MaRDI Open Interfaces for Interoperable Nonlinear Optimization

2026-06-18

AI Summary

This work addresses the lack of interoperability among nonlinear optimization solvers in scientific computing, which typically necessitates extensive code refactoring and revalidation when switching solvers or invoking them across programming languages. To overcome this limitation, the authors propose a universal interface framework tailored for nonlinear optimization problems. The framework employs a modular architecture that integrates multi-language bindings and an automated data marshaling mechanism, enabling plug-and-play solver integration and seamless cross-language invocation. By abstracting solver-specific implementation details, the framework substantially reduces the development and verification overhead associated with solver substitution and cross-language collaboration, thereby significantly enhancing the iteration efficiency of scientific computing workflows.

Abstract

MaRDI Open Interfaces is a software package that aims to improve interoperability in scientific computing, particularly, for nonlinear optimization. To this end, this package holds two main characteristics. First, it provides unified interfaces for typical numerical problems to help switching between solvers for the same problem type. Second, it automates data marshalling between programming languages. Hence, computational scientists can conduct experiments faster by using the package, with fewer code-modification and testing efforts. In this work we describe the general structure of the software package and show examples with the interface for nonlinear optimization.

Problem

• interoperability
• nonlinear optimization
• scientific computing
• solver switching
• data marshalling

Innovation

• interoperability
• nonlinear optimization
• unified interfaces
• data marshalling
• scientific computing