Cerfacs Enter the world of high performance ...

IT management and user support

Beyond standard computing frames (mail, web, accounting systems …) CERFACS’  research works rely on availability and intensive use of High Performance Computing.

  • computers allowing to run the numerical simulations developed by the teams,
  • high storage capacities housing the simulations’ results,
  • pre and post-processing services that prepare and ease the analysis of obtained results,
  • fast interconnection networks for both internal and external exchanges between the different computing and storage resources.

Today, CERFACS hosts 2 clusters for a global peak of about 600 Tflops/s, a primary space disk of 750 Terabytes inside the clusters and a secondary space disk of 1.2 Petabytes hosted on a EMC NAS server.

All these servers are interconnected on a backbone at 10 Gb/sec. Accesses to our partners’ computers (Météo-France and CCRT), and to national resources for research (Genci) and international ones (Prace and Incite) enlarge our global simulation capacities.

Head of I.T. Management :

Nicolas Monnier

HPC software engineer :

Isabelle d’Ast

System engineers :

Fred Blain

Gérard Dejean

Fabrice Fleury

Patrick Laporte

Engineer :

Gabriel Jonville

 

 

 

NEWS

Continuity of activity of the Cerfacs during the Covid-19 pandemic

superadmin |  20 March 2020

On Monday 16 March 2020, in the context of the rapidly evolving COVID-19 epidemic, Cerfacs decided to reorganize its activities by implementing a Business Continuity Plan (BCP) and deploying teleworking facilities for all its employees. All staff members thus continue to carry out their full mission.Read more


A fiery wakeup call for climate science

superadmin |  26 February 2020

The extent of the recent wildfires in Australia significantly exceeded the projections of any member of the multi-model CMIP archive.  This highlights how current multi-model ensembles may be under-representing the risks of natural disasters under climate change.  Limited coupled system process representation in most models coupled with a lack of parameter uncertainty exploration means that some risks are not explored by the existing international multi-model framework.  This calls for a reassessment of how to focus climate model development on providing robust risk quantification for those impacts which most directly affect society. Sanderson, B.M., Fisher, R.A. A fiery wake-up call for climate science. Nat. Clim. Chang. (2020) nature.com Media coverage BBC Sydney Morning Herald The Guardian Wired The Daily Express YahooRead more

ALL NEWS