Nowadays there are more than 43 billion devices connected to the Internet, from IoT ones up to bigger cloud and edge servers, all of them generating, collecting and sharing data. The interconnection of these devices, even allowing human interaction, is creating a new era where more intelligent and self-adaptive environments are needed, being able to understand user needs and react accordingly. In this context, ICOS is developing the next generation of meta-operating systems, offering a platform for managing the applications’ lifecycle in a more secure, smart and efficient way in a continuum (cloud-edge-IoT, and vice versa) environment.
Following its principles of openness, ICOS has released its Beta version (fully available in its GitHub public repository: https://github.com/icos-project), with the corresponding documentation and instructions needed to download, install and test its functionalities.
ICOS modular platform is structured into four main independent layers, that can be combined to cover specific customer needs. There is also the possibility to make use of an additional artificial intelligence (AI) marketplace connected to other registries. Additionally, the platform can be used as a fully integrated stack, benefiting from all ICOS functionalities.
In general lines, each layer can be described as follows:
Data Layer
It is a single data platform to facilitate the management of data distributed across the continuum, able to seamlessly run on heterogeneous and possibly small devices.
New functionalities included in the Beta release are:
Security Layer
It guarantees the security of users, resources and applications at all times, considering security and trustworthiness as key fundamental pillars.
New functionalities included in the Beta release are:
Meta-kernel Layer
It is an adaptation engine for monitoring, scheduling and orchestrating workloads along the continuum. It can be considered as the core layer of the meta-operating system.
New functionalities included in the Beta release are:
Intelligence Layer
It consists of a set of techniques for optimizing and pruning machine learning models to be adapted to the characteristics of edge devices without losing accuracy.
New functionalities included in the Beta release are:
AI Marketplace
ICOS AI online model catalogue connected to intelligence model registries, using the Intelligence Layer, to create a community of projects using and building intelligent applications that can be managed using ICOS platform.
Full Stack
The platform offers a complete set of features and functionalities easing the IoT-edge-cloud continuum management, including all the abovementioned layers.
This project has received funding from the European Union’s HORIZON research and innovation programme under grant agreement No 101070177.