IGTMD

Grids are high performance and large scale distributed computational and storage systems, used by large user communities. Today many operational grids and grid middleware exist like UNICORE , ARC , EGEE/LCG/gLite/NorduGrid , Globus , Condor , SRB. The emergence and the important deployment of different middleware raise the interoperability problem.
These software environments propose global services for job management, data management, access control, status information collect and retrieval. The Global Grid Forum develop standard like OGSA to make these services interoperable, but the set up of the global interoperability requires more research works. On another hand, no standard has been defined for bulk and reliable data transfers in grids.

General Information

Reference: IGTMD
Source of funding: Agence Nationale pour la Recherche, Programme Blanc / STIC
Project Duration: November 1, 2006 - October 31, 2008
Official Project Home Page:
To be defined

Project Overview

The aim of this project is to design, develop and validate mechanisms that concretely make the interoperability of heterogeneous grids a reality. The project concentrates on the following topics:

  1. Bulk data transfers
  2. Replication and referring mechanisms
  3. Information system and job management interoperability
  4. Grid control and monitoring
  5. Usage of statistics and accounting data.


A particular emphasis will be put on disk to disk bulk data transfers over very long distance with optimal performance. The key idea is to fully exploit the specificity of LCG applications and their real infrastructures to analyse and experiment new communication and replication models, alternative transport protocols emerging within the international scientific community. The participation to a standardization process for a generic grid transport service for bulk exchanges between heterogeneous grids will be a strong goal of the project.
Despite the fact that the interoperability and the unification of a generic data transport in Grids are very often perceived as a necessity, they are in fact very little studied. The present project would allow France to get a leading position in this computing area that will be absolutely crucial to insure the LHC data exploitation. The very strong experience of the partners in deployment and exploitation of international research and production computing instruments gives a promising perspective to this project and its ambitious experimental approach.

Project Partners

Personnel

Publication List

Contact


Dr. Pascale Primet Vicat-Blanc INRIA Rhone-Alpes ENS Lyon, LIP, projet RESO 46 allée d'Italie 69364 Lyon Cedex 07 France


Last modified: Wed Jul 11 15:16:21 Paris, Madrid 2007