A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches
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 meth...
| Главные авторы: | , , , , , |
|---|---|
| Формат: | Journal article |
| Опубликовано: |
American Astronomical Society
2018
|
| _version_ | 1826268587269029888 |
|---|---|
| author | Dimoudi, S Adamek, K Thiagaraj, P Ransom, S Karastergiou, A Armour, W |
| author_facet | Dimoudi, S Adamek, K Thiagaraj, P Ransom, S Karastergiou, A Armour, W |
| author_sort | Dimoudi, S |
| collection | OXFORD |
| description | 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. |
| first_indexed | 2024-03-06T21:11:59Z |
| format | Journal article |
| id | oxford-uuid:3e7e588f-be78-4e8e-bf23-eb33a2947e6b |
| institution | University of Oxford |
| last_indexed | 2024-03-06T21:11:59Z |
| publishDate | 2018 |
| publisher | American Astronomical Society |
| record_format | dspace |
| spelling | oxford-uuid:3e7e588f-be78-4e8e-bf23-eb33a2947e6b2022-03-26T14:25:48ZA GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searchesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:3e7e588f-be78-4e8e-bf23-eb33a2947e6bSymplectic Elements at OxfordAmerican Astronomical Society2018Dimoudi, SAdamek, KThiagaraj, PRansom, SKarastergiou, AArmour, WThe 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. |
| spellingShingle | Dimoudi, S Adamek, K Thiagaraj, P Ransom, S Karastergiou, A Armour, W A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches |
| title | A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches |
| title_full | A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches |
| title_fullStr | A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches |
| title_full_unstemmed | A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches |
| title_short | A GPU implementation of the correlation technique for real-time Fourier domain pulsar acceleration searches |
| title_sort | gpu implementation of the correlation technique for real time fourier domain pulsar acceleration searches |
| work_keys_str_mv | AT dimoudis agpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT adamekk agpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT thiagarajp agpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT ransoms agpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT karastergioua agpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT armourw agpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT dimoudis gpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT adamekk gpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT thiagarajp gpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT ransoms gpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT karastergioua gpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches AT armourw gpuimplementationofthecorrelationtechniqueforrealtimefourierdomainpulsaraccelerationsearches |