Enabling a high throughput real time data pipeline for a large radio telescope array with GPUs |
| |
Authors: | RG Edgar MA Clark K Dale |
| |
Affiliation: | a Initiative in Innovative Computing, 29 Oxford Street, Cambridge, MA 02138, United States b Harvard-Smithsonian Center for Astrophysics, 60 Oxford Street MS-42, Cambridge, MA 02138, United States c School of Engineering & Applied Sciences, 33 Oxford Street, Cambridge, MA 02138, United States |
| |
Abstract: | The 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 exa-scale facilities. |
| |
Keywords: | Radio telescopes and instrumentation Heterodyne receivers Mathematical procedures and computer techniques Computer science and technology |
本文献已被 ScienceDirect 等数据库收录! |
|