Ressource-Sharing on the Tera-Flop Scale for the Biomedical Research and Care Sector - The Erasmus Computing Grid and MediGRID

Today advances in scientific research as well as clinical diagnostics and treatment are inevitably connected with information solutions concerning computation power and information storage. The needs for information technology are enormous and are in many cases the limiting factor for new scientific results or clinical diagnostics and treatment. At the same time huge computing and storage resources (e.g. ~109 personal computers in the private, public, and industrial domains) have been installed which outwheigh the resources at high-performance computing centers ~50-100 times and thus could contribute to the challenges mankind faces. Both the Erasmus Computing Grid (ECG) and the MediGRID are two major working resource-sharing entities at public funded organizations:

i) The ECG in Rotterdam exploits currently the desktop computers of the Erasmus Medical Center (the biggest biomedical research and hospital center in The Netherlands) and the Hogheschool Rotterdam (one of the biggest city universities in The Netherlands) and consists of ~104 hosts with ~ 10 TeraFlops capacity (30 TeraFlops at the end of 2008). The ECG has grown into a vital part of the work of the ~10 user groups as well as the organizations, with results which could not have been done without the ECG.

ii) The MediGRID being part of the German D-Grid initiatives connects local dedicated cluster resources at biomedical universities throughout Germany with a capacity of ~2000 hosts with ~2 TeraFlops capacity. Within MediGRID ~15 user groups conduct basic research as well as clinical applications for diagnosis and treatment. Again the results could hardly be obtained otherwise and thus have provided breakthroughs.

To build these infrastructures two e-social influences had to be overcome: i) the sharing attitude and socialization of the individual, i.e. the micro-sociality, and ii) the organization culture of the embedding institution, i.e. the macro-sociality, as e.g. for the ECG the public funded organizations. Operationally, an these factors were adressed by: i) the participative integration of fundamental IT applications of major users, and ii) the setup of an open and sustainable management structure.Consequently, we show that the IT challenges mankind faces in the biomedical research and health-care sectors can be successfully approached by appropriate exploitation of the huge existing resources by grid technology combined with micro and macro e-social means to stimulate sharing on the individual as well as organizational level.