Development of a hardware-accelerated simulation kernel for ultra-high vacuum with Nvidia RTX GPUs. (March 2022)

• Record Type: Journal Article
• Authors: Bähr, Pascal R; Lang, Bruno; Ueberholz, Peer; Ady, Marton; Kersevan, Roberto
• Is Part Of: International journal of high performance computing applications. Volume 36:Number 2(2022)
• Page Start: 141
• Page End: 152
• Publication Date: 2022-03
• Subjects: Molflow -- vacuum -- Monte Carlo -- simulations -- graphics processing units -- RTX -- Nvidia; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/10943420211056654
• Languages: English
• ISSNs: 1094-3420

Abstract

Molflow+ is a Monte Carlo (MC) simulation software for ultra-high vacuum, mainly used to simulate pressure in particle accelerators. In this article, we present and discuss the design choices arising in a new implementation of its ray-tracing–based simulation unit for Nvidia RTX Graphics Processing Units (GPUs). The GPU simulation kernel was designed with Nvidia's OptiX 7 API to make use of modern hardware-accelerated ray-tracing units, found in recent RTX series GPUs based on the Turing and Ampere architectures. Even with the challenges posed by switching to 32 bit computations, our kernel runs much faster than on comparable CPUs at the expense of a marginal drop in calculation precision.