Scaling exact multi-objective combinatorial optimization by parallelization

Guo, Jianmei; Zulkoski, Edward; Olaechea, Rafael; Rayside, Derek; Czarnecki, Krzysztof; Apel, Sven; Atlee, Joanne M.

doi:10.1145/2642937.2642971

Cited by 32 publications

(26 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beside various facets of performance, performance-influence models may be beneficial to reason about other non-functional properties and quality attributes, most notably, energy consumption. Moreover, we can supply the models we learned to other performance-modeling and optimization tools, such as Clafer [19] and EPOAL [9].…”

Section: Discussionmentioning

confidence: 99%

Performance-influence models for highly configurable systems

Siegmund

Grebhahn

Apel

et al. 2015

Proceedings of the 2015 10th Joint Meeting on Foundations of Software Engineering

Self Cite

200

245

View full text Add to dashboard Cite

Almost every complex software system today is configurable. While configurability has many benefits, it challenges performance prediction, optimization, and debugging. Often, the influences of individual configuration options on performance are unknown. Worse, configuration options may interact, giving rise to a configuration space of possibly exponential size. Addressing this challenge, we propose an approach that derives a performance-influence model for a given configurable system, describing all relevant influences of configuration options and their interactions. Our approach combines machine-learning and sampling heuristics in a novel way. It improves over standard techniques in that it (1) represents influences of options and their interactions explicitly (which eases debugging), (2) smoothly integrates binary and numeric configuration options for the first time, (3) incorporates domain knowledge, if available (which eases learning and increases accuracy), (4) considers complex constraints among options, and (5) systematically reduces the solution space to a tractable size. A series of experiments demonstrates the feasibility of our approach in terms of the accuracy of the models learned as well as the accuracy of the performance predictions one can make with them.

show abstract

Section: Discussionmentioning

confidence: 99%

Performance-influence models for highly configurable systems

Siegmund

Grebhahn

Apel

et al. 2015

Proceedings of the 2015 10th Joint Meeting on Foundations of Software Engineering

Self Cite

200

245

View full text Add to dashboard Cite

show abstract

“…The data set covers five large constrained real‐world FMs taken from the LVAT repository . Parts of FMs have been used widely to evaluate different SPLs configuration optimization problems in a series of papers . We report the software version, the number of features, and the number of cross‐tree constraints for each subject FM in Table (columns 1 to 4).…”

Section: Discussionmentioning

confidence: 99%

“…For instance, the L inux SPL, acquired from the Linux variability analysis tools (LVAT) repository, contains 6888 features and 343 944 constraints. In general, a configuration generated by a certain optimization algorithm must be validated to satisfy the predefined constraints . Due to the complexity of constraint solving, the configuration validation task on large SPLs usually depends on automated tool support, such as off‐the‐shelf constraint solvers …”

Section: Introductionmentioning

confidence: 99%

A parallel portfolio approach to configuration optimization for large software product lines

Shi

Guo

et al. 2018

Softw Pract Exp

Self Cite

View full text Add to dashboard Cite

Summary Software product line (SPL) engineering demands for optimal or near‐optimal products that balance multiple often competing and conflicting objectives. A major challenge for large SPLs is to efficiently explore a huge space of various products and satisfy a large number of predefined constraints simultaneously. To improve the optimality and convergence speed, we propose a parallel portfolio approach, called IBEAPORT, which designs three algorithm variants by incorporating constraint solving into the indicator‐based evolutionary algorithm in different ways and performs these variants by utilizing parallelization techniques. Our approach utilizes the exploration capabilities of different algorithms and improves optimality as far as possible within a limited time budget. We evaluate our approach on five large‐scale real‐world SPLs. Empirical results demonstrate that our approach significantly outperforms the state of the art for all five SPLs on a quality indicator and a diversity indicator. Moreover, IBEAPORT quickly converges to a relatively stable hypervolume value even for the largest SPL with 6888 features.

show abstract

“…In our research we tackle a different problem but also having in mind that distributing analysis tasks helps in terms of time to perform the analysis; even though, since PAVIA is a MapReduce application on feature models, we step out the distributed and parallel computing issues. Guo et al [8] addressed the problem of multi-objective combinatorial optimization. They developed parallelization algorithms and show substantial gains in three case studies.…”

Section: Related Workmentioning

confidence: 99%

“…Feature models delimit the scope of a configurable system (i.e., an SPL) and formally document what configurations are supported. Once specified, feature models can be used for model checking an SPL [24], for testing SPLs, for automating product configuration [8], or for computing relevant information [4].…”

Section: Introductionmentioning

confidence: 99%

Exploiting the enumeration of all feature model configurations

Galindo

Acher

Tirado³

et al. 2016

Proceedings of the 20th International Systems and Software Product Line Conference

View full text Add to dashboard Cite

Feature models are widely used to encode the configurations of a software product line in terms of mandatory, optional and exclusive features as well as propositional constraints over the features. Numerous computationally expensive procedures have been developed to model check, test, configure, debug, or compute relevant information of feature models. In this paper we explore the possible improvement of relying on the enumeration of all configurations when performing automated analysis operations. The key idea is to precompile configurations so that reasoning operations (queries and transformations) can then be performed in polytime. We tackle the challenge of how to scale the existing enumeration techniques. We show that the use of distributed computing techniques might offer practical solutions to previously unsolvable problems and opens new perspectives for the automated analysis of software product lines.

show abstract

Scaling exact multi-objective combinatorial optimization by parallelization

Cited by 32 publications

References 46 publications

Performance-influence models for highly configurable systems

Performance-influence models for highly configurable systems

A parallel portfolio approach to configuration optimization for large software product lines

Exploiting the enumeration of all feature model configurations

Contact Info

Product

Resources

About