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:20190613T141500
DTEND;TZID=Europe/Stockholm:20190613T144500
UID:submissions.pasc-conference.org_PASC19_sess166_msa170@linklings.com
SUMMARY:A GPU High Level Trigger 1 for LHCb
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Physi
 cs\n\nA GPU High Level Trigger 1 for LHCb\n\nvom Bruch\n\nThe LHCb experim
 ent at CERN is being upgraded for 2021 to read out the particle detectors 
 at 30 MHz and to select interesting proton bunch collisions (events) with 
 a software-only trigger. This allows unprecedented flexibility for trigger
  selections but at the same time poses a significant computing challenge. 
 We present an R&D project to run the full first trigger stage, High Le
 vel Trigger 1 (HLT1), on GPUs. During HLT1, a sub-set of the full offline 
 track reconstruction for charged particles is run to select particles of i
 nterest based on single or two-track selections. After this first stage, t
 he event rate is reduced by at least a factor 30 and selected events are p
 assed on to the second trigger stage. The HLT1 tasks include decoding the 
 raw data, clustering of hits, pattern recognition, as well as track fittin
 g and the selection of interesting events. We will present the algorithms 
 optimized for many-core architectures, discuss the concept, advantages and
  challenges of performing a full event selection on GPUs and introduce the
  framework we have developed for this task.
END:VEVENT
END:VCALENDAR