Abstract

Contributed Talk - Splinter DataManage

Thursday, 12 September 2024, 14:45   (S22)

The radio astronomy workflow on Compute4PUNCH

Prateek Gupta^1, Alexander Drabent^1, Matthias Hoeft^1, Manuel Giffels^2, Benoit Roland^2, Christoph Wissing^3
1. Thüringer Landessternwarte Tautenburg, Sternwarte 5, 07778 Tautenburg, Germany 2. Karlsruhe Institute of Technology (KIT), Scientific Computing Center (SCC), Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany 3. Deutsches Elektronen-Synchrotron (DESY), Notkestraße 85, 22607 Hamburg, Germany

LOFAR is a novel, fully digitised radio telescope, recording 5-15 TB per regular observation, consisting of 52 stations across Europe. The recorded data are stored in long-term archives at three different sites: Jülich, Poznań, Amsterdam. Data reduction of such large amounts of data is a real challenge. In addition to efficient data management, these tasks require powerful computing resources to produce sky brightness maps in a reasonable time. This is where the Compute4PUNCH infrastructure can be used. The Compute4PUNCH (C4P) infrastructure consists of high performance computing machines offered under the umbrella of the PUNCH4NFDI project, suitable for such a use case. The prototype workflow has been developed within PUNCH, where the staged radio data transfer from the archive to Storage4PUNCH (S4P) and further reduced by the LOFAR Initial Calibration Pipeline on C4P. This talk is intended to discuss the preliminary results of the initial observations processed with the developed proto-type workflow.