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:20191206T142455Z
LOCATION:Substanz 1\, 2
DTSTART;TZID=Europe/Stockholm:20180627T093000
DTEND;TZID=Europe/Stockholm:20180627T100000
UID:isc_hpc_ISC High Performance 2018_sess161_pap175@linklings.com
SUMMARY:Hardware Performance Variation: A Comparative Study using Lightwei
 ght Kernels
DESCRIPTION:Research Paper\n\nHardware Performance Variation: A Comparativ
 e Study using Lightweight Kernels\n\nWeisbach, Kocoloski, Ishikawa, Härtig
 , Gerofi\n\nImbalance among components of large scale parallel simulations
  can adversely affect overall application performance. Software induced im
 balance has been extensively studied in the past, however, there is a grow
 ing interest in characterizing and understanding another source of variabi
 lity, the one induced by the hardware itself. This is particularly interes
 ting with the growing diversity of hardware platforms deployed in high-per
 formance computing (HPC) and the increasing complexity of computer archite
 ctures in general. Nevertheless, characterizing hardware performance varia
 bility is challenging as one needs to ensure a tightly controlled software
  environment.\nIn this paper, we propose to use lightweight operating syst
 em kernels to provide a high-precision characterization of various aspects
  of hardware performance variability. Towards this end, we have developed 
 an extensible benchmarking framework and characterized multiple compute pl
 atforms\n(e.g., Intel x86, Cavium ARM64, Fujitsu SPARC64, IBM Power) runni
 ng on top of lightweight kernel operating systems. Our initial findings sh
 ow up to six orders of magnitude difference in relative variation among CP
 U cores across different platforms.\n\nPasses: Conference Pass, Computer A
 rchitecture, Performance Analysis and Optimization, Reproducibility, Syste
 m Software & Runtime Systems\n\nTag: Conference Pass, Computer Architectur
 e, Performance Analysis and Optimization, Reproducibility, System Software
  & Runtime Systems
URL:https://2018.isc-program.com/presentation/?id=pap175&sess=sess161
END:VEVENT
END:VCALENDAR