Lyncs-API: a Python API for Lattice QCD applications

TL;DR

Lyncs-API is a Python framework that unifies and simplifies access to various Lattice QCD libraries, enabling performance-portable, high-level workflows across CPU and GPU architectures.

Contribution

It introduces a flexible, high-level Python API that integrates multiple Lattice QCD libraries with performance portability and automated bindings, enhancing usability and extensibility.

Findings

Achieves performance portability across CPU and GPU architectures.

Provides automated bindings for multiple Lattice QCD libraries.

Facilitates high-level workflows with distributed computing tools.

Abstract

We present Lyncs-API, a Python API for Lattice QCD applications currently under development. Lyncs aims to bring several widely used libraries for Lattice QCD under a common framework. Lyncs flexibly links to libraries for CPUs and GPUs in a way that can accommodate additional computing architectures as these arise, achieving performance-portability for the calculations while maintaining the same high-level workflow. Lyncs distributes calculations using Dask and mpi4py, with bindings to the libraries automatically generated by cppyy. While Lyncs is designed to allow linking to multiple libraries, we focus on a set of targeted packages that include DDalphaAMG, tmLQCD, QUDA and c-lime. More libraries will be added in the future. We also develop generic-purpose tools for facilitating the usage of Python in Lattice QCD and HPC in general. The project is open-source, community-oriented and available on Github.