BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160903Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess324_post195@linklings.com
SUMMARY:Improving the I/O Performance and Memory Usage of the Xolotl Clust
 er Dynamics Simulator
DESCRIPTION:Poster\nTech Program Reg Pass, Exhibits Reg Pass\n\nImproving 
 the I/O Performance and Memory Usage of the Xolotl Cluster Dynamics Simula
 tor\n\nRoth, Blondel, Bernholdt, Wirth\n\nXolotl is a cluster dynamics sim
 ulator used to predict gas bubble evolution in solids. It is currently bei
 ng used to simulate bubble formation in the plasma-facing surface within f
 usion reactors and the nuclear fuel used in fission reactors. After observ
 ing performance problems in coupled-code simulations of fusion reactors, w
 e used Xolotl's built-in performance data collection infrastructure and an
  external profiling tool to identify inefficiencies when writing Xolotl's 
 two types of checkpoint files. We changed the code to use true parallel wr
 ites via the HDF5 data management library, resulting in a code that is app
 roximately 57x faster when writing the program's main checkpoint file at t
 he scale used in the coupled-code simulations, and that exhibits less perf
 ormance variability due to external system activity. We also identified an
 d addressed a memory usage problem that reduced Xolotl peak memory usage b
 y approximately 88% per compute node.
URL:https://sc18.supercomputing.org/presentation/?id=post195&sess=sess324
END:VEVENT
END:VCALENDAR