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:20190719T085745Z
LOCATION:HG E 3
DTSTART;TZID=Europe/Stockholm:20190614T150000
DTEND;TZID=Europe/Stockholm:20190614T153000
UID:submissions.pasc-conference.org_PASC19_sess164_msa303@linklings.com
SUMMARY:Patatrack: Towards a Heterogeneous High Level Trigger Farm for CMS
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Physi
 cs\n\nPatatrack: Towards a Heterogeneous High Level Trigger Farm for CMS\n
 \nDi Florio, Di Pilato\n\nThe CMS experiment has been designed with a two-
 level trigger system: the L1 Trigger, hardware based, and the High Le
 vel Trigger (HLT), a streamlined version of the CMS reconstruction softwar
 e running on a computer farm. During its “Phase 2” the LHC wil
 l reach a luminosity of 7×10³⁴ cm⁻²s⁻&su
 p1; with a pileup of 200 collisions and the L1 output rate will increase u
 p to 750 kHz. All of this will present an unprecedented challenge to 
 the HLT, requiring a processing power larger than today by a factor 20. Th
 is exceeds by far the expected speed-up for conventional CPUs, demanding a
 n alternative approach. Industry and HPC have been using heterogeneou
 s computing platforms achieving higher throughput and better energy effici
 ency by matching each job to the most appropriate architecture. The reliab
 le use of a heterogeneous platform at the CMS HLT requires the assessment 
 of its performance and characteristics, which can be attained by running a
  prototype in production during Run 3. The integration of heterogeneous co
 mputing in the CMS reconstruction software depends upon improvements to it
 s framework and scheduling, tailoring the algorithms to fit the different 
 architectures. This presentation will describe the results of the developm
 ent and the characteristics of the system, along with its future perspecti
 ves.
END:VEVENT
END:VCALENDAR