Exa4mind – Extreme Analytics for Mining Data spaces

Laying the technological foundations

for future European data analytics and computation

How is

EXA4MIND structured?

EXA4MIND relies on a co-design approach, where technology partners from computing centres and universities and application partners from industry, academia and SMEs design an Extreme Data infrastructure in close collaboration.



Architecture Co-design

This work package supervises the co-design and release process including requirement analysis and continual improvement of the architecture.
With its activities, including teleconferences and workshops as needed, it serves as a communication platform between work packages laying the technical foundations (WP2 and WP3) and the Application Case Work Packages (WP4, WP5 and WP6).



Distributed Data Space Management

WP2 is responsible for Distributed Data Management. On the one hand, it generates efficient interfaces between Database Management System (DBMS) and HPC ecosystems for data processing on some of Europe’s largest supercomputers, and on the other hand it creates interfaces between DBMS and the FAIR and European data world, such as EUDAT, EOSC, Gaia-X or European Data Spaces.



Extreme Data Analytics & Processing

WP3 is responsible for the optimisation of database usage in an innovative sense.

Bringing together significant expertise in data analytics and AI, this work package develops tools for data preprocessing and harmonisation as well as the Advanced Query and Indexing System (AQIS) of the Extreme Data Database. It optimises data analytics techniques used in the Application Cases and provides user interfaces to the AQIS for semi-automatised data analytics for special applications.



Scientific Use Case

This work package, led by IT4I@VSB Modelling for Nanotechnologies Lab, will supervise the systematic improvement of molecular dynamics simulations and accuracy.
The main objective of this work package is to set up an Extreme Data Mining System, called ADAMS4SIMS, using the EXA4MIND Extreme Data Database.



Industry Application Case

This work package, led by Valeo, will investigate automated annotation & evaluation of automotive camera recordings.
The ultimate goal of this work is to develop methods and technologies for Advanced Driver Assistance Systems, which include AI models trained to annotate the data avoiding tedious work and allowing smart queries based on weather, speed, location, etc.



SME Application Case
For this work package Alternativ and Terraview will define specifications and integration plans applied to smart farming/viticulture data processing (sensor, satellite and public data), and secure health data mining through big data systems. The key objective of work package 6 is to prove, emphasise and maximise the benefits of the EXA4MIND Extreme Data Database platform for small-medium enterprises.



Outreach, Exploitation & Collaboration

This work package fast-tracks the transfer of knowledge between the project and its stakeholder base, funneling the inwards/outwards value streams with the scientific and industrial communities, while capitalising on the results obtained.



Project Management

WP8 provides financial, administrative and reporting capabilities. This WP is also responsible for quality control, the Data Management Plan, and collaborative tools for working within the consortium.



Ethics Requirements

WP9 sets out the ‘ethics requirements’ that the project must comply with. The Ethics Advisor -an external independent person specialised in EU data protection law and the ethics of AI- provides expertise and recommendations on topics related to data identification, data minimisation, and the legal requirements for data processing, including for the four application cases.