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:20211207T055412Z
LOCATION:240-241-242
DTSTART;TZID=America/Chicago:20211118T153000
DTEND;TZID=America/Chicago:20211118T170000
UID:submissions.supercomputing.org_SC21_sess166@linklings.com
SUMMARY:Storage and Application Characteristics
DESCRIPTION:Paper\n\nPinpointing Crash-Consistency Bugs in the HPC I/O Sta
 ck: A Cross-Layer Approach\n\nSun, Huang, Snir\n\nWe present ParaCrash, a 
 testing framework for studying crash recovery in a typical HPC I/O stack, 
 and demonstrate its use by identifying 15 new crash-consistency bugs in va
 rious parallel file systems (PFS) and I/O libraries. ParaCrash uses a "gol
 den version'' approach to test the entire HPC I/O stac...\n\n-------------
 --------\nCharacterization and Prediction of Deep Learning Workloads in La
 rge-Scale GPU Datacenters\n\nHu, Sun, Yan, Wen, Zhang\n\nModern GPU datace
 nters are critical for delivering Deep Learning (DL) models and services i
 n both the research community and industry. When operating a datacenter, o
 ptimization of resource scheduling and management can bring significant fi
 nancial benefits. Achieving this goal requires a deep underst...\n\n------
 ---------------\nExploiting User Activeness for Data Retention in HPC Syst
 ems\n\nZhang, Byna, Sim, Lee, Vazhkudai...\n\nHPC systems typically rely o
 n the fixed-lifetime (FLT) data retention strategy, which only considers t
 emporal locality of data accesses to parallel file systems. Our extensive 
 analysis based on the leadership-class HPC system traces, however, suggest
 s that the FLT approach often fails to capture the...\n\n\nTag: Applicatio
 ns, Big Data, Datacenter, File Systems and I/O, Machine Learning and Artif
 icial Intelligence, State of the Practice, Storage\n\nRegistration Categor
 y: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR