Affordable and sustainable large-scale distributed radio telescope processing

Abstract

Producing science-ready data products from large-scale radio telescopes is computationally, financially and environmentally expensive. In light of the current climate crisis, we can no longer accept that improving scientific instruments inevitably results in an unconstrained increase in processing requirements. Starting this year, ASTRON aims to increase the capabilities of the international LOFAR telescope, within the current energy budget. In this talk we will highlight some envisioned challenges and we will discuss in some detail the efforts to achieve this ambitious goal. We will take a look at the technologies used to improve efficiency and measure the environmental impact of various operations in the signal processing pipeline. A key component of this is operator and end-user awareness. We will close with an outlook at the applicability of emerging and future technologies and their impact on environmental impact.

Authors 1

1. P. Chris Broekema corresponding Aachen Leiden Institute of Advanced Computer Science

   Leiden University · Netherlands Institute for Radio Astronomy

   Affiliation as printed

   ASTRON (Netherlands)

   Leiden Institute of Advanced Computer Science (Netherlands)