Section: Overall Objectives
Experimental Grid Infrastructures
The Paris Project-Team is engaged in research along five research topics: Operating System and Runtime for Clusters and Grids , Middleware for Computational Grids , Large-scale Data Management for Grids , Advanced Models for the Grid and Experimental Grid Infrastructures . The concepts proposed by each of these topics must be validated against real applications on realistic hardware. The project-team manages a computation platform dedicated to operating system and middleware experimentations. This platform is integrated within Grid 5000 , a national computing infrastructure dedicated to large-scale Grid and peer-to-peer experiments. The Grid 5000 infrastructure federates experimental platforms (currently 9 platforms) across France. These platforms are connected through Renater using dedicated Gigabit Ethernet links.
Our experimental platform is heterogeneous: PowerPC and PC families of processors, 32-bit and 64-bit architectures, Linux and Mac OS X operating systems. Various high-performance interconnection technologies such as Myrinet and InfiniBand are available on groups of nodes. Heterogeneity allows realistic validation of interoperability of middleware and P2P systems. On the other hand, our platform is composed of sufficiently large groups of homogeneous computation nodes: 66 dual Xeon, 166 dual Opterons, 33 Xserve G5. This allows to evaluate the scalability of operating systems, runtimes and applications on various architectures.
Our experimental platform is dedicated to operating system and middleware experimentation: it is possible to repeat experiments in the same environment (same machines, same network, etc.). The allocation of the resources to the experiments is handled through GridPrems , a collaborative resource manager developed in our group and through OAR , a job manager developed by the Grenoble partner group of Grid 5000 .