BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:2.0
BEGIN:VEVENT
DTSTART:20141117T143000Z
DTEND:20141117T151500Z
LOCATION:288-89
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Exascale computing will enable combustion simulations in parameter regimes relevant to next-generation combustors burning alternative fuels needed to provide the underlying science base for developing predictive combustion models and the analysis of massive volumes of data.  It will enable combustion research with complex fuels in high-pressure, turbulent environments associated with IC engines and gas turbines for power generation.  The current practice at the petascale of writing out raw data from simulations to persistent disk and re-reading in the data later for analysis won’t scale at the exascale due to the disparity among computational power, machine memory and I/O bandwidth. It is anticipated that much of the current analysis and visualization will need to be integrated with the combustion simulation in situ in a comprehensive overall work flow on an exascale machine. In addition to understanding the specific performance characteristics of individual solver, analysis and visualization algorithms, we need to understand the overall workload and impact of various designs in the context of the end-to-end workflow.  Finally, programming models and runtimes that optimize performance for complex in situ workflows will be discussed.
SUMMARY:Big Data and Combustion Simulation
PRIORITY:3
END:VEVENT
END:VCALENDAR