A model-driven framework for guided design space exploration

Ábel Hegedüs, Ákos Horváth, D. Varró

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Design space exploration (DSE) aims at searching through various models representing different design candidates to support activities like configuration design of critical systems or automated maintenance of IT systems. In model-driven engineering, DSE is applied to find instance models that are (i) reachable from an initial model with a sequence of transformation rules and (ii) satisfy a set of structural and numerical constraints. Since exhaustive exploration of the design space is infeasible for large models, the traversal is often guided by hints, derived by system analysis, to prioritize the next states to traverse (selection criteria) and to avoid searching unpromising states (cut-off criteria). In this paper, we define an exploration approach where selection and cut-off criteria are defined using dependency analysis and algebraic abstraction of transformation rules. Additionally, we apply different state encoding techniques to identify recurring states and reduce the number of visited states. Finally, we illustrate our approach on a cloud infrastructure configuration problem and provide detailed evaluation on both synthetic and real applications. This evaluation includes (i) the comparison of several exploration techniques, (ii) performance measurements on multiple state encoding techniques and (iii) comparing two implementation architectures of our design space exploration framework.

Original languageEnglish
Pages (from-to)399-436
Number of pages38
JournalAutomated Software Engineering
Volume22
Issue number3
DOIs
Publication statusPublished - Aug 20 2014

Fingerprint

Systems analysis

Keywords

  • Design space exploration
  • Model-driven engineering
  • Search-based software engineering

ASJC Scopus subject areas

  • Software

Cite this

A model-driven framework for guided design space exploration. / Hegedüs, Ábel; Horváth, Ákos; Varró, D.

In: Automated Software Engineering, Vol. 22, No. 3, 20.08.2014, p. 399-436.

Research output: Contribution to journalArticle

Hegedüs, Ábel ; Horváth, Ákos ; Varró, D. / A model-driven framework for guided design space exploration. In: Automated Software Engineering. 2014 ; Vol. 22, No. 3. pp. 399-436.
@article{84099e1a91844958aadf486c13ac406b,
title = "A model-driven framework for guided design space exploration",
abstract = "Design space exploration (DSE) aims at searching through various models representing different design candidates to support activities like configuration design of critical systems or automated maintenance of IT systems. In model-driven engineering, DSE is applied to find instance models that are (i) reachable from an initial model with a sequence of transformation rules and (ii) satisfy a set of structural and numerical constraints. Since exhaustive exploration of the design space is infeasible for large models, the traversal is often guided by hints, derived by system analysis, to prioritize the next states to traverse (selection criteria) and to avoid searching unpromising states (cut-off criteria). In this paper, we define an exploration approach where selection and cut-off criteria are defined using dependency analysis and algebraic abstraction of transformation rules. Additionally, we apply different state encoding techniques to identify recurring states and reduce the number of visited states. Finally, we illustrate our approach on a cloud infrastructure configuration problem and provide detailed evaluation on both synthetic and real applications. This evaluation includes (i) the comparison of several exploration techniques, (ii) performance measurements on multiple state encoding techniques and (iii) comparing two implementation architectures of our design space exploration framework.",
keywords = "Design space exploration, Model-driven engineering, Search-based software engineering",
author = "{\'A}bel Heged{\"u}s and {\'A}kos Horv{\'a}th and D. Varr{\'o}",
year = "2014",
month = "8",
day = "20",
doi = "10.1007/s10515-014-0163-1",
language = "English",
volume = "22",
pages = "399--436",
journal = "Automated Software Engineering",
issn = "0928-8910",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - A model-driven framework for guided design space exploration

AU - Hegedüs, Ábel

AU - Horváth, Ákos

AU - Varró, D.

PY - 2014/8/20

Y1 - 2014/8/20

N2 - Design space exploration (DSE) aims at searching through various models representing different design candidates to support activities like configuration design of critical systems or automated maintenance of IT systems. In model-driven engineering, DSE is applied to find instance models that are (i) reachable from an initial model with a sequence of transformation rules and (ii) satisfy a set of structural and numerical constraints. Since exhaustive exploration of the design space is infeasible for large models, the traversal is often guided by hints, derived by system analysis, to prioritize the next states to traverse (selection criteria) and to avoid searching unpromising states (cut-off criteria). In this paper, we define an exploration approach where selection and cut-off criteria are defined using dependency analysis and algebraic abstraction of transformation rules. Additionally, we apply different state encoding techniques to identify recurring states and reduce the number of visited states. Finally, we illustrate our approach on a cloud infrastructure configuration problem and provide detailed evaluation on both synthetic and real applications. This evaluation includes (i) the comparison of several exploration techniques, (ii) performance measurements on multiple state encoding techniques and (iii) comparing two implementation architectures of our design space exploration framework.

AB - Design space exploration (DSE) aims at searching through various models representing different design candidates to support activities like configuration design of critical systems or automated maintenance of IT systems. In model-driven engineering, DSE is applied to find instance models that are (i) reachable from an initial model with a sequence of transformation rules and (ii) satisfy a set of structural and numerical constraints. Since exhaustive exploration of the design space is infeasible for large models, the traversal is often guided by hints, derived by system analysis, to prioritize the next states to traverse (selection criteria) and to avoid searching unpromising states (cut-off criteria). In this paper, we define an exploration approach where selection and cut-off criteria are defined using dependency analysis and algebraic abstraction of transformation rules. Additionally, we apply different state encoding techniques to identify recurring states and reduce the number of visited states. Finally, we illustrate our approach on a cloud infrastructure configuration problem and provide detailed evaluation on both synthetic and real applications. This evaluation includes (i) the comparison of several exploration techniques, (ii) performance measurements on multiple state encoding techniques and (iii) comparing two implementation architectures of our design space exploration framework.

KW - Design space exploration

KW - Model-driven engineering

KW - Search-based software engineering

UR - http://www.scopus.com/inward/record.url?scp=84937641875&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937641875&partnerID=8YFLogxK

U2 - 10.1007/s10515-014-0163-1

DO - 10.1007/s10515-014-0163-1

M3 - Article

AN - SCOPUS:84937641875

VL - 22

SP - 399

EP - 436

JO - Automated Software Engineering

JF - Automated Software Engineering

SN - 0928-8910

IS - 3

ER -