English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Modern software systems are usually highly configurable, providing users withcustomized functionality through various configuration options. Understandinghow system performance varies with different option combinations is importantto determine optimal configurations that meet specific requirements. Due to thecomplex interactions among multiple options and the high cost of performancemeasurement under a huge configuration space, it is challenging to study howdifferent configurations influence the system performance. To address thesechallenges, we propose HINNPerf, a novel hierarchical interaction neuralnetwork for performance prediction of configurable systems. HINNPerf employsthe embedding method and hierarchic network blocks to model the complicatedinterplay between configuration options, which improves the prediction accuracyof the method. Besides, we devise a hierarchical regularization strategy toenhance the model robustness. Empirical results on 10 real-world configurablesystems show that our method statistically significantly outperformsstate-of-the-art approaches by achieving average 22.67% improvement inprediction accuracy. In addition, combined with the Integrated Gradientsmethod, the designed hierarchical architecture provides some insights about theinteraction complexity and the significance of configuration options, whichmight help users and developers better understand how the configurable systemworks and efficiently identify significant options affecting the performance.

HINNPerf: Hierarchical Interaction Neural Network for Performance Prediction of Configurable Systems