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:20260522T150111Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess327_spost139@linklings.com
SUMMARY:Identifying Network Data Transfer Bottlenecks in HPC Systems
DESCRIPTION:Karen Tu (Lawrence Berkeley National Laboratory; University of
  California, Berkeley)\n\nImproving network data transfer performance is a
  major factor for improving high performance computing systems. Most studi
 es analyze data transfer and file system IO performance separately, but un
 derstanding the relationship between the two is essential for optimizing s
 cheduling and resource management. Intuitively, if data is being transferr
 ed to a busy file system the transfer rate would be slower than a file sys
 tem at regular activity levels.\n\nThis study analyzes patterns between fi
 le system activity and network throughput for several use cases of file wr
 iting and data transfers using a parallel file system. The parameters chan
 ged among the use cases were file striping for the file system, and buffer
  size and parallelism for data transfer. The main bottleneck for network d
 ata transfer rate was the number of OSTs the data was striped across. For 
 a large number of OSTs (16 or greater), writing to the file system was the
  bottleneck.\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg
  Pass\n\n
END:VEVENT
END:VCALENDAR