# Database support of detector operation and data analysis in the DEAP-3600 Dark Matter experiment

## TL;DR

This paper details the integration of CouchDB and PostgreSQL databases into the DEAP-3600 dark matter experiment to manage vast data and metadata efficiently over a decade-long research project.

## Contribution

It introduces a unified database system combining structured and unstructured data management for large-scale, long-term physics experiments.

## Key findings

- Successful implementation of CouchDB and PostgreSQL for data management
- Enhanced data accessibility and operational efficiency
- Supports long-term data analysis and reporting

## Abstract

The DEAP-3600 detector searches for dark matter interactions on a 3.3 tonne liquid argon target. Over nearly a decade, from start of detector construction through the end of the data analysis phase, well over 200 scientists will have contributed to the project. The DEAP-3600 detector will amass in excess of 900 TB of data representing more than 10$^{10}$ particle interactions, a few of which could be from dark matter. At the same time, metadata exceeding 80 GB will be generated. This metadata is crucial for organizing and interpreting the dark matter search data and contains both structured and unstructured information. The scale of the data collected, the important role of metadata in interpreting it, the number of people involved, and the long lifetime of the project necessitate an industrialized approach to metadata management. We describe how the CouchDB and the PostgreSQL database systems were integrated into the DEAP detector operation and analysis workflows. This integration provides unified, distributed access to both structured (PostgreSQL) and unstructured (CouchDB) metadata at runtime of the data analysis software. It also supports operational and reporting requirements.