Large scale problem solving using automatic code generation and distributed visualization
Scientific computation faces multiple scalability challenges in trying to take advantage of the latest generation compute, network and graphics hardware. We present a comprehensive approach to solving four important scalability challenges: programming productivity, scalability to large numbers of processors, I/O bandwidth, and interactive visualization of large data. We describe a scenario where our integrated system is applied in the field of numerical relativity. A solver for the governing Einstein equations is generated and executed on a large computational cluster; the simulation output is distributed onto a distributed data server, and finally visualized using distributed visualization methods and high-speed networks. A demonstration of this system was awarded first place in the IEEE SCALE 2009 Challenge.
Publication Source (Journal or Book title)
Hutanu, A., Schnetter, E., Benger, W., Bentivegna, E., Clary, A., Diener, P., Ge, J., Kooima, R., Korobkin, O., Liu, K., Löffler, F., Paruchuri, R., Tao, J., Toole, C., Yates, A., & Allen, G. (2010). Large scale problem solving using automatic code generation and distributed visualization. Scalable Computing, 11 (2), 205-220. Retrieved from https://digitalcommons.lsu.edu/physics_astronomy_pubs/1437