Pfister, HanspeterEdgar, Richard GMitchell, DanielOrd, StephenGreenhill, LincolnClark, Michael A.Dale, KevinWayth, Randall B.2011-11-102010Edgara, Richard G., Michael A. Clark, Kevin Dale, Daniel A. Mitchell, Stephen M. Ord, Randall B. Wayth, Hanspeter Pfister, and Lincoln J. Greenhill. 2010. Enabling a high throughput real time data pipeline for a large radio telescope array with GPUs. Computer Physics Communications 181(10): 1707-1714.0010-4655http://nrs.harvard.edu/urn-3:HUL.InstRepos:5344187The Murchison Widefield Array (MWA) is a next-generation radio telescope currently under construction in the remote Western Australia Outback. Raw data will be generated continuously at 5 GiB s\(^{−1}\), grouped into 8 s cadences. This high throughput motivates the development of on-site, real time processing and reduction in preference to archiving, transport and off-line processing. Each batch of 8 s data must be completely reduced before the next batch arrives. Maintaining real time operation will require a sustained performance of around 2.5 TFLOP s\(^{−1}\) (including convolutions, FFTs, interpolations and matrix multiplications). We describe a scalable heterogeneous computing pipeline implementation, exploiting both the high computing density and FLOP-per-Watt ratio of modern GPUs. The architecture is highly parallel within and across nodes, with all major processing elements performed by GPUs. Necessary scatter-gather operations along the pipeline are loosely synchronized between the nodes hosting the GPUs. The MWA will be a frontier scientific instrument and a pathfinder for planned peta- and exascale facilities.en-USradio telescopes and instrumentationheterodyne receiversmathematical procedures and computer techniquescomputer science and technologyJournal Article2011-11-1010.1016/j.cpc.2010.06.019