Portability for the Patatrack Pixel Track Reconstruction with Alpaka

The Compact Muon Solenoid (CMS) experiment is one of the largest experiments at Large Hadron Collider (LHC) that has been built to search for new physics. CMS Software (CMSSW) is the framework utilized by the CMS experiment for Data Acquisition, Trigger and Event Reconstruction. The future upgrade of the LHC will add new challenges for CMS detector due to the larger amount of data that will be produced. One solution is to look towards a heterogeneous High-Level Trigger (HLT) Computing Farm, where the computing load can be distributed among different hardware (CPU, GPU, FPGAS). This project aims to use Portability Libraries like Alpaka to write code for the CMS Software (CMSSW), in order to have a single implementation which will be compiled on different architectures. In the last years, a new pixel track reconstruction that can run on GPUs (CUDA) has been implemented. The goals are to port the current CUDA implementation with Alpaka, in order to understand if this library is feasible for CMSSW and to compare the timing performances between Alpaka implementation and native CUDA.