Thread-Scalable Evaluation of Multi-Jet Observables

TL;DR

This paper presents a GPU-accelerated event generator for multi-jet observables, achieving significant speed-ups and analyzing its feasibility, convergence, and accuracy for high-energy physics simulations.

Contribution

It introduces a multi-threaded GPU-based event generator for multi-jet processes, demonstrating substantial performance improvements over CPU implementations.

Findings

Speed-up factors of 150-300 times over CPU implementations

Feasibility of GPU-based event generation for complex observables

Monte Carlo convergence and accuracy for up to 10-jet events

Abstract

A leading-order, leading-color parton-level event generator is developed for use on a multi-threaded GPU. Speed-up factors between 150 and 300 are obtained compared to an unoptimized CPU-based implementation of the event generator. In this first paper we study the feasibility of a GPU-based event generator with an emphasis on the constraints imposed by the hardware. Some studies of Monte Carlo convergence and accuracy are presented for PP -> 2,...,10 jet observables using of the order of 1e11 events.