iSGTW Feature - GRelC at Euro-Mediterranean Centre for Climate Change


Feature - GRelC @ Euro-Mediterranean Centre for Climate Change


RealClimate is a commentary site on climate science, founded by working climate scientists-mostly from Columbia University's Lamont-Dohery Earth Observatory- written for the general public.

Image courtesy of RealClimate

Drowning in data

Talk to any climate scientist, and they'll tell you that one of their biggest problems is dealing with the sheer volume of data they acquire when dealing with something as complex as worldwide climate change. They don't lack for sources of data; if anything, they have too much to choose from-everything from tree ring data to ice cores to collections of air samples from hot-air balloons.

With this in mind, managing the metadata can be a bottleneck. But distributed, peer-to-peer and grid-enabled solutions can help to provide a secure, decentralized, transparent and scalable solution for managing petabytes of datasets spread among several sites, say researchers at the EuroMediterranean Center for Climate Change (CMCC, Italy). They have developed what they call the CMCC Grid Metadata Handling System (GMHS), which provides both access to and integration of climate metadata stored in different and widespread data sources, while allowing scalability, transparency and efficiency.

The middleware for their system is based upon GRelC DAIS, adopted as a data grid enabling technology, developed by the SPACI Consortium and the University of Salento under the Grid Relational Catalog (GRelC) Project. The software, compatible both with gLite and Globus, is part of the IGI release, and was recently included in the EGEE Respect Program and presented at the last two EGEE events, at the EGEE07 conference in Budapest and the 3rd EGEE User Forum in Clermont-Ferrand. It was mentioned as a runner-up during the related demo sessions-in both demos, CMCC researchers Sandro Fiore, Salvatore Vadacca and Alessandro Negro showed practical scenarios involving the distributed climate metadata management at CMCC.

In the CMCC infrastructure, distributed production centers span Italy, including everything from Tier-0 to Tier-2.

Image courtesy of Euro-Mediterranean Center for Climate Change

Supporting science

"The CMCC Data Grid framework," says Giovanni Aloisio, head of the Scientific Computation and Operation Division Head at CMCC, "will play a strategic and active role in the Euro-Mediterranean Center for Climate Change infrastructure since it will support the scientific community in organizing and managing the huge amount of distributed climate datasets, and will provide enhanced metadata search and discovery services leveraging a transparent and scalable P2P/Grid approach. In addition, due to its distributed nature, it represents a perfect incubator for new data grid developments."

In the CMCC infrastructure, the GRelC DAIS will connect several distributed production centers, in Lecce (Tier-0), Bologna, Capua (Tier-1), Sassari, Venice, Milan, Sannio (Tier-2) through a P2P grid framework. It will allow CMCC climate scientists to perform distributed metadata search and discovery activities on the climate datasets that will be produced starting as of September 2008.

A distributed European testbed, exploiting the GRelC DAIS, will be presented at the EGEE '08 conference in Istanbul in September during the poster session.

-Giovanni Aloisio and Sandro Fiore, Euro-Mediterranean Centre for Climate Change