End-to-end deep learning inference with CMSSW via ONNX using docker

TL;DR

This paper presents an implementation of end-to-end deep learning inference within the CMS software framework using ONNX and Docker, enabling real-time particle identification with GPU acceleration.

Contribution

It introduces a novel integration of deep learning inference into CMSSW with GPU support via ONNX and Docker, addressing computational challenges in high-energy physics applications.

Findings

GPU inference with ONNX is faster than CPU.

Docker containerization simplifies deployment within CMSSW.

Benchmark results demonstrate improved performance on NERSC Perlmutter.

Abstract

Deep learning techniques have been proven to provide excellent performance for a variety of high-energy physics applications, such as particle identification, event reconstruction and trigger operations. Recently, we developed an end-to-end deep learning approach to identify various particles using low-level detector information from high-energy collisions. These models will be incorporated in the CMS software framework (CMSSW) to enable their use for particle reconstruction or for trigger operation in real-time. Incorporating these computational tools in the experimental framework presents new challenges. This paper reports an implementation of the end-to-end deep learning inference with the CMS software framework. The inference has been implemented on GPU for faster computation using ONNX. We have benchmarked the ONNX inference with GPU and CPU using NERSCs Perlmutter cluster by building a docker image of the CMS software framework.