Filters








577,132 Hits in 6.8 sec

Performance-influence models for highly configurable systems

Norbert Siegmund, Alexander Grebhahn, Sven Apel, Christian Kästner
2015 Proceedings of the 2015 10th Joint Meeting on Foundations of Software Engineering - ESEC/FSE 2015  
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  ...  Almost every complex software system today is configurable. While configurability has many benefits, it challenges performance prediction, optimization, and debugging.  ...  Batory for comments on earlier drafts of this paper, A. Simbürger for his help with the measurements, and the Jülich Supercomputing Center for providing access to the supercomputer JuQueen.  ... 
doi:10.1145/2786805.2786845 dblp:conf/sigsoft/SiegmundGAK15 fatcat:34j2frnsbjgwrh3qg37bymht5q

Performance-Influence Models

Norbert Siegmund, Alexander Grebhahn, Sven Apel, Christian Kästner
2016 Software Engineering  
Even domain experts and the developers themselves often do not (fully) understand the performance influences of all configuration options and their combined influence when options interact.  ...  End-users, developers, and administrators are often overwhelmed with the possibilities to configure asoftware system.  ...  Performance-influence models are meant to ease understanding,debugging,and optimization of highly configurable software systems.  ... 
dblp:conf/se/SiegmundGAK16 fatcat:ryvzaj4qtnh37hb4de6bgajh2i

Transfer Learning for Performance Modeling of Configurable Systems: An Exploratory Analysis [article]

Pooyan Jamshidi, Norbert Siegmund, Miguel Velez, Christian Kästner, Akshay Patel, Yuvraj Agarwal
2017 arXiv   pre-print
Modern software systems provide many configuration options which significantly influence their non-functional properties.  ...  While this line of research is promising to learn more accurate models at a lower cost, it is unclear why and when transfer learning works for performance modeling.  ...  We would like to thank Tim Menzies, Vivek Nair, Wei Fu, and Gabriel Ferreira for their feedback.  ... 
arXiv:1709.02280v1 fatcat:44hubsccozegrcgnmfscjzq7ba

White-Box Analysis over Machine Learning: Modeling Performance of Configurable Systems [article]

Miguel Velez, Pooyan Jamshidi, Norbert Siegmund, Sven Apel, Christian Kästner
2021 arXiv   pre-print
Performance-influence models can help stakeholders understand how and where configuration options and their interactions influence the performance of a system.  ...  We present Comprex, a white-box approach to build performance-influence models for configurable systems, combining insights of local measurements, dynamic taint analysis to track options in the implementation  ...  We thank Chu-Pan Wong, Jens Meinicke, and Florian Sattler for their comments during the development of this work, the FOSD 2019 meeting participants for their feedback on the iterative dynamic taint analysis  ... 
arXiv:2101.05362v1 fatcat:dcya4usgsze7boggmrrwb7tveq

On the Relation of External and Internal Feature Interactions: A Case Study [article]

Sergiy Kolesnikov, Norbert Siegmund, Christian Kästner, Sven Apel
2018 arXiv   pre-print
Detecting feature interactions is imperative for accurately predicting performance of highly-configurable systems.  ...  We expect that the information about potentially interacting features can be obtained by statically analyzing the source code of a highly-configurable system, which is computationally cheaper than performing  ...  highly configurable systems.  ... 
arXiv:1712.07440v2 fatcat:sgrhbrxyqbf3xohmkwq376xh3i

Unicorn: Reasoning about Configurable System Performance through the lens of Causality [article]

Md Shahriar Iqbal, Rahul Krishna, Mohammad Ali Javidian, Baishakhi Ray, Pooyan Jamshidi
2022 arXiv   pre-print
Understanding and reasoning about the performance behavior of highly configurable systems, over a vast and variable space, is challenging.  ...  Further, unlike the existing methods, the learned causal performance models reliably predict performance for new environments.  ...  For highly configurable systems, gathering high-quality data is challenging.  ... 
arXiv:2201.08413v2 fatcat:i45rkdbsjbdtregsnyj6ydrmoy

Learning Contextual-Variability Models

Paul Temple, Mathieu Acher, Jean-Marc Jezequel, Olivier Barais
2017 IEEE Software  
Modeling how contextual factors relate to the configuration space of a software system is most of the time a manual and error-prone task, highly dependent on expert knowledge.  ...  As a result, software developers and product managers can automatically extract the rules that specialize highly-configurable systems for operating on specific contexts.  ...  It can be tuned for internally influencing the results of subsequent computations.  ... 
doi:10.1109/ms.2017.4121211 fatcat:ykrd4uordrep3n2baiyt3rpucq

Experiments on Optimizing the Performance of Stencil Codes with SPL Conqueror

Alexander Grebhahn, Sebastian Kuckuk, Christian Schmitt, Harald Köstler, Norbert Siegmund, Sven Apel, Frank Hannig, Jürgen Teich
2014 Parallel Processing Letters  
For HSMGP, we predict performance with an accuracy of 97 % including the performance-optimal configuration, while measuring 3.2 % of all configurations.  ...  For DUNE, we predict performance of all configurations with an accuracy of 86 % after measuring 3.3 % of all configurations.  ...  Acknowledgments We thank Richard Membarth for providing the tested multi-grid implementation in HIPA cc as well as for advice on suitable evaluation parameters; we thank Peter Bastian for providing the  ... 
doi:10.1142/s0129626414410011 fatcat:ccakutmjkbdk7cvdjosrbrii4u

ACTS in Need

Yuqing Zhu, Jianxun Liu, Mengying Guo, Wenlong Ma, Yungang Bao
2017 Proceedings of the 8th Asia-Pacific Workshop on Systems - APSys '17  
However, finding the best setting for the tens or hundreds of configuration parameters is mission impossible for ordinary users.  ...  As these co-deployed systems can interact to affect the overall performance, they must be tuned together.  ...  ACKNOWLEDGMENTS We would like to thank our shepherd, Cheng Li, and the anonymous reviewers for their constructive comments and inputs to improve our paper.  ... 
doi:10.1145/3124680.3124730 dblp:conf/apsys/ZhuLGMB17 fatcat:uo3otpgkhff5jaijwqxgmdglg4

Optimal Configuration in Production Planning and Control

Christoph Wolfsgruber, Gerald Lichtenegger
2016 Berg- und Huttenmännische Monatshefte (BHM)  
Based on these fundamental insights, a framework for an optimal configuration in production planning and control is presented.  ...  The decision on the optimal PPC configuration should be based on the overall system costs for the information system, the PPC system, the resulting system performance and the costs for the indirect factors  ...  Despite the best practices in PPC configuration, standard measures using lean tools and complexity management approaches have to be taken to gain control over these highly influencing factors.  ... 
doi:10.1007/s00501-016-0474-6 fatcat:2indpzcirfezpjm6cjogmrhkdq

Reliability analysis of shear strengthening externally bonded FRP models

João L. Lima, Joaquim António Barros
2011 Proceedings of the Institution of Civil Engineers : Structures and buildings  
models was highly unconservative, which may be a serious concern as these formulations may be currently being used in design practice.  ...  Guide for the design and construction of externally bonded FRP systems for strengthening of concrete structures. Reported by ACI Committee 440. Ali, M., Oehlers, D. and Seracino, R. (2006).  ...  Fig. 8 -Fig. 9 - 89 PAS and PSS ratio variation for different configurations with the: a) IDB, b) RDB Influence of strengthening configuration on the predictive performance of the models Fig. 10 -Influence  ... 
doi:10.1680/stbu.9.00042 fatcat:nkhs2dncyvajhje4a5wttifg3y

Learning to sample: exploiting similarities across environments to learn performance models for configurable systems

Pooyan Jamshidi, Miguel Velez, Christian Kästner, Norbert Siegmund
2018 Proceedings of the 2018 26th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering - ESEC/FSE 2018  
The increased flexibility raises challenges for understanding the configuration space and the effects of options and their interactions on performance and other non-functional properties.  ...  To identify how options and interactions affect the performance of a system, several sampling and learning strategies have been recently proposed.  ...  THE BIG PICTURE 2.1 Performance Analysis For understanding the performance behavior of a software system, performance models that predict the performance of the system in different configurations using  ... 
doi:10.1145/3236024.3236074 dblp:conf/sigsoft/JamshidiVKS18 fatcat:ctbkc6uqxvexlbzcivw5mtmjl4

Monte Carlo Simulations for Variability Analyses in Highly Configurable Systems

José Miguel Horcas Aguilera, A. Germán Márquez, José A. Galindo, David Benavides
2021 International Configuration Workshop  
Highly configurable systems expose numerous variation points to be configured by the stakeholders.  ...  Deciding which variant to select for a given variation point is hard to know a priori because each variant affects the configuration properties (e.g., performance, efficiency, fault tolerance) differently  ...  Troyano for having inspired us in the usage of Monte Carlo methods in software product line analyses.  ... 
dblp:conf/confws/AguileraMG021 fatcat:xvnall2eurg5ziceyyseut5vpa

Transfer Learning for Improving Model Predictions in Highly Configurable Software [article]

Pooyan Jamshidi, Miguel Velez, Christian Kästner, Norbert Siegmund, Prasad Kawthekar
2017 arXiv   pre-print
Usually, we learn a black-box model based on real measurements to predict the performance of the system given a specific configuration.  ...  In a self-adaptation context, we are often interested in reasoning about the performance of the systems under different configurations.  ...  To do so, we use black-box performance models that describe how configuration options and their interactions influence the performance of a system (e.g., execution time).  ... 
arXiv:1704.00234v2 fatcat:4rja3jewinfjrb2zvkxy6kegmq

Performance Health Index for Complex Cyber Infrastructures [article]

Sanjeev Sondur, Krishna Kant
2021 arXiv   pre-print
In this paper, we propose a Configuration Health Index (CHI) framework specifically attuned to the performance attribute to capture the influence of CVs on the performance aspects of the system.  ...  Most IT systems depend on a set of configuration variables (CVs), expressed as a name/value pair that collectively define the resource allocation for the system.  ...  The discussions with them were highly valuable in devising the solution and added to the techniques presented in the paper.  ... 
arXiv:2109.01254v1 fatcat:wdxpbnhovzc4vkvdg4yvjvdctu
« Previous Showing results 1 — 15 out of 577,132 results