Cloud Computing is an emerging computing paradigm where applications, data and infrastructures are provided as services, ubiquitously accessible from any device connected to the Internet.

Such a concept will be at the heart of the future Internet - implementing the vision of computing and information processing as a utility. Clouds are currently hosted in data centers and provide an elastic set of resources that are used by software that is deployed and accessed as services.

Clouds will also extend to exploit resources of devices at the edge of the network.  Eventually a service may be created, or a content may be provided, by a local device, by a network element or a hosting data center.

Cloud computing has many advantages: economy of scale, frugality in energy consumption,elasticity - the ability to quickly and dynamically scale up and down in response to changes in load,the lack of any entry threshold, and the possibility to pay only for actual usage or “pay as you go”.  

Clouds are expected to be a boon to innovation, as
• Clouds lower the entry threshold for start-ups and SMEs,
• Cloud elasticity allows IT-startups and SMEs to quickly and easily expand the IT-services that they offer.
• Clouds allow for large-scale testing and prototyping of sophisticated novel services with no capital investment.

Clouds have an important place in the two most important trends we see in IT today, the increasing proliferation of data-intensive services and the digital convergence of Telecommunications, Media and ICT.

However, currently clouds have a number of limitations. The promise of computing as a utility is not yet realized, as it is difficult for consumers to move between and combine services from different cloud providers, and because there is no way for new providers to offer their resources to the utility. For some applications existing clouds do not have sufficient elasticity, making them unsuitable for real-time media and entertainment.

Resource allocation frameworks are coarse-grained and bandwidth is not treated as a managed resource, making it difficult to test and predict the behavior of distributed applications running in the cloud. Existing software tools for data-intensive applications lack functionality and are difficult to use. The complexity, scale, uncertainty, and dynamism of the infrastructure and hosted applications, require new methods for cloud management.

Security in clouds needs to be improved, not least to protect users from the provider. There is a need to augment contractual security with technology-based, verifiable security features. Access networks infrastructures must cope with cloud services’ demands of high speed and QoE. 

In the action line, research is conducted on a number of topics to address these shortcomings:

• Mechanisms for elasticity and resource allocation of all types of resources including networking
• Distributed storage systems and indexing
• Benchmarking
• Computational frameworks and infrastructures for data-intensive applications
• Marketplaces for Data and Software as a Service
• Cloud management and autonomics
• Integration of Information-Centric Networking and data centers
• Service composition and engineering 
• Cloud security

Currently there is no large scale cloud testbed in Europe for experimentation and research. Therefore EIT promotes the creation of a large European cloud infrastructure. In addition, mobility of researchers and engineers, between partners (and countries), both academic and industrial, is promoted through the mobility program. 

Lead: Prof. Seif Haridi, SICS/KTH.