Enabling a high throughput real time data pipeline for a large radio telescope array with GPUs

DOI10.1016/J.CPC.2010.06.019zbMATH Open1220.85002arXiv1003.5575OpenAlexW2126734610WikidataQ59391807 ScholiaQ59391807MaRDI QIDQ548974FDOQ548974

Authors: Yong-Cai Geng, Sumit K. Garg 

Publication date: 30 June 2011

Published in: Computer Physics Communications (Search for Journal in Brave)

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 5GiB/s, grouped into 8s 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 8s data must be completely reduced before the next batch arrives. Maintaining real time operation will require a sustained performance of around 2.5TFLOP/s (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.

Full work available at URL: https://arxiv.org/abs/1003.5575

Recommendations

• Big telescope, big data: towards exascale with the square kilometre array
• Real-time signal parallel processing scheme and realization of HF ground wave radar based on CPU+GPU architecture
• Application of graphics processing units to search pipelines for gravitational waves from coalescing binaries of compact objects
• Modern parallel architectures to speed up numerical simulation
• Performance evaluation of a parallel pipeline computational model for space-time adaptive processing

zbMATH Keywords

computer science and technologyheterodyne receiversmathematical procedures and computer techniquesradio telescopes and instrumentation

Mathematics Subject Classification ID

General questions in astronomy and astrophysics (85A04) Computational methods for problems pertaining to astronomy and astrophysics (85-08) Experimental work for problems pertaining to astronomy and astrophysics (85-05)

Uses Software

• Thrust

This page was built for publication: Enabling a high throughput real time data pipeline for a large radio telescope array with GPUs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q548974)