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:20260522T150117Z
LOCATION:C141/143/149
DTSTART;TZID=America/Chicago:20181115T163000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess186_pap206@linklings.com
SUMMARY:A Year in the Life of a Parallel File System
DESCRIPTION:Glenn K. Lockwood (Lawrence Berkeley National Laboratory), Sha
 ne Snyder (Argonne National Laboratory), Teng Wang and Suren Byna (Lawrenc
 e Berkeley National Laboratory), Philip Carns (Argonne National Laboratory
 ), and Nicholas J. Wright (Lawrence Berkeley National Laboratory)\n\nI/O p
 erformance is a critical aspect of data-intensive scientific computing.  W
 e seek to advance the state of the practice in understanding and diagnosin
 g I/O performance issues through investigation of a comprehensive I/O perf
 ormance data set that captures a full year of production storage activity 
 at two leadership-scale computing facilities.  We demonstrate techniques t
 o identify regions of interest, perform focused investigations of both lon
 g-term trends and transient anomalies, and uncover the contributing factor
 s that lead to performance fluctuation.\n\nWe find that a year in the life
  of a parallel file system is comprised of distinct regions of long-term p
 erformance variation in addition to short-term performance transients.  We
  demonstrate how systematic identification of these performance regions, c
 ombined with comprehensive analysis, allows us to isolate the factors cont
 ributing to different performance maladies at different time scales.  From
  this, we present specific lessons learned and important considerations fo
 r HPC storage practitioners.\n\nTag: Architectures, Data Management, File 
 Systems, Networks, State of the Practice, System Software, Workflows\n\nRe
 gistration Category: Tech Program Reg Pass\n\nSession Chair: Robin J. Gold
 stone (Lawrence Livermore National Laboratory)\n\n
END:VEVENT
END:VCALENDAR
