GPU-accelerated machine learning inference as a service for computing in neutrino experiments

TL;DR

This paper presents SONIC, a GPU-accelerated web service for neutrino experiment data processing that significantly speeds up event reconstruction tasks, reducing total processing time and costs.

Contribution

We introduce a GPU-based inference service integrated into neutrino data processing, achieving substantial speedups and cost efficiency without disrupting existing workflows.

Findings

Track and shower hit identification accelerated by 17x

Total processing time reduced by a factor of 2.7

Only 1 GPU per 68 CPU threads needed for efficiency

Abstract

Machine learning algorithms are becoming increasingly prevalent and performant in the reconstruction of events in accelerator-based neutrino experiments. These sophisticated algorithms can be computationally expensive. At the same time, the data volumes of such experiments are rapidly increasing. The demand to process billions of neutrino events with many machine learning algorithm inferences creates a computing challenge. We explore a computing model in which heterogeneous computing with GPU coprocessors is made available as a web service. The coprocessors can be efficiently and elastically deployed to provide the right amount of computing for a given processing task. With our approach, Services for Optimized Network Inference on Coprocessors (SONIC), we integrate GPU acceleration specifically for the ProtoDUNE-SP reconstruction chain without disrupting the native computing workflow. With our integrated framework, we accelerate the most time-consuming task, track and particle shower hit identification, by a factor of 17. This results in a factor of 2.7 reduction in the total processing time when compared with CPU-only production. For this particular task, only 1 GPU is required for every 68 CPU threads, providing a cost-effective solution.