Beowulf computational class in Wide-Field Astronomy

Enrico Cascone; Aniello Grado; Mike Pavlov; Giulio Capasso

doi:10.1117/12.461097

13 December 2002 Beowulf computational class in Wide-Field Astronomy

Enrico Cascone, Aniello Grado, Mike Pavlov, Giulio Capasso

Proceedings Volume 4848, Advanced Telescope and Instrumentation Control Software II; (2002) https://doi.org/10.1117/12.461097
Event: Astronomical Telescopes and Instrumentation, 2002, Waikoloa, Hawai'i, United States

Abstract

Nowadays, computer clusters offer the best price/performance ratio for a huge range of scientific and industrial applications. Among the various cluster architectures so far proposed, the most promising surely is the Beowulf: a distributed computational system formed by a server plus one or more client nodes connected to the server through a fast communication network. A Beowulf computer class looks like the most promising solution to cope with the processing of the enormous amount of data produced by the present and, even more, planned Wide Field Telescopes. Indeed Wide Field cameras are built as CCD mosaic and the nature embarrassingly parallel of such data make a parallel machine particularly suitable to process them as most of the data process can be execute independently on each CCD of the mosaic. A Beowulf cluster of one master plus eight nodes has been installed at the Astronomical Observatory of Naples. We will present results on tests performed on various hardware configurations in order to find the cluster configuration optimized, in terms of performance, for the specific application of reducing Wide Field images. A comparison with other hardware platform will also be presented.

Citation Download Citation

Enrico Cascone, Aniello Grado, Mike Pavlov, and Giulio Capasso "Beowulf computational class in Wide-Field Astronomy", Proc. SPIE 4848, Advanced Telescope and Instrumentation Control Software II, (13 December 2002); https://doi.org/10.1117/12.461097

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available