BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160554Z
LOCATION:Track 2
DTSTART;TZID=America/New_York:20201112T120000
DTEND;TZID=America/New_York:20201112T123000
UID:submissions.supercomputing.org_SC20_sess208_ws_pmbsf121@linklings.com
SUMMARY:Exploiting the Potentials of the Second Generation SX-Aurora TSUBA
 SA
DESCRIPTION:Workshop\n\nExploiting the Potentials of the Second Generation
  SX-Aurora TSUBASA\n\nEgawa, Fujimoto, Yamashita, Sasaki, Isobe...\n\nNEC 
 SX-series vector supercomputers have provided outstanding memory bandwidth
 s to meet the strong demands for efficient execution of memory-intensive s
 cientific applications in practice. Inheriting the advantage, the 2nd gene
 ration SX-Aurora TSUBASA, Type 20B, provides an extremely high memory band
 width of 1.53 TB/s per vector processor. Unlike conventional SX-series sys
 tems, SX-Aurora TSUBASA also offers various execution modes to execute a d
 iversity of emerging scientific workloads efficiently. As a result, applic
 ation developers need to understand their workloads and the performance ch
 aracteristics of SX-Aurora TSUBASA, and select an optimization strategy as
 suming an appropriate execution mode to fully exploit the system performan
 ce. Therefore, this paper discusses workload characterization by performan
 ce bottleneck analysis to determine an optimization strategy for the 2nd g
 eneration SX-Aurora TSUBASA. The evaluation results with benchmarks and re
 al-world applications demonstrate that the workload characterization appro
 ach can accurately find the bottleneck and characterize various workloads,
  by helping application developers decide the optimization strategies for 
 individual workloads. Since we can consider SX-Aurora TSUBASA as a typical
  example of the latest processors with high memory bandwidths, the workloa
 d characterization approach will also be helpful for other future processo
 rs.\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR