Periodic Reporting for period 1 - EDGELESS (Cognitive edge-cloud with serverless computing)
Período documentado: 2023-01-01 hasta 2024-06-30
The EDGELESS consortium is working toward three goals.
• Design and implement an event-driven architecture able to manage applications’ dynamic state and scalability. This requires methods to efficiently redistribute computation from user devices and remote datacentres across a diverse infrastructure ecosystem, including edge nodes with heterogeneous computational capabilities. By leveraging an event-driven architecture, the stateful FaaS framework of EDGELESS enables seamless communication and coordination among distributed functions. This promotes efficient handling of events and triggers, enhancing the system’s responsiveness.
• Security and privacy concerns are first-class considerations in EDGELESS. Encryption, access controls, and other security measures are employed to ensure that sensitive data is protected and user privacy is maintained even in a distributed edge-cloud environment. To further enhance privacy and security, EDGELESS will implement secure authentication based on Hardware Security Modules and secure stateful FaaS workflow execution in trusted enclaves.
• Longevity, sustainability, and desirability of the EDGELESS project arises from alignment between developments and a value proposition for edge computing ecosystem actors applying the EDGELESS framework, enabling new use and business cases. On the one hand, the framework emphasises ease of use and improved developer experience by applying standard-compliant and maintainable APIs, and state-of-art tools for the different actors interacting with EDGELESS, e.g. graphical, low-code interfaces. On the other hand, EDGELESS has, from the start, adopted an open-source approach, to facilitate collaboration among developers, researchers, and industry experts, fostering innovation and sustainable development of the EDGELESS cutting-edge FaaS framework.
• The refined definition of the project requirements and the technology gap analysis, the application programming model, and the EDGELESS architecture.
• The initial implementations of the key platform components and an early integration of those components into a functioning prototype of the platform.
• The refined definition and implementation of the EDGELESS services, the reference implementation of a baseline orchestrator (ε-ORC), which realizes the key interactions with the other layers: ε-CON at a cluster level and nodes within its orchestration domain.
• Definition and set-up of the Continuous Integration and Continuous Deployment (CI/CD) approach, tools and processes, description of the three Use Cases, their objectives and requirements.
• Refinement of our dissemination strategy, enhancement of the project’s digital presence, regular newsletters and social media updates, collaborative initiatives with other projects and programs, participation and organization of events.
• Initial definition of the overall exploitation strategy for the project’s results, and identification of key stakeholders, business values and related value proposition.
During the next period most of the work will be focused on the finalization of the intermediate implementation of the EDGELESS platform, and then the final development tasks of all components and their integration into the full implementation of the EDGELESS platform, as well as the implementation of the EDGELESS use cases.
• Real-time data distribution. Transparent management of large-scale deployments of small devices.
• Physically isolated security module. Add security to devices with very limited capabilities.
• Universal orchestrator. Take orchestration beyond containers.
• SLA manager. Seamless manage QoS across multiple serverless domains Internal developments, infrastructure set up, training of marketing teams.
• Anomaly detection. Use distributed AI to detect network/load anomalies in real-time.
• Data-centric function middleware. Low-code development with 1:1 mapping to deployment.
• Smart city surveillance system. High accuracy detection of events with less resources.
• Activity identification and anomaly detection. More accurate and highly personalised detection of health-related events.
After conducting an in-depth impact study on technical aspects and results by project partners, the results of the analysis show that all results of EDGELESS have high impact on at least one of the partners after the end of the project, and each technical result has medium impact on more than 50% of the partners. This strongly emphasises the right selection of target results with respect to potential exploitation of the project results and the sustainable developments addressed in the EDGELESS project.