Evaluating #SAT solvers on industrial feature models

Sundermann, Chico; Thüm, Thomas; Schaefer, Ina

doi:10.1145/3377024.3377025

Cited by 28 publications

(29 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent work, we tried to create a BDD for hundreds of large feature models and failed for 98% of them (i.e., 100% of models having more than 550 features) [80]. Furthermore, even though other knowledge compilation techniques scaled to many large feature models, not a single knowledge compilation tool scaled to Linux when we tried to compute the number of valid configurations [80]. Similarly, t-wise sampling algorithms typically do not scale to Linux [69].…”

Section: Motivationmentioning

confidence: 99%

“…Second, there is a considerable amount of research that is based on the translation of the feature model into a BDD. In the past two decades, researchers proposed the use of BDDs to count the number of valid configurations [8,10,47,61,70,80], to compute feature-model slices [1] and differences [2], for interactive product configuration [44], to check whether product-line artifacts are consistent [29,79], to parse preprocessor-based product lines [41], to simplify preprocessor annotations [91,93], and to lift test-suite generation [15], data-flow analyses [12,13], or model checking [5, 7, 17, 18, 20, 22-25, 43, 90, 91] to product lines. If we aim to apply that research to Linux or similarly complex configuration spaces, it is an open question whether BDDs can be created for them.…”

Section: Motivationmentioning

confidence: 99%

“…Czarnecki and Pietroszek used BDDs for consistency checking in model-based product lines and evaluated their approach on an e-commerce platform with about 200 features [29]. Numerous authors used BDDs to compute the number of valid configurations [10,55,70,80]. Benavides et al employed multiple solvers, such as SAT, CSP, and BDDs for the analysis of feature models in the FaMa framework [8,10].…”

Section: State-of-the-artmentioning

confidence: 99%

“…Pohl et al investigated nine knowledge compilation techniques, including BDDs, CSP, and SAT solvers to compute the number of valid configurations [70]. While they only evaluated small academic product lines from the SPLOT repository, we recently extended their study with the largest known feature models from open-source projects and proprietary models [80]. While some knowledge compilation techniques scaled to all systems except Linux and one automotive product line, BDDs did not scale for almost all models.…”

Section: State-of-the-artmentioning

confidence: 99%

See 3 more Smart Citations

A BDD for Linux?

Thüm

2020

Proceedings of the 24th ACM Conference on Systems and Software Product Line: Volume a - Volume A

Self Cite

View full text Add to dashboard Cite

What is the number of valid configurations for Linux? How to generate uniform random samples for Linux? Can we create a binary decision diagram for Linux? It seems that the product-line community tries hard to answer such questions for Linux and other configurable systems. However, attempts are often not published due to the publication bias (i.e., unsuccessful attempts are not published). As a consequence, researchers keep trying by potentially spending redundant effort. The goal of this challenge is to guide research on these computationally complex problems and to foster the exchange between researchers and practitioners.

show abstract

Section: Motivationmentioning

confidence: 99%

Section: Motivationmentioning

confidence: 99%

Section: State-of-the-artmentioning

confidence: 99%

Section: State-of-the-artmentioning

confidence: 99%

See 2 more Smart Citations

A BDD for Linux?

Thüm

2020

Proceedings of the 24th ACM Conference on Systems and Software Product Line: Volume a - Volume A

Self Cite

View full text Add to dashboard Cite

show abstract

“…The time needed by our modules to provide a solution varies according to the size of the problem [34]. This section evaluates their…”

Section: Scalabilitymentioning

confidence: 99%

Supporting the evolution of applications deployed on edge-based infrastructures using multi-layer feature models

Cañete

Amor

Fuentes

2020

Proceedings of the 24th ACM International Systems and Software Product Line Conference - Volume B

View full text Add to dashboard Cite

The proliferation of cyber-physical systems has encouraged the emergence of new technologies and paradigms to improve the performance of IoT-based applications. Edge Computing proposes using the nearby devices in the frontier/Edge of the access network for deploying application tasks. However, the functionality of cyberphysical systems, which is usually distributed in several devices and computers, imposes specific requirements on the infrastructure to run properly. The evolution of an application to meet new user requirements and the high diversity of hardware and software technologies in the edge can complicate the deployment of evolved applications. The aim of our approach is to apply Multi Layer Feature Models, which capture the variability of applications and the infrastructure, to support the deployment in edge-based environments of cyber-physical systems applications. This separation can support the evolution of application and infrastructure. Considering that IoT/Edge/Cloud infrastructures are usually shared by many applications, the SPL deployment process has to assure that there will be enough resources for all of them, informing developers about the alternatives of deployment. Prior to its deployment and leaning on the infrastructure feature models, the developer can calculate what is the configuration of minimal set of devices supporting application requirements of the evolved application. In addition, the developer can find which is the application configuration that can be hosted in the current evolved infrastructure. CCS CONCEPTS • Software and its engineering → Software product lines; • Computer systems organization → Distributed architectures; Embedded and cyber-physical systems.

show abstract