The following email has been sent to GALLORINI, Stefano:

===

Dear Stefano Gallorini,

The submission of your abstract has been successfully processed.

Abstract submitted: https://indico.cern.ch/event/192695/call-for- abstracts/my-abstracts.

Status of your abstract: https://indico.cern.ch/event/192695/call- for-abstracts/328/.

See below a detailed summary of your submitted abstract:

Conference: Tipp 2014 - Third International Conference on Technology and Instrumentation in Particle Physics

Submitted by: GALLORINI, Stefano

Submitted on: 28 February 2014 10:01

Title: Many-core studies on pattern-recognition in the LHCb experiment

Abstract content The LHCb experiment is entering in its upgrading phase, with its detector and read-out system re-designed to cope with the increased LHC energy after the long shutdown of 2018. In this upgrade, a trigger-less data acquisition is being developed to read-out the full detector at the bunch-crossing rate of 40 MHz. In particular, the High Level Trigger (HLT) system, where the bulk of the trigger decision is implemented via software on a CPU farm, has to be heavily revised. Since the small LHCb event size (about 100 kB after the upgrade), many-core architectures such as General Purpose GPU (GPGPU) and multi-core CPUs can be used to process many events in parallel for real-time selection, and may offer a solution for reducing the cost of the HLT farm. Track reconstruction and vertexing are the more time-consuming applications running in HLT and therefore are the first to be ported on many-core. In this talk we discussed the studies ongoing in LHCb for implementing pattern-recogniton algorithms for the Velo detector on many-core systems. We present our solution for porting the existing Velo tracking algorithm (FastVelo) on GPGPU, and we show the achieved performance. We plan to test the parallelized version of FastVelo during the data-taking in 2015 and assess the impact of the many-core solution on the HLT infrastructure. We discuss also other tracking algorithms in view of the upgrade and their preliminary performaces.

Summary

Primary Authors: GALLORINI, Stefano (Universita e INFN (IT)) stefano.gallorini@cern.ch

Co-authors: LUCCHESI, Donatella (Universita e INFN (IT)) donatella.lucchesi@cern.ch Mr. GIANELLE, Alessio (Universita e INFN (IT)) alessio.gianelle@pd.infn.it CORVO, Marco (INFN) corvo@pd.infn.it COLLAZUOL, Gianmaria (U) giancoll@cern.ch SESTINI, Lorenzo (Universita e INFN (IT)) lorenzo.sestini@cern.ch LUPATO, Anna (Universita di Ferrara (IT)) anna.lupato@cern.ch NEUFELD, Niko (CERN) niko.neufeld@cern.ch CAMPORA PEREZ, Daniel Hugo (CERN) daniel.hugo.campora.perez@cern.ch VILASIS CARDONA, Xavier (University of Barcelona (ES)) xavier.vilasis.cardona@cern.ch

Abstract presenters: GALLORINI, Stefano

Track classification: Data-processing: 3b) Trigger and Data Acquisition Systems

Presentation type: Oral

Comments: