# CMS physics reconstruction running on NextSilicon Maverick accelerators

- Institution: CERN
- Speaker: Fernando Antonio Marques Schettini
- Year: 2026
- Language: English
- Subjects: Physics
- Source: [CERN](https://videos.cern.ch/record/3026668)
- Licence: [CERN](https://copyright.web.cern)
- Who can watch: Tweens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

**Project Description**:

This project evaluates the suitability of NextSilicon's Maverick-2 dataflow accelerator for HPC workloads at CERN using the CLUE clustering algorithm as a benchmark. The CLUE implementation is adapted and optimized for execution on Maverick-2 and compared against a CPU OpenMP baseline. The study analyzes kernel-level performance, end-to-end execution time, memory movement, and accelerator offload overheads. The results provide insight into the benefits and current limitations of dataflow architectures for CMS reconstruction workloads.

**Supervisor**: Andrea Bocci and Mario Gonzalez

00:00:00 Slide 1

00:00:01 Slide 2

00:00:25 Slide 3

00:01:20 Slide 4

00:02:30 Slide 5

00:03:07 Slide 6

00:04:18 Slide 7

00:04:48 Slide 8

00:05:04 Slide 9

00:05:22 Slide 10

## Lectures

1. [CMS physics reconstruction running on NextSilicon Maverick accelerators](https://invisiblenetwork.ai/learn/cern/3026668/64f041a1f30f), 7 min (CERN's player)