BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191206T142454Z
LOCATION:Analog 1\, 2
DTSTART;TZID=Europe/Stockholm:20180626T110000
DTEND;TZID=Europe/Stockholm:20180626T113000
UID:isc_hpc_ISC High Performance 2018_sess110_pap128@linklings.com
SUMMARY:Bayesian Optimization of HPC Systems for Energy Efficiency
DESCRIPTION:Research Paper\n\nBayesian Optimization of HPC Systems for Ene
 rgy Efficiency\n\nMiyazaki, Sato, Shimizu\n\nEnergy efficiency is a crucia
 l factor in developing large supercomputers and cost-effective datacenters
 . However, tuning a system for energy efficiency is difficult because the 
 power and performance are conflicting demands. We applied Bayesian optimiz
 ation (BO) to tune a graphics processing unit (GPU) cluster system for the
  benchmark used in the Green500 list, a popular energy-efficiency ranking 
 of supercomputers. The resulting benchmark score enabled our system, named
  ``kukai'',  to earn second place in the Green500 list in June 2017, showi
 ng that BO is a useful tool. By determining the search space with minimal 
 knowledge and preliminary experiments beforehand, BO could automatically f
 ind a sufficiently good configuration. Thus, BO could eliminate laborious 
 manual tuning work and reduce the occupancy time of the system for benchma
 rking. Because BO is a general-purpose method, it may also be useful for t
 uning any practical applications in addition to Green500 benchmarks.\n\nPa
 sses: Conference Pass, AI/Machine Learning/Deep Learning, Energy consumpti
 on, Performance Analysis and Optimization\n\nTag: Conference Pass, AI/Mach
 ine Learning/Deep Learning, Energy consumption, Performance Analysis and O
 ptimization
URL:https://2018.isc-program.com/presentation/?id=pap128&sess=sess110
END:VEVENT
END:VCALENDAR