Sites Inria

Version française



Inria at TERATEC 2018

The TERATEC Forum, a major meeting for international experts in HPC, big data and simulation, will take place on 19 and 20 June 2018 at the École Polytechnique. Once again this year, Inria is a partner of the event, where it will present a selection of innovative technologies. Inria will also be a partner of the European Research Café, which aims to showcase European scientific projects and initiatives in these fields.

On the programme

By bringing together more than 1,300 professionals, the TERATEC Forum illustrates the technological and industrial dynamism of HPC and the important role played by France in this field. The participation by - and testimonies of - major industry players, the presentations by the technological companies who are leaders in the field, the diversity and level of the technical workshops, the representativeness of the exhibitors and the innovation of the solutions presented are all strengths that make this an unmissable event for everyone concerned by high performance design and digital simulation.

Isabelle Ryl, interim CEO for Technology Transfer and Industrial Partnerships, will present Inria's main strategic objectives on HPC, big data and simulation. Bruno Raffin, head of the DataMove project team, will then present the new Inria Project Lab HPC - Big Data, whose aim is to study the problems linked to the convergence between HPC, big data and learning.

Organised by TERATEC and Usine Digitale, the Digital Simulation Awards recognise the most innovative simulation and digital technology projects, both within major groups and SMEs. Inria, one of the awards' partners, will present the Innovation Award.

  • 19 and 20 June - Exhibition of the main simulation, HPC and big data players

Among the 70 stands bringing together these major players, Inria will present a selection of its technologies relating to these fields on stand 18 in the forum exhibition area:

  1. Scikit-learn: Scikit-learn is a software library written in Python which provides the building blocks of statistical learning, or 'machine learning'. Scikit-learn is regularly used by 500,000 people around the world. Its simple but powerful predictive models enable the extraction of data comprehension from the effective linear model on random forest texts, well-adapted to heterogeneous databases.  In industrial contexts, the detection of anomalies or reductions in data easily add value. Originating from the PARIETAL project team, Scikit-learn has brought together various partners - in particular industry players - in order to form a consortium around this research project. The official signature is planned for the end of June 2018 and the launch in September 2018.
  2. Presentation of the expertise and software tools around ‘Uncertainty Quantification’ by Pietro Congedo from the DEFI project team: the heart of DEFI's expertise lies in its uncertainty quantification methods for computational engineering, optimisation and inverse problems. The activities of the project team are based around the formulation, implementation and validation of numerical methods in order to use scientific computing to manage the experiments and data available (coming from models, simulation and experiments) by taking into account the uncertainty of the system. Our activities are also based on the internal development of specialised software which, for example, are used as a design tool in the aerospace industry.
  3. Presentation of the Inria Project Lab HPC-Big Data

An Inria start-up founded in 2007 will also be present on stand 26 - ActiveEon, open source software and solutions publisher, specialised in the automation and acceleration of IT processes. Its president, Denis Caromel, will present the plenary session HPC with Hybrid Clouds for Critical Big Compute Workloads .

Inria at the European Research Café

Inria is a member of the European technological platform ETP4HPC (European Technology Platform for High Performance Computing) and contributes, through the presence of its researchers in different working groups, to the development of the platform's research and innovation strategic roadmap.

As a result, the Inria teams are contributing to the European ecosystem of high performance computing research, and maintain numerous collaborations in Europe, in particular within the framework of their participation in around 10 Horizon 2020 programme collaborative projects in the field of HPC and in the PRACE programme, via shareholding and partnership with GENCI.

Discover the European projects ANTAREX, and POP, originating from Inria, at the European Research Café.

Keywords: Teratec IPL HPC Big data Digital simulation