Virtualization is a key enabler of cloud computing that allows to run multiple virtual machines (VM) with their own software environment and applications on top of physical hardware with the promise to increase efficient usage of hardware resources at lower cost and increased elasticity. Typically VMs are created using provider-specific templates (so called VM images) that are stored in proprietary repositories which leads to provider lock-in and hinder portability or simultaneous use of multiple federated Clouds. Optimization at the level of the VM images is needed both by the Cloud applications as well as by the underlying Cloud providers for improved resource usage, speed, elasticity, redundancy, fault tolerance and other much desired Quality of Service (QoS)-related features. Critical barriers exist that prevent many users from industry, business and academia to effectively use cloud resources and virtualization environments for their computing and data processing needs. In this project we will create a novel (ENTICE) environment targeting federated Cloud infrastructures for: (i) simplifying the creation of lightweight and highly optimized VM images; (ii) automatic decomposition and distribution of VM images based on multi-objective optimization (performance, economic costs, storage size, and QoS requirements) and a knowledge collection and reasoning infrastructure, and (iii) auto-scaling of Cloud resources that supports interoperability of VMs across Cloud infrastructures without provider lock-in. We gathered an interesting selection of complementary use cases from energy management to earth observation and cloud orchestration which will be used to validate the ENTICE environment and that are provided by two SME and one industrial partners of the project.

Cloud federation enables cloud providers to collaborate and share their resources to create a large virtual pool of resources at multiple network locations. Different types of federation architectures for clouds and datacenters have been proposed and implemented (e.g. cloud bursting, cloud brokering or cloud aggregation) with different level of resource coupling and interoperation among the cloud resources, from loosely coupled, typically involving different administrative and legal domains, to tightly coupled federation, usually spanning multiple datacenters within an organization. In both situations, an effective, agile and secure federation of cloud networking resources is key to impact the deployment of federated applications. The main goal of this project is two-fold: research and develop techniques to federate cloud network resources, and to derive the integrated management cloud layer that enables an efficient and secure deployment of federated cloud applications. Our proposal will deliver a homogeneous virtualization layer, on top of heterogeneous underlying physical networks, computing and storage infrastructures, providing enablement for automated federation of applications across different clouds and datacenters. The project is fully committed to open source software. Cloud networking aspects will be based on OpenDaylight, a collaborative project under The Linux Foundation, and specifically we will leverage and extend the OpenDOVE project with new rich inter-cloud APIs to provision cross-site virtual networks overlays. The new inter-cloud network capabilities will be leveraged by existing open source cloud platforms, OpenNebula and OpenStack, to deploy multi-cloud applications. In particular, different aspects of the platforms will be extended to accommodate the federated cloud networking features like multi-tenancy, federated orchestration of networking, compute and storage management or the placement and elasticity of the multi-cloud applications.

The rapid increase in demand for data-intensive applications capable of exploiting Big Data technologies such as Hadoop/MapReduce, NoSQL, cloud-based storage, and stream processing is creating massive growth opportunities for European independent software vendors (ISVs). However, developing software that meets the high-quality standards expected for business-critical cloud applications remains a barrier to this market for many small and medium ISVs, which often lack resources and expertise for advanced quality engineering. DICE will tackle this challenge by defining a quality-driven development methodology and related tools that will markedly accelerate the development of business-critical data-intensive applications running on public or private clouds. Building on the principles of model-driven development (MDD) and on popular standards such as UML, MARTE and TOSCA, the project will first define a novel MDD methodology that can describe data and data-intensive technologies in cloud applications. A quality engineering toolchain offering simulation, verification, and numerical optimisation will leverage these extensions to drive the early design stages of the application development and guide software quality evolution. DevOps-inspired methods for deployment, testing, continuous integration and monitoring feedback analysis will be used to accelerate the incorporation of quality in data-intensive cloud application both in public and private deployments, enhancing the capability of small and medium European ISVs to enter the Big Data market.