Espri

Ensemble des Services pour la Recherche à l'IPSL

What is ESPRI ?

The IPSL Computing and Data Centre (ESPRI) was orn out from several initiatives at the Institut Pierre-Simon Laplace (IPSL) and its laboratories, wishing to share IT resources and joint projects through a numerical facility for research:

  • the ClimServ data centre for the calculation, archiving and distribution of satellite and long-term data at the École Polytechnique,
  • the Ether data and service centre for atmospheric chemistry,
  • the CICLAD calculation server at the Sorbonne University (previously Pierre et Marie Curie University) to which were attached the services related to the storage, analysis and distribution of climate simulations.

A computing and storage infrastructure, including:

  • Different types of storage spaces,
  • CPU and GPU computing servers,
  • Virtualisation products,
  • High-speed network infrastructure,
  • Scientific computing and analysis environment/softwares.

Data management services, including:

  • Data acquisition or production,
  • Data access and processing,
  • Data curation and archiving,
  • Data documentation,
  • Data distribution and publication.

Thus, ESPRI’s activities cover various fields to support multidisciplinary projects carried out at IPSL. It organises, for the last 20 years, access for IPSL laboratories and their partners to different reference data sets and high-performance computing resources. Thus ESPRI plays a transverse role as a shared, multi-site and multi-thematic platform, recognized by the Institut National des Sciences de l’Univers (INSU) from the Centre National de la Recherche Scientifique (CNRS) and by Sorbonne University.

Teams

ESPRI engineers ensure the harmonization of the development of this mutualized and multi-site data analysis platform and work in close relationship with scientific teams, allowing them to understand and efficiently take into account their requirements. They identify and implement new technologies on prototypes and select the best operational solutions to be deployed.

The ESPRI staff is organised into three teams:

  • ESPRI-Infra, in charge of the infrastructure itself (workstations, servers, disks, network, and software) and user support,
  • ESPRI-Obs, which offers expertise and services around observation data,
  • ESPRI-Mod, which offers expertise and services around modelling data.

Governance

The governance of ESPRI is organised at different levels:

  • An executive board, composed of the leaders of ESPRI teams, meets monthly to coordinate the ESPRI roadmap, resources and to monitor transverse actions.
  • A technical committee, composed of all ESPRI engineers, quarterly meets to share expertise, best practices, methodologies, tools and to create synergies.
  • A steering committee, in which the centres, the IPSL management and the institutional partners are represented, meets once a year. This committee is also composed of key users.

Roadmap

Several thematic working groups (WG) have been created to design and focus on three major axes for the ESPRI roadmap:

Axis 1: Enhancing the value of environnemental data produced by IPSL and its partners

  • Theme 1.1: Certification of data warehouses and DOI service (Data Object Identifier)
  • Theme 1.2: Improving IPSL data catalogs
  • Theme 1.3: Website

Axis 2: Mobilising new technologies and training users

  • Theme 2.1: Building relevant Virtual Research Environments (VRE)
  • Theme 2.2: Automate processing chains
  • Theme 2.3: Monitoring and metrics

Axis 3: Broadening the fields of expertise in data management

  • Theme 3.1: Developing Geographic Information Systems (GIS) standards at IPSL
  • Theme 3.2: Coordinating infrastructures and services around artificial intelligence IPSL initiatives though ESPRI-IA services
  • Theme 3.3: Collecting and coordinating the production of new energy data sets through the E4C DataHub

ESPRI Community Infrastructure

As ESPRI provides a numerical infrastructure dedicated to climate data and services, it is aimed at the wider climate science community.

Acknowledgement

Data or Computing resources from ESPRI should be cited in any article or presentation making use of the data or computing facility.