GPU-accelerated observational forecasting of superconducting (sub)millimeter integral field units with GATEAU

Abstract

Scientific interpretation of data obtained with superconducting (sub)millimeter integral field units (IFUs) requires careful design of reduction and analysis pipelines. However, the quality and behavior of the designed pipelines must be assessed before they are applied, requiring simulated observational data. To this end, we have developed GATEAU, a GPU-accelerated software framework for simulating realistic (sub)millimeter observations of arbitrary astronomical sources with IFUs using MKIDs. The framework consists of a Python interface powered by a C/C++ backend, which achieves GPU-acceleration through Nvidia’s CUDA library. The output is stored in FITS format, which the user can feed to their pipeline. We present the software implementation of GATEAU and several comparisons between GATEAU and on-sky DESHIMA 2.0 data, serving as validation tests. Finally, we present a use case focusing on the TIFUUN instrument, illustrating the kind of analysis that can be done with GATEAU.

Authors 7

1. Arend Moerman

   Delft University of Technology

   Affiliation as printed

   Technische Univ. Delft (Netherlands)

2. Nikita A. Soshnin

   Delft University of Technology

   Affiliation as printed

   Technische Univ. Delft (Netherlands)

3. Stefanie Brackenhoff

   Delft University of Technology

   Affiliation as printed

   Technische Univ. Delft (Netherlands)

4. K. Karatsu Aachen SRON Space Research Organization Netherlands

   Space Research Organisation Netherlands · Delft University of Technology

   Affiliation as printed

   SRON Space Research Organization Netherlands (Netherlands)

   Technische Univ. Delft (Netherlands)

5. Louis Marting Aachen SRON Space Research Organization Netherlands

   Space Research Organisation Netherlands · Delft University of Technology

   Affiliation as printed

   SRON Space Research Organization Netherlands (Netherlands)

   Technische Univ. Delft (Netherlands)

6. Bernhard Brandl Aachen Leiden Univ. (Netherlands)

   Leiden University · Delft University of Technology

   Affiliation as printed

   Leiden Univ. (Netherlands)

   Technische Univ. Delft (Netherlands)

7. Akira Endo

   Delft University of Technology

   Affiliation as printed

   Technische Univ. Delft (Netherlands)