Application of Graphics Processing Units to Search Pipeline for Gravitational Waves from Coalescing Binaries of Compact Objects

TL;DR

This paper demonstrates that using GPUs can significantly accelerate the search for gravitational waves from coalescing binaries, achieving a 16-fold speed-up over traditional CPU methods and reducing CPU requirements.

Contribution

The study introduces a novel GPU-based approach to gravitational wave search pipelines, substantially improving processing speed and efficiency over CPU-only methods.

Findings

16-fold speed-up achieved with GPU implementation

Reduced CPU count needed for gravitational wave detection

Potential for substantial improvements in search pipeline efficiency

Abstract

We report a novel application of graphics processing units (GPUs) for the purpose of accelerating the search pipelines for gravitational waves from coalescing binaries of compact objects. A speed-up of 16 fold has been achieved compared with a single central processing unit (CPU). We show that substantial improvements are possible and discuss the reduction in CPU count required for the detection of inspiral sources afforded by the use of GPUs.