CERN Accelerating science

This website is no longer maintained. Its content may be obsolete. Please visit for current CERN information.

CERN home pageCERN home pageDocuments by ReferenceDocuments by ReferenceCNLsCNLsYear 2000Year 2000Help, Info about this page


Editorial Information
If you need help
What our Users Have to Say ...
Announcements Physics Computing Desktop Computing Internet Services and Network Scientific Applications and Software Engineering The Learning Zone User Documentation
Previous:Physics Computing
Next:What is SCADA?
 (See printing version)

DataGrid Project successfully Negotiated with the EU

Fabrizio Gagliardi , IT/DI


Presentation of the DataGrid Project, a collaboration between CERN and several European laboratories.

The European Union FrameWork 5 IST programme has recently successfully concluded the negotiation for a Data Grid research and development project submitted by a consortium of 21 scientific applications (HEP, Earth Observation and Biology), computer sciences and industry. CERN is the leading and coordinator partner. The project will receive in excess of 9.8 Million Euros for three years staring early 2001. The project will develop middleware software necessary to deploy real production test beds in a phased approach over the three-year period.

As far as CERN is concerned, this programme of work will integrate well in the already planned LHC computing test bed activity. Indeed the model of the distributed computing architecture, which DataGrid will implement, is mostly based on the results of the Monarc project. CERN efforts will be essentially integrated into the IT programme of work and jointly staffed by EU and CERN funded personnel.

CERN major contribution will be in WorkPackage 2, dedicated to data management and data replication, WP4, computing fabric management and WP 8, HEP applications. WP8 is coordinated by EP Division, with resources mostly coming from the four LHC collaborations. Other areas of work include workload management (coordinated by INFN), monitoring and mass storage (coordinated by Rutherford laboratory), and testbed and networking (coordinated by IN2P3 and the CNRS).

The data management work package will develop and demonstrate the necessary middleware to assure secure access to Petabyte size data collections via a universal global name space, including efficient high speed data movement between Grid sites with caching and replication of data as appropriate. Strategies will be developed for optimising and costing queries on the data, including the effect of dynamic usage patterns. A generic interface to various mass storage management systems in use at different Grid sites will also be provided. The objective of the fabric management work package is to develop new automated system management techniques that will enable the deployment of very large computing fabrics constructed from tens of thousands of mass market components with reduced systems administration and operations costs. All aspects of management will be covered from system installation and configuration through monitoring, alarms and problem resolution. WP8 aims at deploying and running large scale distributed simulation, reconstruction and analysis programs using the GRID technology. This package is central to the project, as it provides, together with an earth observation work package (coordinated by ESA) and a biology activity (coordinated by the CNRS), testing on a very large scale of the middle-ware developed by the other work-packages and provides the user requirements that drive the definition of the architecture of the project. Several tens of physicists, mostly from Europe, will participate to this effort, while performing their day-by-day research activity in a geographically distributed environment.

CERN is also contributing to the project testbeds and networking with IT resources. CERN is also responsible for overall project management and administration with resources partially funded by the EU.

A project Architecture Task Force has been recently appointed with participants from the relevant Middleware work packages and a representative from the applications. Leading US computer scientists such as Ian Foster and Carl Kesselman are participating also in this task force to guarantee that our developments and similar developments in USA will not diverge. DataGrid is also hosting the first Global Grid Forum in Amsterdam early March next year. This Forum is trying to coordinate Grid activity at a world-wide scale.

More information on the project can be found on the (temporary) DataGrid Project WEB site (

About the author(s): F. Gagliardi is the coordinator of the DataGrid project.

For matters related to this article please contact the author.

Vol. XXXV, issue no 3

Last Updated on Mon Dec 18 13:51:00 GMT+03:30 2000.
Copyright © CERN 2000 -- European Organization for Nuclear Research