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:20181221T160728Z
LOCATION:D163
DTSTART;TZID=America/Chicago:20181112T155500
DTEND;TZID=America/Chicago:20181112T162000
UID:submissions.supercomputing.org_SC18_sess142_ws_pdsw109@linklings.com
SUMMARY:Toward Understanding I/O Behavior in HPC Workflows
DESCRIPTION:Workshop\nI/O, Storage, Workshop Reg Pass\n\nToward Understand
 ing I/O Behavior in HPC Workflows\n\nLuettgau, Snyder, Carns, Wozniak, Kun
 kel...\n\nScientific discovery increasingly depends on complex workflows c
 onsisting of multiple phases and sometimes millions of parallelizable task
 s or pipelines. These workflows access storage resources for a variety of 
 purposes, including preprocessing, simulation output, and postprocessing s
 teps. Unfortunately, most workflow models focus on the scheduling and allo
 cation of computational resources for tasks while the impact on storage sy
 stems remains a secondary objective and an open research question. I/O per
 formance is not usually accounted for in workflow telemetry reported to us
 ers.\n\nIn this paper, we present an approach to augment the I/O efficienc
 y of the individual tasks of workflows by combining workflow description f
 rameworks with system I/O telemetry data. A conceptual architecture and a 
 prototype implementation for HPC data center deployments are introduced. W
 e also identify and discuss challenges that will need to be addressed by w
 orkflow management and monitoring systems for HPC in the future. We demons
 trate how real-world applications and workflows could benefit from the app
 roach, and we show how the approach helps communicate  performance-tuning 
 guidance to users.
URL:https://sc18.supercomputing.org/presentation/?id=ws_pdsw109&sess=sess1
 42
END:VEVENT
END:VCALENDAR

