Using bad learners to find good configurations

Using bad learners to find good configurations
复制标题

使用糟糕的学习器来找到好的配置

DOI:
10.1145/3106237.3106238
复制
发表时间:
2017
期刊:
Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering
影响因子:
--
通讯作者:
S. Apel
S. Apel
中科院分区:
--
文献类型:
--
作者:
V. Nair;T. Menzies;N. Siegmund;S. Apel

文献摘要

参考文献

被引文献

相似文献

对于给定的设置,找到软件系统的最佳性能配置通常是具有挑战性的。最近的方法通过基于配置样本集学习性能模型来解决这一挑战。然而,建立精确的性能模型可能非常昂贵(并且在实践中通常是不可行的)。本文的核心观点是,准确的性能值(例如,软件系统的响应时间)不需要对配置进行排序并识别最佳配置。正如我们的实验所示,学习成本低但不准确(相对于实际和预测性能之间的差异)的性能模型仍然可以用于排名配置,从而找到最佳配置。这种新颖的基于秩的方法使我们能够显着降低成本(在样品配置的测量数量方面)以及建立性能模型所需的时间。我们评估了我们的方法与21个场景的基础上9个软件系统,并证明我们的方法是有益的16个场景;对于其余5个场景,可以建立一个准确的模型,无论如何,使用很少的样本,而不需要一个基于排名的方法。
Finding the optimally performing configuration of a software system for a given setting is often challenging. Recent approaches address this challenge by learning performance models based on a sample set of configurations. However, building an accurate performance model can be very expensive (and is often infeasible in practice). The central insight of this paper is that exact performance values (e.g., the response time of a software system) are not required to rank configurations and to identify the optimal one. As shown by our experiments, performance models that are cheap to learn but inaccurate (with respect to the difference between actual and predicted performance) can still be used rank configurations and hence find the optimal configuration. This novel rank-based approach allows us to significantly reduce the cost (in terms of number of measurements of sample configuration) as well as the time required to build performance models. We evaluate our approach with 21 scenarios based on 9 software systems and demonstrate that our approach is beneficial in 16 scenarios; for the remaining 5 scenarios, an accurate model can be built by using very few samples anyway, without the need for a rank-based approach.
DOI: 10.1109/icse.2012.6227196
发表时间: 2012-06
期刊: 2012 34th International Conference on Software Engineering (ICSE)
影响因子: --
作者:
Norbert Siegmund;Sergiy S. Kolesnikov;Christian Kästner;S. Apel;D. Batory;Marko Rosenmüller;G. Saake
通讯作者: Norbert Siegmund;Sergiy S. Kolesnikov;Christian Kästner;S. Apel;D. Batory;Marko Rosenmüller;G. Saake
GALE:基于搜索的软件工程的几何主动学习
DOI: --
发表时间: 2015
影响因子: 7.4
作者:
Joseph Krall;T. Menzies;M. Davies
通讯作者: M. Davies
DOI: 10.1145/2786805.2786852
发表时间: 2015-08
期刊: Proceedings of the 2015 10th Joint Meeting on Foundations of Software Engineering
影响因子: --
作者:
Tianyin Xu;Long Jin;Xuepeng Fan;Yuanyuan Zhou;S. Pasupathy;Rukma Talwadker
通讯作者: Tianyin Xu;Long Jin;Xuepeng Fan;Yuanyuan Zhou;S. Pasupathy;Rukma Talwadker
DOI: 10.1109/ase.2015.45
发表时间: 2015-11
期刊: 2015 30th IEEE/ACM International Conference on Automated Software Engineering (ASE)
影响因子: --
作者:
Atrisha Sarkar;Jianmei Guo;Norbert Siegmund;S. Apel;K. Czarnecki
通讯作者: Atrisha Sarkar;Jianmei Guo;Norbert Siegmund;S. Apel;K. Czarnecki
可变性感知性能预测:一种统计学习方法
DOI: 10.1109/ase.2013.6693089
发表时间: 2013
期刊: 2013 28th IEEE/ACM International Conference on Automated Software Engineering (ASE)
影响因子: --
作者:
J. Guo;K. Czarnecki;S. Apel;N. Siegmund;A. Wasowski
通讯作者: A. Wasowski