iKaaS set out to pave the way towards trouble free multi-cloud service provisioning, envisioning a scenario where users are not just consumers of services and resources but can also play the role of a provider of -in many case otherwise underutilized- resources.
During the project, significant work has taken place to deliver a secure, robust and scalable multi-cloud platform. Towards this end a set of use cases (5 in total) in the broader smart cities/homes area was identified, not only matching the interests and daily business of project partners, but also calling inherently for the joint consideration of IoT, Big Data and Cloud challenges. The ethical and privacy issues that may arise especially in a cross-border setup were also identified and appropriate processes and measures have been put in place.
Following a structured and joint consideration of IoT, Big Data and Cloud requirements, the architecture of the iKaaS platform together with the interfaces specification was developed. The architecture further decomposes the original idea of a multi-cloud environment onto a set of components and their needed functionalities for accommodating the whole lifecycle of a service rolled out using the iKaaS platform. The platform design also supports the notion of "Knowledge as a Service (KaaS)" allowing third parties to become both consumers as well as providers of data (raw or processed) and even functionalities of/to the platform.
The identified components and supported functionalities were further elaborated with their final specification and algorithmic implementation being delivered together with data model components of the unified iKaaS data model. The project also implemented the components (iKaaS toolbox) for use and manipulation in cloud infrastructures in formats widely supported to maximize further uptake and integration of components in iKaaS platform instantiations, which allowed for real-life use case validation.
The real-life validation confirmed that the use of Docker containers (https://www.docker.com/) in conjunction with Kubernetes (https://kubernetes.io/) provides a lightweight yet powerful combination that can support services with diverse requirements in terms of performance, scalability, underlying technologies and algorithmic complexity.
As iKaaS has made its toolbox publicly available at https://gitlab.atosresearch.eu/ari/ikaas_pub and https://hub.docker.com/u/ikaas/, this can be used by stakeholders of similar nature to the iKaaS project partners to roll out services for their citizens (in the case of Public Administrations) or commercial customers (in case of Health Support Service Providers, Town Management Service Providers as well as Technology/Solutions Providers).
It is also worth noting that as many of the iKaaS developed components are very generic in nature, their applicability is not tight strictly to Smart City scenarios but might as well constitute a basis for cloud based deployments in other application domains (e.g. manufacturing, supply chain etc). More details on the use of the developed components can be found in the D4.5 deliverable (all publicly available deliverables can be found under the Impact/Deliverables section of the website).