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:20190719T085744Z
LOCATION:HG E 3
DTSTART;TZID=Europe/Stockholm:20190613T154500
DTEND;TZID=Europe/Stockholm:20190613T161500
UID:submissions.pasc-conference.org_PASC19_sess166_msa302@linklings.com
SUMMARY:Accelerating HEP with GPUs at LHC
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Physi
 cs\n\nAccelerating HEP with GPUs at LHC\n\nDi Florio, Di Pilato\n\nThe HEP
  community is approaching an era were the excellent performances of the pa
 rticle accelerators in delivering collision at high rate is forcing t
 he experiments to record and analyse a huge amount of information. On the 
 other hand, CPU evolution will not be able to cope with the increasing dem
 and of performance and, depending on the application, GPUs may provide bet
 ter performance and energy efficiency. For all the LHC experiments th
 e LS2 represents an excellent occasion to embrace a paradigm shift towards
  heterogeneous computing. The acceleration of algorithms with GPUs is expe
 cted to benefit both online computing by decreasing the overall cost of th
 e event selection farm or increasing its discovery potential, and offline 
 computing by enabling software frameworks to execute efficiently on HPC ce
 nters. Also GPUs have proven to be of invaluable importance for data scien
 tists by accelerating by factors the training and inference times of Neura
 l Networks.  The multiple non-coordinated efforts made to develop and
  deploying GPU-based solutions within the HEP computing community come wit
 h novel implicit criticalities such as code duplication and performance po
 rtability. This contribution will describe both the latest achievemen
 ts and the open issues the LHC experiments needs to cope with in the next 
 years.
END:VEVENT
END:VCALENDAR