• English 
  • Spanish 

GRID computing in Earth Sciences

logo_grid.jpg Short description: Development and deployment of earth science applications (e.g., global and regional climate models) to run in geographically distributed data and computing environments (GRID computing).

The GRID provides transparent access to geographically distributed computational and storage resources. Several applications areas as high energy physics or bio-applications have proven to benefit from this computational paradigm. Applications from the Earth Science community are starting to take advantage of this technology (see e.g. www.eu-degree.eu).

Earth Science (ES) applications and, in particular, weather and climate models, are nowadays among the most computer-power demanding tasks. Computers at different national weather services can often be found at the top 10 and most of the top general-purpose supercomputers run ES-related parallel applications. Moreover, a number of tasks such as ensemble prediction, sensitivity analysis, etc., consist of running the same application many times with small modifications in the configuration parameters, thus requiring an appropriate production infrastructure. GRID computing consists of a geographically distributed computing and data infrastructure linking computer resources (clusters) around the world in a transparent way using appropriate middleware. This infrastructure allows parametric and parallel computation in a simple form.

The figure below shows an illustrative example of sensitivity analysis where the same application (the CAM atmospheric global model) is run several times from slightly perturbed Sea Surface Temperature (SST) conditions. Appropriate execution workflow programs control the computation of the experiment within GRID by restarting interrupted jobs, managing the input and output data, etc. (see the EELA CAM-demo for more details).

CAMgrid.jpg

In the last two decades, a number of other computer-demanding applications, such as High Energy Physics and Biomedicine applications, have migrated towards GRID tecnologies as a complementary way to fulfill their increasing CPU power and storage requirements. Thus, during the last few years a lot of effords has been made in order to have an European production GRID infrastructure for e-Science. The EGEE Project, the most important iniciative in Europe has 70 European partners and over 10000 CPUs and approximately 10 Petabytes of storage.

Key reading:

  • The GridCafe portal is a good place to get a brief introduction about the GRID technollogy and its main components.
  • The Anatomy of the Grid: Enabling Scalable Virtual Organizations . Defines Grid computing and the associated research field, proposes a Grid architecture, and discusses the relationships between Grid technologies and other contemporary technologies.
  • The Data Grid: Towards an Architecture for the Distributed Management and Analysis of Large Scientific Datasets Describe two basic services that are fundamental to the design of a data grid, namely, storage systems and metadata management.

    Activities of the Santander Meteorology Group:

  • We have participated in the following european and national GRID initiatives: crossgrid, EELA, EELA2, EGEE, DEGREE, Climate-G
  • The group has succesfully ported to GRID Earth Science applications: CAM (Global model), WRF (Regional model) and WAM wave model (work in progress).
  • Experience with execution workflow and data management in GRID (very important for almost all the Earth Science applications).
  • Study of the different GRID middleware and development of GRID services.
  • Data mining.
  • Our computing cluster is part of the EGEE infrastructure and the group is involved in the EGEE Earth Sciences Research (ESR) Virtual Organization.

    This research topic has his own resources (computing infrastructure) used by the members of the group to test and tune the applications and midlleware before running them in other production resources.

    UC Grid Application Support Centre:

  • Provide assistance to current and future grid users and application developers during the process of porting their application to run in a GRID enviroment.
  • Grid Application Support Centre certified by the EELA-2 European project and the Spanish e-Science Network .

    Gente: , , , , ,
    • Palabras clave: