MMO: Meta Multi-Objectivization for Software Configuration Tuning

TL;DR

The paper tackles local optima in software configuration tuning by meta-multi-objectivizing at the optimization-model level rather than refining the optimizer itself. It introduces the MMO framework with a normalization that obviates the weight parameter and uses an auxiliary objective to diversify similarly-performing configurations, preventing Pareto-equal clustering. Across 22 cases from 11 real-world systems, MMO with the new normalization often surpasses single-objective baselines and prior MMO variants, delivering up to 2.09x speedups and reducing the resource cost of weight tuning, while also consolidating outputs from model-based tuners such as Flash and BOCA. The work offers a promising direction for SBSE by generalizing MMO to more auxiliary objectives and extending to other problems, potentially improving robustness and efficiency of configuration tuning.

Abstract

Software configuration tuning is essential for optimizing a given performance objective (e.g., minimizing latency). Yet, due to the software's intrinsically complex configuration landscape and expensive measurement, there has been a rather mild success, particularly in preventing the search from being trapped in local optima. To address this issue, in this paper we take a different perspective. Instead of focusing on improving the optimizer, we work on the level of optimization model and propose a meta multi-objectivization (MMO) model that considers an auxiliary performance objective (e.g., throughput in addition to latency). What makes this model distinct is that we do not optimize the auxiliary performance objective, but rather use it to make similarly-performing while different configurations less comparable (i.e. Pareto nondominated to each other), thus preventing the search from being trapped in local optima. Importantly, by designing a new normalization method, we show how to effectively use the MMO model without worrying about its weight -- the only yet highly sensitive parameter that can affect its effectiveness. Experiments on 22 cases from 11 real-world software systems/environments confirm that our MMO model with the new normalization performs better than its state-of-the-art single-objective counterparts on 82% cases while achieving up to 2.09x speedup. For 68% of the cases, the new normalization also enables the MMO model to outperform the instance when using it with the normalization from our prior FSE work under pre-tuned best weights, saving a great amount of resources which would be otherwise necessary to find a good weight. We also demonstrate that the MMO model with the new normalization can consolidate recent model-based tuning tools on 68% of the cases with up to 1.22x speedup in general.