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

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.

15 pages, 7 figures

References in corpus (5)

• High Performance Direct Gravitational N-body Simulations on Graphics Processing Units -- II: An implementation in CUDA
• Search for Gravitational Waves from Low Mass Binary Coalescences in the First Year of LIGO's S5 Data
• The path to the enhanced and advanced LIGO gravitational-wave detectors
• High Performance Direct Gravitational N-body Simulations on Graphics Processing Unit I: An implementation in Cg
• Searching for Gravitational Waves from Binary Inspirals with LIGO

Cited by in corpus (5)

• Capturing the electromagnetic counterparts of binary neutron star mergers through low latency gravitational wave triggers
• Towards low-latency real-time detection of gravitational waves from compact binary coalescences in the era of advanced detectors
• PyCOOL - a Cosmological Object-Oriented Lattice code written in Python
• Gravitational wave astrophysics, data analysis and multimessenger astronomy
• cuInspiral: prototype gravitational waves detection pipeline fully coded on GPU using CUDA