Periodic Reporting for period 1 - GLACIATION (Green responsibLe privACy preservIng dAta operaTIONs)
Berichtszeitraum: 2022-10-01 bis 2024-03-31
As edge technologies mature, edge applications will be increasingly deployed for a dynamically changing, distributed and growing volume and heterogeneity of data, users, devices and infrastructure. An optimisation of local computation services is crucial for minimising costs and latencies while ensuring reliability, security and privacy. Energy consumption, latency and reliability critical applications will drive data capturing, computation and distribution towards the edge in close proximity of data producers and data consumers. Current edge storage optimisations are insufficient, often based on traditional cloud-based perspectives and lacking consideration for the efficiency of data movement and placement in a distributed infrastructure, as well as the relevancy of data throughout its lifecycle.
The GLACIATION project aims to enhance the efficiency and use of trustworthy digital technologies to meet the demands of citizens, industry, and public administrations concerning privacy, commercial and administrative confidentiality, as well as responsible, fair, and environmentally friendly data operations across the data lifecycle. To achieve this goal, GLACIATION aims to develop a novel Distributed Knowledge Graph (DKG) spanning the edge-core-cloud architecture and reducing energy consumption for data processing through Artificial Intelligence (AI) enforced minimal data movement operations.
The GLACIATION project aims to enhance the efficiency and use of trustworthy digital technologies to meet the demands of citizens, industry, and public administrations concerning privacy, commercial and administrative confidentiality, as well as responsible, fair, and environmentally friendly data operations across the data lifecycle. To achieve this goal, GLACIATION aims to develop a novel Distributed Knowledge Graph (DKG) spanning the edge-core-cloud architecture and reducing energy consumption for data processing through Artificial Intelligence (AI) enforced minimal data movement operations.
To achieve these goals, GLACIATION followed a structured approach made up of three phases: (i) Early Platform and Basic Component, (ii) Complete Platform and Completed Components and (iii) Complete Platform and Advanced Components. During the first eighteen months of the project, the consortium focused on the first phase, and it is currently in the midst of the Complete Platform and Completed Components phase. Until now, the main activities developed include defining software and hardware requirements, researching the state of the art, including industry standards, defining benchmarks, and roadmaps. Other relevant activities include also the design of prototypes, the conduction and integration of technical and non-technical tasks, and the definition and execution of lab testing procedures.
More concretely, the project successfully submitted 14 out of 25 deliverables during the covered period, including technical reports, prototypes, research findings, and dissemination and exploitation strategies. Additionally, 5 out of 9 milestones were achieved, namely the initiation of scientific and technical work in each task and use case, the consolidation of the reference architecture, and the initiation of lab tests on sample of methods and use case data.
More concretely, the project successfully submitted 14 out of 25 deliverables during the covered period, including technical reports, prototypes, research findings, and dissemination and exploitation strategies. Additionally, 5 out of 9 milestones were achieved, namely the initiation of scientific and technical work in each task and use case, the consolidation of the reference architecture, and the initiation of lab tests on sample of methods and use case data.
GLACIATION’s aspiration goes beyond the state of the art in several aspects. The main advancements that will be brought by the project can be summarised as follows.
• A mechanism to deploy and operate a knowledge graph in a distributed cloud/edge/core environment, facilitating interoperability and supporting key platform objectives such as AI-enabled data movement, data privacy enforcement, and data analytics and visualisation.
• An AI/ML-enhanced workload orchestration engine to minimise energy consumption and data transfer latency to maximize processing performance across edge/cloud services.
• A controlled sharing of data within the GLACIATION platform across a range of services and providers while maintaining and enforcing data privacy requirements. The core component providing such controlled sharing ability will consist of a data agnostic policy language which annotates the privacy requirements of data.
• A mechanism to deploy and operate a knowledge graph in a distributed cloud/edge/core environment, facilitating interoperability and supporting key platform objectives such as AI-enabled data movement, data privacy enforcement, and data analytics and visualisation.
• An AI/ML-enhanced workload orchestration engine to minimise energy consumption and data transfer latency to maximize processing performance across edge/cloud services.
• A controlled sharing of data within the GLACIATION platform across a range of services and providers while maintaining and enforcing data privacy requirements. The core component providing such controlled sharing ability will consist of a data agnostic policy language which annotates the privacy requirements of data.