Pith. sign in

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

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
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.

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Track reconstruction as a service for collider physics

physics.ins-det · 2025-01-09 · conditional · novelty 4.0

Running the Patatrack and Exa.TrkX tracking algorithms through NVIDIA Triton as a remote service gives near-local GPU throughput while letting one GPU serve many more CPU clients.

citing papers explorer

Showing 1 of 1 citing paper.

  • Track reconstruction as a service for collider physics physics.ins-det · 2025-01-09 · conditional · none · ref 12 · internal anchor

    Running the Patatrack and Exa.TrkX tracking algorithms through NVIDIA Triton as a remote service gives near-local GPU throughput while letting one GPU serve many more CPU clients.