ecosystems theory is used to formulate a set of qualitative criteria for the role divisions, structures and performance of multi-sided platforms. The criteria have been applied in fourteen semi-structured interviews with mobile industry experts, resulting in a range of success factors and inhibitors for mobile cloud ecosystems. With regards to ecosystem role divisions it is noted that mobile cloud platforms are most likely to succeed when positioned in existing strong ecosystems with large user bases of service providers and service consumers, such as those of the handset operating systems and internet based service providers. Operators lack such ecosystems and innovation capabilities and are therefore less likely mobile cloud platform owners. However, their support may still be required as end-users incur data subscription costs and cause inter-operator traffic handoff when sharing smartphone resources via the operator network. Billing mechanisms between users and between operators may need to be adapted to take this into account. With regards to ecosystem structure a mobile cloud platform owner can benefit from the revenues obtained by providing third parties and end-users access to a shared mobile resource pool, which may include unique and desired resources such as sensors. However, this coordinating platform ownership position may be difficult to maintain as similar access to mobile resources may be obtained by installing specific-purpose applications on smartphones. Operators and handset OS developers may therefore struggle to maintain platform control as internet-based service providers can use web-based cross-platform applications to gain access to any number of smartphone resources, regardless of geography and hardware. The performance of mobile cloud ecosystems is currently considered to be hampered due to current technological limitations and market conditions. Scarce and perishable smartphone resources, limited, expensive bandwidth and lack of clear end-user sharing incentives are the most visible hurdles, along with the security, privacy and legal concerns associated with smartphone resource sharing. While some of these issues may be addressed over time with technical and network improvements, these improvements may also disrupt the need for smartphone resource sharing. Faster operator networks diminish the need to share connectivity, while increasing smartphone power and cloud computing datacenters reduce the need to offload computing tasks to other smartphones. This potential performance of mobile cloud platforms is expected to change when more collaborative services making use of multiple devices are conceived, rather than considering mobile devices as a straightforward resource pool akin to cloud computing datacenters. A shared pool of smartphone sensors is expected to lead to innovations of which both end-users and service providers can benefit. Additionally it is noted that in private or community environments such as homes and offices some of these performance hurdles may not apply, as they provide an environment where trust and cost are less of an issue and appliances are more visible than in a general purpose resource pool. Follow-up research towards mobile cloud platforms could focus on mechanisms to award smartphones users for opting in to a shared resource pool. Furthermore, operator, device and service provider centric ecosystems could be further specified in terms of role divisions, relations and platform assets shared using the success factors and inhibitors uncovered in this research as a basis. Finally, as mobile cloud can potentially seamlessly integrate with the traditional cloud, future research could explore ways to optimize whether the resources of a traditional cloud datacenter, a local mobile cloud resource or a remote mobile cloud resource are acquired depending on cost and connection properties.

Introduction: The nowadays economical developments in emerging countries, like Brazil, Russia, India and China (BRIC) ask for a new approach in global practice for multinational companies. Currently most products for emerging countries are simple modifications or defeatured products based on products first designed for developed economies. Reverse innovation is a strategy where products are developed meeting the specific needs and budgets of customers in emerging countries. With a local approach in development and manufacturing innovative products can emerge with a possible spin off for developed countries, which gives the strategy its name: Reverse innovation. Project: This project has been exploring the possibilities for developing an affordable large format printing system by Océ, appropriate for emerging markets. By conducting a field research in India giving insights about printing practice, usage, context, environment, and technical and maintenance related aspects, a design proposal has been developed. This research was mainly focused on small and medium enterprises in the markets of print service providers (copy shops), architects, engineers and constructors. Research: The field research showed five key aspects that should be taken in to account for further product development: With India’s history in caste system and enormous population it has a service focused society, with services that offer solutions for people with every budget. Prepaid payment systems, pay per use and exact and transparent cost communications are elementary in this service practice. Although, with the printed volumes the total cost of ownership of Océ products is much more interesting than inkjet competitors, the initial cost price of the current product is too high and not affordable for offices in India. Offices and copy shops do have serious space constrains, especially in bigger cities. There is no space for media drawers and the printer is only accessible at the front and top side because of other equipment or office furniture. India has a very big and ambitious young population. The focus group of engineers and architects is well-educated, follows the latest technical developments and is an early adaptor of new technologies. They have studied or worked abroad and do know what they want. One of the most noticeable developments in India is the rapid growth in mobile communication, where wired infrastructure is largely skipped. Smart phones are becoming cheaper and will be the first access to Internet for a large population in India. Focus: The research showed that just defeaturing the product is not the solution. Buyers compare products on offered features and price, and all employees have to able to work with the product. They look for most value for money. A new product should be affordable, appropriate for the context and appreciated by the buyer and user. Strategy: In a strategy where product features are offered as pay per use services a more affordable initial cost price could be achieved without taking off features. By transferring the local image processing and data storage to cloud computing, and replacing the local interface by already used mobile devices, first steps in material cost price savings can be made. Business model: Using cloud computing offers a platform for new business models and automated services, for in time supply and maintenance and on distance support. It also offers possibilities for all kinds of cloud services that can be introduced over time, like accounting and document management. Conceptualization: In the design phase of this project the possibilities have been explored to make a smaller and more accessible multifunctional table top printer, with convenient handling, an affordable cost price, meeting the expectations of young professionals. Various workflow scenarios have been studied to improve document handling in an office environment for inexperienced users, with simple interface and cloud solutions. Proposal: The design proposal consists of three layers: printer engine, mobile interface and cloud services. The printer is a full functioning table top printer that anticipates to the space limited context in India. By making the printer smaller, using fabric parts, taking off the local interface and minimizing the local storage and processing the printer has a material cost price reduction of 30% compared to Océ’s current product. The mobile interface connects with the printer trough near field communication. The interface is context aware and adjusts the presented functionalities according to the placement of the mobile device: Mobile device on the printer for scanning, copying, fast print start and media change. Mobile device from the printer for making print jobs, managing scans and documents, accounting and task history. The proposed cloud functionalities can be categorized into six elements: Cloud information and cloud control to support the use of mobile devices and create a more efficient workflow. Cloud processing and cloud storage to compensate the defeatured engine and offer device independent document access. Cloud accounting and cloud monitoring are new services that offer users, Océ and other stakeholders up to date information to make their practice more efficient. Conclusion: This project showed that developing professional products for small and medium enterprises is more focussing on the emerging middle class with young ambitious professionals, than on button of the pyramid. Defeaturing products especially for these markets is not a strategy that will succeed in India. Using new technologies and smart innovations affordable solutions have to be achieved.

Research over the past several years has identified a convergence of the technology trends mobile computing and cloud computing into mobile cloud computing. While the term ‘mobile cloud computing‘ is often used to refer to the complementary benefits of mobile cloud based services, some translate the cloud vision from access to remote cloud resources in large data centers to local shared and pooled cloud resources consisting of mobile devices. Technical literature has showed the technical feasibility of this approach, however literature fails to show the practical viability of the local cloud approach for mobile business services that could be exploited to create value for organizations. This study evaluates the feasibility and viability of mobile business services enabled by mobile cloud computing that can be exploited by business enterprises. The research approach was design oriented and exploratory in nature. A literature review was conducted in order to identify technological opportunities and enabling capabilities of mobile cloud computing for mobile business services were. In addition, based on business model literature a framework was constructed that provides a blueprint of the process from the identification of a potential business service idea to business exploitation within an organization. This framework was applied to the business model outline of a potential local cloud based mobile business service. The results of this study indicate that the feasibility and viability of mobile business services enabled by mobile cloud computing is influenced by four aspects. First of all, the feasibility and viability depends on depends on the approach to mobile cloud computing. Secondly, the feasibility and viability is negatively influenced by security, privacy and trust issues, limitations regarding performance, reliability and scalability, and compatibility issues. Thirdly, the feasibility and viability seems to be negatively influenced by a lack of knowledge of how the technology should be designed in order to realize potential local cloud based business services. Fourthly, the feasibility and viability seems to be negatively influenced by the mismatch between technology and market need. The suitability of the technological opportunities and enabling capabilities of local cloud approach to mobile cloud computing are very context dependent and are not driven by solving business problems. All four aspects impact the feasibility and viability of mobile business services enabled by mobile cloud computing that can be exploited by business enterprises. With regard to the future of mobile cloud computing it can be suggested that organizations should be more involved in the development phase for identifying and matching customer needs and for the development of potential mobile cloud service concepts. It is expected that this will drive the development and diffusion of mobile cloud computing concepts in the future.

Cloud services such as Facebook and Google search started to use personalization algorithms in order to deal with growing amount of data online. This is often done in order to reduce the “information overload”. User’s interaction with the system is recorded in a single identity, and the information is personalized for the user using this identity. However, as we argue, such filters often ignore the context of information and they are never value neutral. These algorithms operate without the control and knowledge of the user, leading to a “filter bubble”. In this paper, by building on existing philosophical work, we discuss three human values implicated in personalized filtering: autonomy, identity, and transparency.

The sourcing of cloud services is a relatively new type of service delivery model in which an organization gets access to IT services via a cloud service provider that is delivering services over the web to many users on a pay per use or period basis. Even though the importance of IT governance is often underlined, there is limited literature available regarding the required IT governance capabilities that public sector organizations need to have in place to successfully implement and manage a cloud service delivery model. Using an existing governance framework of IT core capabilities as basis, the required cloud computing capabilities are investigated using interviews and studying reports. The analyses helped to identify 16 discriminating capabilities that are essential when effectively implementing and managing cloud services in the public sector. Different factors, including the cloud service and deployment model, the strategic intent underlying cloud sourcing, the degree and complexity of cloud sourcing and the IT governance structure, were found to influence the relevance of cloud capabilities and the relevance might also change over time.

Cloud computing market is expected to grow rapidly in the next five years. Even though Software-as-a-Service and Infrastructure-as-a-Service solutions will dominate most of the market, Platform-as-a-Service solution is forecasted to have the fastest-growing segment, especially in Western Europe. On the other hand, cloud computing is also argued to be the new playing ground for Telecommunication industry. As the industry facing a threat for just being the network “bit piper”, telecom operators might find a new revenue source within the cloud computing domain. The issues explained above have led us to conduct a research that aimed at two objectives. First is to give guidance to larger enterprise companies in the Netherlands on how to capture values from PaaS offering. Second, this research aimed to identify whether telecom companies could create particular new business value and opportunities from the PaaS service provision. Therefore this study raised a challenging main questions: “What factors are important for large enterprise clients to adopt Platform-as-a-Service in order to capture the business value that is offered from Platform-as-a-Service offering, and what are the strategic implications of the Platform-as-a-Service adoption for Telecom industry?” To answer the main questions, five research sub-questions were formulated. A theoretical approach from business model literature is used to guide answering the main research questions. A conceptual model of PaaS adoption factors was developed, which is based on the STOF business model framework. Analysis of PaaS market in the Netherlands was conducted among 9 client companies and 3 PaaS provider companies, in order to validate the conceptual model. The results showed that the PaaS market in the Netherlands are still immature, thus making the validation of the conceptual model for PaaS adoption hard to be generalized. Modification of the analysis approach was taken that enable us to analyze the conceptual model for SaaS adoption as well. Our study found that there are several factors within the Service domain that are important for clients when they consider which applications to be moved to PaaS or SaaS environment. The most important factor is criticality of application that wants to be moved to the cloud. These factors from the Service domain influence client’s technical requirements on Security, Quality of Service, and System Integration issues. Furthermore, our research found that there are several factors that client take into account when they about to choose PaaS and SaaS providers. These factors serve as assessment framework whether the providers’ profile can meet clients’ requirements. Our finding shows that Branding is the most important factor of PaaS and SaaS providers. Branding is important because it is closely associated with trustworthiness. Moreover, trust also becomes really important, because in PaaS and SaaS implementation clients are losing control of their application data. In addition, our finding also shows that the factor of data location is really important in PaaS and SaaS adoption. A key stakeholder, namely the regulators, drives this data location factor. We also found that clients perceived system integrators and network providers as the other important key stakeholder in the provision of PaaS and SaaS services. Furthermore, our study also found that telecom companies could play two roles in cloud computing domain in order to create new business value, and hence making new revenue source. The first role is to be Cloud service provider. This role means that telecom companies could exploit their main resources of Internet network infrastructure to deliver a guaranteed and high level end-to-end cloud services (both PaaS or SaaS services). The second role is to be Cloud service broker. This role exploits telecom companies’ resource of broad customer base and customer relationship in order to mediate cloud providers and the clients. Being cloud service broker enables telecom companies to gain monetary incentives by means of revenue sharing. Our study concludes that clients need to take into account factors in the four domains of Service, Technology, Organization, and Finance in order to capture the value of PaaS and SaaS offering. On top of that, clients need to take into consideration the regulation applied in order to legally implement cloud computing solutions. Clients also need to acknowledge the importance of system integrators and network providers in utilizing cloud computing services. This study adds several main contributions to the literature. Firstly, this study has made the first attempt to construct an adoption model for PaaS and SaaS specifically. Secondly, this study has contributed to business model theory, particularly by extending the applicability of STOF model. Thirdly, this study confirms the existing literature that describes key stakeholders on cloud computing. On top of that, we contribute by adding Network Provider as another key stakeholder. Nevertheless, this study bears some limitations. Due to small sample size, this research serves as means of first validation of the constructed model. We are aware that further research can be conducted to do more extensive validation by means of survey research. However due to the immaturity of cloud computing market, especially PaaS, we suggest that further research on PaaS is better to be conducted later to wait for the market to be developed. Besides the sampling issue, our research was not able to validate our findings on telecom companies role in cloud computing. Further research could focus on validating our results empirically through involvement of industry experts.

With the increasing adoption of distributed systems in both academia and industry, and with the increasing computational and storage requirements of distributed applications, users inevitably demand more from these systems. Moreover, users also depend on these systems for latency and throughput sensitive applications, such as interactive perception applications and MapReduce applications, which make the performance of these systems even more important. Therefore, for the users it is very important that distributed systems provide consistent performance, that is, the system provides a similar level of performance at all times. In this thesis we address the problem of understanding and improving the performance consistency of state-of-the-art distributed computing systems. Towards this end, we take an empirical approach and we investigate various resource management, scheduling, and statistical modeling techniques with real system experiments in diverse distributed systems, such as clusters, multi-cluster grids, and clouds, using various types of workloads, such as Bags-of-tasks (BoTs), interactive perception applications, and scientific workloads. In addition, as failures are known to be an important source of significant performance inconsistency, we also provide fundamental insights into the characteristics of failures in distributed systems, which is required to design systems that can mitigate the impact of failures on performance consistency.

Software as a Service (SaaS) model allows subscription to a wide variety of application services that are developed specifically for and delivered over the Internet on an as-needed basis without the need to install and manage third-party software in-house. According to Salesforce.com Inc the wide adoption of SaaS model will eventually lead to the end of on-premise software era. Currently the success of SaaS model goes hand-in-hand with popularity of cloud computing. For instance, recently Google in collaboration with Samsung introduced to the mass-market their Chromebook with cloud-based operational system Chrome OS on board, which is also delivered as a service. Nevertheless, the idea of outsourcing the software or hardware is not new. Before SaaS there was Application Service Provider (ASP) model that in the past considered being very promising as well, but failed to meet the requirements of the wide market and serves a niche market today. The interesting fact is that ASP and SaaS models are very similar and some authors even don’t make distinction between them. However, we believe that there are differences between them that affected the adoption of the models. Thus, we have set two objectives for this study. First is to conduct comparison analysis of two software delivery models from business model perspective and study factors that possibly affected the adoption. Second is to identify components of the business model that require changes in order to make a shift from one software delivery model to another and barriers that hamper these alterations. Therefore this research raised the following main research question: “Why is Software-as-a-Service model more successful today than the model applied by Application Service Providers?” In order to answer the main research question above, we have formed four sub-questions, answers for which were used as a foundation for the final analysis. We have adopted two theoretical frameworks to answer research questions. Firstly, in order to get an overview of the past and current state of the cloud computing industry, the diffusion and development patterns theory was used. Secondly, for the purpose of software delivery models comparison we took the STOF business model framework through which it was possible to highlight differing components of SaaS and ASP. Subsequently, we have formed and validated six propositions through two completely independent methods – case studies and expert interviews. The results showed that business models of SaaS and ASP differ in four components: Technical Architecture particularly Software Architecture, Pricing, Cost and Market Segment. As predicted the most important component turned out to be the Technical Architecture that practically co-determined differences in other aspects. We confirmed that the Technical Architecture of the SaaS model is better in reaching ‘economies of scale’ compared to the ASP model, therefore the SaaS model was able to cut costs, drop service prices and serve wider markets which positively contributed to the large-scale diffusion of the model. 2 Evolution of the Software-as-a-Service model Furthermore, a set of recommendations for managers on ways of switching from one software delivery model to another were formed. We also conclude that although the SaaS has more advanced software architecture that makes it more successful on the market, it is still not a perfect solution for all types of companies. Mainly because of security issues that multi-tenant architecture entails, the enterprise application delivered over the ASP model could be a solution for certain market niches dealing with sensitive data, lacking IT expertise and willing to pay extra for the service. Therefore before committing to a certain model, providers have to carefully consider which type of companies they are able and willing to serve. It is also very important for companies that already have large installed base of customers and legacy software with single-tenant architecture. We found that virtualization technologies are rapidly developing and practically enable single-tenant applications to “fake” multi-tenancy and run on comparatively high levels of resource utilization as the SaaS model. Therefore the first recommendation for the traditional enterprise software providers would be to decide whether they are willing to maintain their current market segment or capture a wider market of SMEs and even individual end-consumers. Based on that decision, the most suitable architecture could be chosen. However, this study has some limitations among which the most serious is the lack of attention to external macro economical factors that could play an important role in diffusion and development of the innovation. Although, to partially cover that topic was done, it still requires more careful and extensive research. Therefore suggestion would be to consider it as an opportunity for the further research in this field.

The study of the history of computing infrastructures reveals an integration trend. For example, the explosive growth of the Internet in the 1990s was the result of an integration process started in the 1960s with the emerging networks of computers. By using the Internet, millions of users were capable of accessing information anytime and anywhere, much like other daily utilities such as water, electricity, and telephone. However, an important category of users remained under-served: the users with large computational and storage requirements, e.g., the scientists, the companies that focus on data analysis, and the governmental departments that manage the interaction between the state and the population (such as census, tax, and public health). Thus, in the mid-1990s, the vision of the (Computing) Grid as a universal computing utility was formulated. The main benefits promised by the Grid are similar to those of other integration efforts: extended and optimized service of the integrated network, and significant reductions of maintenance and operation costs through sharing and better scheduling. While the universal Grid has yet to be developed, large-scale distributed computing infrastructures that provide their users with seamless and secure access to computing resources, individually called Grid parts or simply grids, have been built throughout the world---in different countries, for different sciences, and both for production work and for computer-science research. At the same time, the main technological alternatives to grids, that is, supercomputers and large clusters, have evolved into much larger, scalable, and reliable systems. Thus, the integration of existing grids into larger infrastructures and finally into The Grid is key in keeping the grid vision attractive for its potential users. The integration of grids raises a double challenge, one related with the efficient scaling of a distributed computing system, the second associated with the operation of a system across different ownership and administrative domains. Thus, many of the traditional approaches for inter-operating computer systems, such as those based on completely centralized or purely decentralized system architectures, are eliminated from the start. To mark the distinction between the typical problem of integrating smaller components into a larger system and the double challenge of grid integration, we call the latter the problem of grid inter-operation. In this thesis we approach the problem of grid inter-operation with two main objectives: to design a comprehensive framework for the study of grid inter-operation mechanisms, and to provide an initial but good solution for this problem. We design a framework for the study of grid inter-operation that includes a toolbox for grid inter-operation research and a method for the study of grid inter-operation mechanisms. In the research toolbox we include the Grid Workloads Archive (GWA), a comprehensive model for grid resources and workloads, the GrenchMark performance evaluation framework, and the Delft Grid Simulation (DGSim) framework for repeated and realistic simulations of multi-cluster and multi-grid environments. The GWA and our comprehensive model show that grid computing is mostly used in practice for single-processor jobs and not for parallel computing, which raises previously ignored challenges related to the volume of jobs to be managed. We also devise in this thesis a method for studying grid inter-operation mechanisms. We answer using our framework important questions regarding existing grid operation mechanisms, and in particular show that these mechanisms are too limited to cope with real and realistic conditions. We further demonstrate the usefulness of our framework by designing Delegated MatchMaking, a novel mechanism for inter-operating grids. This mechanism is used to operate an architecture that is a hybrid between hierarchical and purely decentralized architectures. The Delegated Matchmaking mechanism attempts to use the local resources of a grid as much as possible and also transparently extends the local environment with resources obtained (delegated) from other sites when resources are not available locally. Our approach is compared with five alternatives through trace-based simulations, and is found to deliver the best performance, especially when the system is heavily loaded. While many other mechanisms can be designed in the future, our experiments prove that the Delegated MatchMaking approach already is a good solution for the problem of grid inter-operation. Our experiments also demonstrate that having grids inter-operate leads to better performance than having the same grids operate independently.