BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20141116T201500Z
DTEND:20141116T203000Z
LOCATION:288-89
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Steady improvements in computing resources enable ever more enhanced simulations, but data input/output (I/O) constraints are impeding their impact. Despite increases in temporal resolution, the gap between time steps saved to disk keeps increasing as computational power continues to outpace I/O capabilities. Consequently, we are seeing a paradigm shift away from the use of prescribed I/O frequencies and post-process-centric data analysis toward a more flexible concurrent paradigm in which raw simulation data is processed as it is computed. This talk will survey applied mathematics and computer science research challenges introduced by this paradigm shift and highlight active research efforts to deal with these issues.
SUMMARY:Extreme-Scale In-Situ Data Analysis
PRIORITY:3
END:VEVENT
END:VCALENDAR