BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:2.0
BEGIN:VEVENT
DTSTART:20141118T193000Z
DTEND:20141118T210000Z
LOCATION:383-84-85
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Since its introduction in 1970, Unix (including Linux) has been a mainstay in scientific computing.  The use of Linux has many advantages, but also is a drag on innovation: Since Linux is used on a broad range of platforms, changes that are specific to HPC are accepted very slowly, if at all, by Linux maintainers. In addition, very little work has focused on the “global operating systems”, i.e., the software system infrastructure needed to manage the entire machine and support multi-node parallel applications. Current Linux software will be stretched in dealing with future nodes with hundred of heterogeneous cores and heterogeneous memory; it will continue to provide little support to the integration of local OS’es into a global OS.  A panel composed of HPC OS/R experts will consider the technical, research, and socio-economic challenges that need to be addressed for future, extreme-scale operating systems.
SUMMARY:Changing Operating Systems is Harder than Changing Programming Languages
PRIORITY:3
END:VEVENT
END:VCALENDAR