GPU-based acceleration of Radio Interferometry point source visibility calculations in the MEQtrees framework
Baxter, Richard , Patrick Marais and Michelle Kuttel (2013) GPU-based acceleration of Radio Interferometry point source visibility calculations in the MEQtrees framework. In Proceedings Astronomical Data Analysis Software and Systems - ADASS XXII 475, pages 53-59, Chicago, USA.
Full text available as:
Modern radio interferometer arrays are powerful tools for obtaining high resolution,
low frequency images of objects in deep space. An inverse Fourier transform of a
model sky-intensity map will produce the corresponding Point Source Visibilities: the
raw output of an interferometer. Simulated visibilities can be used to test models of
factors affecting the accuracy of observed data, such as radio frequency interference.
We describe a GPU/CUDA implementation of the Point Source Visibility calculation
module within theMeqTrees software suite. For a large numbers of sources, we achieve
an 18× speed-up over the existing CPU module, with the parallel component running
up to 120× faster. With modifications to the MeqTrees memory management system to
incorporate GPU memory operations, a speed-up of 24× is achievable.
|EPrint Type:||Conference Paper|
|Subjects:||I Computing Methodologies: I.4 IMAGE PROCESSING AND COMPUTER VISION|
I Computing Methodologies: I.6 SIMULATION AND MODELING
|Deposited By:||Marais, Patrick|
|Deposited On:||28 October 2013|