The Christian Doppler laboratory for client-centric cloud computing is devoted to address primarily the research problems associated with the client side of cloud computing. That is, the key questions addressed are the following:
- What are the expectations and benefits for companies, in particular small and medium enterprises, in cloud computing?
- What are the security and privacy hazards (loss of control, data leaks, etc.) for clients and how can they be addressed to gain trust in cloud computing?
- How is flexibility of the clients guaranteed to join and also leave clouds, whenever they decide to do so?
The research will not aim at solving the many interesting problems associated with the provision of cloud services, but in view of the many world-wide research efforts addressing data management, quality-of-service and security in the cloud we are confident that these problems will be addressed elsewhere. The research will also dispense with an intensive study of performance and scalability for cloud offers, for which similar arguments apply.
We envision the following scenario for the engagement of clients in cloud computing: As the transition to cloud computing will not be done in one step, but instead many lines of business applications will remain on premises, there have to be possibilities for a step-by-step adoption. We also see a great resistance, particularly in small and medium companies, to transfer all their data to the cloud. This means that the clients are still in a sort of \hybrid" operation mode, where they access cloud services and on-premises services simultaneously. The coexistence of these services has to be investigated, and solutions especially for the client side have to be developed. Furthermore, at the moment there is a great concern of not having the directory with the security credentials on site, so that the company can keep total control over the usage of services, regardless whether they are on-cloud or on-premise. Last but not least, also the logging of service usage should be done on-premise to allow correct forensics.
The research programme of the proposed CD laboratory comprises two modules. Our literature review revealed that research on cloud computing takes an exclusive server-oriented view. However, in order to let the vision of cloud computing to become reality, it is equally important to address the problems arising on the side of the client. Therefore, Module I - Client Needs addresses the clients' needs with respect to service mediation, adaptivity, transparency, client management, and all contractual problems between service providers and users associated with cloud computing.
As cloud computing exposes service user to the public, security and privacy are crucial. These problems will be addressed in Module II - Security and Privacy. Here we also observed that most of the research that has been done so far focuses on the protection of the service providers, while client-related security aspects do not receive much attention. Therefore, we move security problems on the client side into the focus of our research.