A GPU Implementation of the Correlation Technique for Real-time Fourier Domain Pulsar Acceleration Searches

• Published in: The Astrophysical journal. Supplement series Vol. 239; no. 2; pp. 28 - 41
• Main Authors: Dimoudi, Sofia, Adamek, Karel, Thiagaraj, Prabu, Ransom, Scott M., Karastergiou, Aris, Armour, Wesley
• Format: Journal Article
• Language: English
• Published: Saskatoon The American Astronomical Society 01.12.2018; IOP Publishing
• Subjects: Acceleration; Algorithms; Astronomy; Binary systems; Convolution; Correlation; Fast Fourier transformations; FIR filters; Fourier transforms; Graphics processing units; instrumentation: miscellaneous; methods: data analysis; methods: numerical; Microprocessors; Periodic variations; Pulsars; pulsars: general; Radio astronomy; Radio signals; Real time; Relativity; Searching; Signal processing; telescopes
• ISSN: 0067-0049; 1538-4365; 1538-4365
• DOI: 10.3847/1538-4365/aabe88

The study of binary pulsars enables tests of general relativity. Orbital motion in binary systems causes the apparent pulsar spin frequency to drift, reducing the sensitivity of periodicity searches. Acceleration searches are methods that account for the effect of orbital acceleration. Existing methods are currently computationally expensive, and the vast amount of data that will be produced by next-generation instruments such as the Square Kilometre Array necessitates real-time acceleration searches, which in turn requires the use of high-performance computing (HPC) platforms. We present our implementation of the correlation technique for the Fourier Domain Acceleration Search (FDAS) algorithm on Graphics Processor Units (GPUs). The correlation technique is applied as a convolution with multiple finite impulse response (FIR) filters in the Fourier domain. Two approaches are compared: the first uses the NVIDIA cuFFT library for applying Fast Fourier transforms (FFTs) on the GPU, and the second contains a custom FFT implementation in GPU shared memory. We find that the FFT shared-memory implementation performs between 1.5 and 3.2 times faster than our cuFFT-based application for smaller but sufficient filter sizes. It is also 4-6 times faster than the existing GPU and OpenMP implementations of FDAS. This work is part of the AstroAccelerate project, a many-core accelerated time-domain signal-processing library for radio astronomy.