CosmoDM and its application to Pan-STARRS data

TL;DR

CosmoDM is an automated, scalable data management system for optical survey data processing, calibration, and catalog creation, demonstrated on Pan-STARRS data for galaxy cluster analysis.

Contribution

The paper introduces CosmoDM, a flexible, high-throughput data management system with novel image compression features, applied to Pan-STARRS data for cosmological research.

Findings

Successfully processed Pan-STARRS data with CosmoDM

Confirmed and measured photometric redshifts of galaxy clusters

Demonstrated scalability and efficiency of the system

Abstract

The Cosmology Data Management system (CosmoDM) is an automated and flexible data management system for the processing and calibration of data from optical photometric surveys. It is designed to run on supercomputers and to minimize disk I/O to enable scaling to very high throughput during periods of reprocessing. It serves as an early prototype for one element of the ground-based processing required by the Euclid mission and will also be employed in the preparation of ground based data needed in the eROSITA X-ray all sky survey mission. CosmoDM consists of two main pipelines. The first is the single-epoch or detrending pipeline, which is used to carry out the photometric and astrometric calibration of raw exposures. The second is the co- addition pipeline, which combines the data from individual exposures into deeper coadd images and science ready catalogs. A novel feature of CosmoDM is that it uses a modified stack of As- tromatic software which can read and write tile compressed images. Since 2011, CosmoDM has been used to process data from the DECam, the CFHT MegaCam and the Pan-STARRS cameras. In this paper we shall describe how processed Pan-STARRS data from CosmoDM has been used to optically confirm and measure photometric redshifts of Planck-based Sunyaev-Zeldovich effect selected cluster candidates.