MindOpt Tuner: Boost the Performance of Numerical Software by Automatic Parameter Tuning

TL;DR

MindOpt Tuner is an automatic parameter tuning tool that significantly enhances the performance of numerical software like solvers, achieving over 100x speedup and outperforming existing tuning tools.

Contribution

It introduces a versatile, cloud-enabled tuning system with web and API interfaces, demonstrating superior efficiency and performance improvements over prior methods.

Findings

Over 100x acceleration on some problem instances.

Higher tuning efficiency than SMAC3.

Effective tuning for a range of numerical software.

Abstract

Numerical software is usually shipped with built-in hyperparameters. By carefully tuning those hyperparameters, significant performance enhancements can be achieved for specific applications. We developed MindOpt Tuner, a new automatic tuning tool that supports a wide range of numerical software, including optimization and other solvers. MindOpt Tuner uses elastic cloud resources, features a web-based task management panel and integration with ipython notebook with both command-line tools and Python APIs. Our experiments with COIN-OR Cbc, an open-source mixed-integer optimization solver, demonstrate remarkable improvements with the tuned parameters compared to the default ones on the MIPLIB2017 test set, resulting in over 100x acceleration on several problem instances. Additionally, the results demonstrate that Tuner has a higher tuning efficiency compared to the state-of-the-art automatic tuning tool SMAC3.