Back to CERN-HSF
GSoC 2026

EKO Oxidation: Maturing the Rust Core of EKO

EKO (Evolution Kernel Operators) is a foundational Python library used in high-precision particle physics that currently faces performance bottlenecks as computational demands grow. This project aims to "oxidize" EKO by maturing its initial Rust port into a production-ready, highly optimized backend. The solution focuses on minimizing cross-language friction by designing a zero-copy Python-Rust interface utilizing PyO3 and rust-numpy, implementing stateful workspaces for zero-allocation loops, and optimizing data layouts for CPU cache locality. Key deliverables include seamless Python-Rust integration, automated CI/CD packaging via Maturin, rigorous accuracy and memory benchmarking against the legacy Python codebase, and the establishment of stable C-ABI headers to allow direct ingestion by C++ and Fortran tools within the broader physics ecosystem.

Project details

Contributor

AkshatRai07

Mentors

Not available

Technologies

Not listed in the archive