一句话：提出 HyPerf 贝叶斯多级性能模型，区分设置不变与可变影响，用更少样本实现准确预测。
调优软件配置对满足性能需求至关重要，但性能还受工作负载等外部因素影响，需理解外部因素设置与配置组合如何共同影响性能。本文提出 HyPerf，一种贝叶斯多级性能建模方法，区分设置不变与设置可变影响，旨在以更少训练样本实现稳健准确的性能预测。与现有方法不同，HyPerf 能识别捕获关键性能变化的最小设置集，帮助判断是否已考虑所有设置可变影响。在十个真实软件系统（最多35种工作负载）及 Linux 内核可扩展性实验上，HyPerf 在更少测量下匹配或优于当前最优方法，并支持可解释推理，识别捕获关键性能变化的最小工作负载子集。
Tuning a software system’s configuration is essential to meet performance requirements. However, not only do configuration options affect performance, but also the system’s interaction with external factors such as the workload. Hence, tuning requires understanding how a specific setting of external factors (e.g., a specific workload) in combination with the system configuration influences performance. To address this issue, we propose HyPerf , a Bayesian multi-level performance modeling approach that systematically distinguishes between setting-invariant and setting-variant influences, that is, influences that remain consistent across settings versus those that exhibit substantial variation. With HyPerf , we aim at balancing accuracy and efficiency, achieving robust performance predictions with significantly fewer training samples. Unlike the state of the art, HyPerf is able to identify a minimal set of settings that captures essential performance variations, so that developers can approximate whether all setting-variant influences have been accounted for. Empirical evaluations on ten real-world software systems across up to 35 workloads and scalability experiments on the Linux kernel demonstrate that HyPerf matches or outperforms state-of-the-art approaches while requiring fewer measurements. Notably, HyPerf is indeed capable of interpretable performance reasoning and can identify minimal workload subsets that capture essential performance variations.
@inproceedings{paperbot11630,
title = {Bayesian Multi-Level Performance Models for Multi-Factor Variability of Configurable Software Systems},
author = {Johannes Dorn and Stefan Mühlbauer and Stefan Jahns and Sven Apel and Norbert Siegmund},
booktitle = {ICSE},
year = {2026},
doi = {10.1145/3744916.3773131}
}