Section: New Results

Distributed Shared Memory approach for the steering of parallel simulations

As a different approach of EPSN, we recently propose in the thesis of J. Soumagne an in-situ visualization approach for parallel coupling and steering of simulations through distributed shared memory files (DSM). Indeed, as simulation codes become more powerful and more interactive, it is desirable to monitor a simulation in-situ, performing not only visualization but also analysis of the incoming data as it is generated. Monitoring or post-processing simulation data in-situ has obvious advantage over the conventional approach of saving to – and reloading data from – the file system; the time and space it takes to write and then read the data from disk is a significant bottleneck for both the simulation and subsequent post-processing steps. Furthermore, the simulation may be stopped, modified, or potentially steered, thus conserving CPU resources.

In this thesis, we propose a loosely coupled approach that enables a simulation to transfer data to a visualization server via the use of in-memory files. We show in this study how the interface, implemented on top of a widely used hierarchical data format (HDF5), allows us to efficiently decrease the I/O bottleneck by using efficient communication and data mapping strategies. For steering, we present an interface that allows not only simple parameter changes but also complete re-meshing of grids or operations involving regeneration of field values over the entire computational domain to be carried out. This approach is generic enough so that no particular knowledge of the underlying model is required and a user can therefore plug any simulation to this framework without any re-compilation work.

A scalability study have demonstrated the performance of this solution up to 2048 cores on a Cray machine. Finally, the environment has been validated on two industrial test cases: the first one is developed by Ecole Centrale de Nantes and HydrOcean and an object placed into a wave maker is dynamically modified and steered, thereby making use of the re-meshing capabilities introduced by the framework; and the other is developed by Ecole Centrale de Lyon and ANDRITZ HYDRO, a Pelton turbine is dynamically controlled and results are analyzed in-situ.

This thesis has been defended by J. Soumagne in december 2012. This work was supported by NextMuSE project receiving funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement 225967. It has been realized in collaboration with the Swiss National Supercomputing Centre (CSCS). J.