High-Performance Image Synthesis for Radio Interferometry
Institute of Space Science and Astronomy, Department of Physics, University of Malta
A radio interferometer indirectly measures the intensity distribution of the sky over the celestial sphere. Since measurements are made over an irregularly sampled Fourier plane, synthesising an intensity image from interferometric measurements requires substantial processing. Furthermore there are distortions that have to be corrected. In this thesis, a new high-performance image synthesis tool (imaging tool) for radio interferometry is developed. Implemented in C++ and CUDA, the imaging tool achieves unprecedented performance by means of Graphics Processing Units (GPUs). The imaging tool is divided into several components, and the back-end handling numerical calculations is generalised in a new framework. A new feature termed compression arbitrarily increases the performance of an already highly efficient GPU-based implementation of the w-projection algorithm. Compression takes advantage of the behaviour of oversampled convolution functions and the baseline trajectories. A CPU-based component prepares data for the GPU which is multi-threaded to ensure maximum use of modern multi-core CPUs. Best performance can only be achieved if all hardware components in a system do work in parallel. The imaging tool is designed such that disk I/O and work on CPU and GPUs is done concurrently. Test cases show that the imaging tool performs nearly 100× faster than another general CPU-based imaging tool. Unfortunately, the tool is limited in use since deconvolution and A-projection are not yet supported. It is also limited by GPU memory. Future work will implement deconvolution and A-projection, whilst finding ways of overcoming the memory limitation.
March 24, 2014 by hgpu
Comments
3 responses to “High-Performance Image Synthesis for Radio Interferometry”
Your response
You must be logged in to post a comment.
Dear hgpu,
I am the author of this thesis. Please note that I have no connection with University of the Witwatersrand. As it is clear from the first page of the thesis this work was done at the University Of Malta. Please change affiliation to Institute of Space Science and Astronomy, Department of Physics, University of Malta.
Regards
Daniel
Dear Daniel,
Thank you very much for your attention to our hgpu site.
We apologize for the erroneous information presented, the affiliation has already changed.
Hope, we will be useful source of information on GPU computation for you as well as a way for dissemination of the results.
Best regards
hgpu
thanks a lot