cTreeBalls: a fast 3-point correlation function code for clustering measurements

TL;DR

cTreeBalls is a high-performance Python/C package that efficiently computes 2- and 3-point clustering statistics for large sky simulations, supporting LSST data analysis.

Contribution

It introduces a fast, scalable algorithm combining octree and kd-tree methods with a user-friendly interface for large-scale clustering measurements.

Findings

Calculates 3-point correlations of over 200 million pixels in under 10 minutes.

Supports analysis of full-sky simulations with Nside=4096.

Includes methods for two-point clustering in periodic boxes.

Abstract

cTreeBalls (cBalls for short) is a Python/C package useful to measure (2,3)-point clustering statistics. cBalls can efficiently calculate 3-point correlations of more than 200 million HEALPix pixels ( a full sky simulation with Nside = 4096) in less than 10 minutes on a single high-performance computing node, enabling a feasible analysis for the upcoming LSST data. It builds upon octree (Barnes & Hut, 1986) and kd-tree algorithms (Bentley, 1975), and supplies a user-friendly interface with flexible input/output (I/O) of catalogue data and measurement results, with the built program configurable through external parameter files and tracked through enhanced logging and warning/exception handling. For completeness and complementarity, methods for measuring two-point clustering statistics for periodic boxes are also included in the package. cTreeBalls was developed for its use in the Dark Energy Science Collaboration (DESC) of the Rubin Observatory Legacy Survey of Space and Time (LSST).