The IDRIS (Institut du Développement et des Ressources en Informatique Scientifique = Institute for Development and Resources in Scientific Computing), founded in November 1993, is the CNRS national center for high-performance computing (HPC) and artificial intelligence (AI) using deep learning. It serves the scientific communities of public or private research (provided open research with publication of results), relying on extreme computing.
Both a computing resource center and a competency cluster in HPC and AI, IDRIS is a CNRS support and research unit (UAR 851). It reports to the Open Research Data Directorate (DDOR) of the CNRS and is administratively attached to the CNRS Computer Sciences institute (formerly INS2I), though its role within the CNRS is multidisciplinary. The operating procedures of IDRIS are similar to those of the IR* ("star" Research Infrastructures) of the Ministry of Higher Education and Research (MESR).
The main objective assigned to IDRIS is to contribute as effectively as possible to the excellence of scientific research in the fields of modeling, HPC, and AI.
Furthermore, IDRIS has acquired recognized expertise on a European scale and is currently involved in the MINERVA and AI Factory France projects.
Jean Zay is the national supercomputer purchased by GENCI and operated by the CNRS at IDRIS. It consists of an HPE SGI 8600 partition and an Eviden BullSequana XH3000 partition for a total peak performance of 125,9 PFlop/s.
Made available to scientific research communities relying on intensive computing, the configuration of Jean Zay extends the traditional usage modes of high-performance computing (HPC) to new applications for artificial intelligence (AI).
The complete hardware and software characteristics of the Jean Zay supercomputer are detailed here.
IDRIS offers training courses in high-performance computing (HPC) and artificial intelligence (AI). These courses are aimed at anyone wishing to deepen their knowledge in these fields, whether they belong to the scientific community or corporate personnel.
The courses combine theoretical presentations and practical sessions, allowing participants to directly apply the acquired skills. They cover a wide range of topics, ranging from introductory courses to artificial intelligence to specialized training on parallelizing HPC and AI codes, for optimal use of supercomputers.
Your opinion matters!
To give your feedback, report an error, or suggest an improvement, click here:
quick anonymous questionnaireThis questionnaire is temporary and will take less than a minute, so take the opportunity!