Abstract
The future High Luminosity LHC (HL-LHC) is expected to deliver about 5 times higher instantaneous luminosity than the present LHC, resulting in pile-up up to 200 interactions per bunch crossing (PU200). As part of the phase-II upgrade program, the CMS collaboration is developing a new endcap calorimeter system, the High Granularity Calorimeter (HGCAL), featuring highly-segmented hexagonal silicon sensors and scintillators with more than 6 million channels. For each event, the HGCAL clustering algorithm needs to group more than 105 hits into clusters. As consequence of both high pile-up and the high granularity, the HGCAL clustering algorithm is confronted with an unprecedented computing load. CLUE (CLUsters of Energy) is a fast fullyparallelizable density-based clustering algorithm, optimized for high pile-up scenarios in high granularity calorimeters. In this paper, we present both CPU and GPU implementations of CLUE in the application of HGCAL clustering in the CMS Software framework (CMSSW). Comparing with the previous HGCAL clustering algorithm, CLUE on CPU (GPU) in CMSSW is 30x (180x) faster in processing PU200 events while outputting almost the same clustering results.
Highlights
The luminosity of the future High Luminosity Large Hadron Collider (HL-LHC) is going to achieve up to 7.5 × 1034 cm−2 s−1 [1] in its ultimate scenario, which is 5 times that delivered at present
GPU implementations of CLUsters of Energy (CLUE) in the application of high granularity calorimeter system (HGCAL) clustering in the CMS Software framework (CMSSW)
One of the major tasks in the CMS Phase-II upgrade is to replace the current endcap calorimeters, including both endcap electromagnetic and hadronic calorimeters, with a new high granularity calorimeter system (HGCAL) which is based on highly-segmented Silicon sensors and plastic scintillators
Summary
The luminosity of the future High Luminosity Large Hadron Collider (HL-LHC) is going to achieve up to 7.5 × 1034 cm−2 s−1 [1] in its ultimate scenario, which is 5 times that delivered at present. Among this 30x surge of the computing demand, improvement in the CPU performance by 2026 is expected to account for only 4x. In the HL-LHC era, the HLT time budget for CMS HGCAL clustering is roughly estimated to be less than a few tens of milliseconds It is a huge challenge of computing for the HGCAL clustering algorithm to process n ∼ 5 × 105 hits within such a limited time budget. We demonstrate that CLUE on CPU (GPU) is about 30x (180x) faster than the previous HGCAL clustering algorithm [8] in CMSSW for PU200 events
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
