1 / 58

Welcome to CREST

Centre for Research in Evolution, Search & Testing. Welcome to CREST. King’s College London Mark Harman. CREST. Engineering: Better, faster and cheaper through automation and Improved insight World leading in Program Slicing Search Based Software Engineering Software Testing

walterbrennan
Download Presentation

Welcome to CREST

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Centre for Research in Evolution, Search & Testing Welcome to CREST King’s College London Mark Harman

  2. CREST Engineering: Better, faster and cheaper • through automation and • Improved insight World leading in • Program Slicing • Search Based Software Engineering • Software Testing Also widely known for work in • Service-Oriented Software Engineering • Foundations of Source Code Analysis

  3. CREST King’s College London CREST: Where we are

  4. CREST King’s College London CREST: Where we are 20 minutes walk

  5. CREST Funding

  6. Industrial Links

  7. Industrial Links

  8. What is SBSE? In SBSE we apply search techniques to search large search spaces, guided by a fitness function that captures properties of the acceptable software artefacts we seek. Genetic Algorithms, Hill climbing, Simulated Annealing, Random, Tabu Search, Estimation of Distribution Algorithms, Particle Swarm Optimization

  9. Why is SBSE?

  10. Why is SBSE?

  11. Why not SBSE? ?

  12. Why not SBSE? EPSRC network 1999 – 2002 Laid foundation for SBSE

  13. Current state of SBSE research 18 countries 20 in UK alone CREST is undisputed world leader

  14. Current state of SBSE research 18 countries 20 in UK alone CREST is undisputed world leader Mark Harman The Current State and Future of Search Based Software Engineering 29th International Conference on Software Engineering (ICSE 2007), Future of Software Engineering (FoSE). Minneapolis, USA, May 2007

  15. SEBASE Project Overview Overall: £2.7M KCL: £1.2M Aim: To provide an entirely new way of understanding and practicing software engineering Software Engineering as search and optimize

  16. … plus EvoTest Overall: £2.1M KCL: £250k Aim: To provide evolutionary search for software test automation Software testing as search and optimize

  17. The Eight Queens Problem

  18. The Eight Queens Problem

  19. The Eight Queens Problem Perfect

  20. The Eight Queens Problem Score 0

  21. The Eight Queens Problem

  22. The Eight Queens Problem

  23. The Eight Queens Problem Two Attacks

  24. The Eight Queens Problem Score -2

  25. The Eight Queens Problem

  26. The Eight Queens Problem

  27. The Eight Queens Problem Three Attacks

  28. The Eight Queens Problem Score -3

  29. That was easy

  30. Generate a solution Place 8 Queens With no attacks

  31. Scale up: Generate a solution

  32. Scale up: Generate a solution

  33. Evolution beats random

  34. SBSE is so generic Testing Fitness function: coverage, time, … Representation: input vector Restructuring Fitness function: cohesion and coupling Representation: mapping nodes to clusters

  35. SBSE is so generic Testing Fitness function: coverage, time, … Representation: input vector Restructuring Fitness function: cohesion and coupling Representation: mapping nodes to clusters

  36. SBSE is so generic Testing is a search problem Fitness function: coverage, time, … Representation: input vector Requirements is a search problem Fitness function: cost, value, dependence constraints Representation: bitset

  37. SBSE is so generic Testing is a search problem Fitness function: coverage, time, … Representation: input vector Regression is a search problem Fitness function: testing goals Representation: retained test set

  38. SBSE is so generic Testing is a search problem Fitness function: coverage, time, … Representation: input vector Planning is a search problem Fitness function: time, schedule robustness Representation: work package allocation

  39. SBSE is so generic Testing is a search problem Fitness function: coverage, time, … Representation: input vector Planning is a search problem Fitness function: time, schedule robustness Representation: work package allocation

  40. SBSE Applications Transformation Cooper, Ryan, Schielke, Subramanian, Fatiregun, Williams Requirements Bagnall, Mansouri, Zhang Effort prediction Aguilar-Ruiz, Burgess, Dolado, Lefley, Shepperd Management Alba, Antoniol, Chicano, Di Pentam Greer, Ruhe Heap allocation Cohen, Kooi, Srisa-an Regression test Li, Yoo, Elbaum, Rothermel, Walcott, Soffa, Kampfhamer SOA Canfora, Di Penta, Esposito, Villani Refactoring Antoniol, Briand, Cinneide, O’Keeffe, Merlo, Seng, Tratt Test Generation Alba, Binkley, Bottaci, Briand, Chicano, Clark, Cohen, Gutjahr, Harrold, Holcombe, Jones, Korel, Pargass, Reformat, Roper, McMinn, Michael, Sthamer, Tracy, Tonella,Xanthakis, Xiao, Wegener, Wilkins Maintenance Antoniol, Lutz, Di Penta, Madhavi, Mancoridis, Mitchell, Swift Model checking Alba, Chicano, Godefroid Probe dist’ion Cohen, Elbaum UIOs Derderian, Guo, Hierons Comprehension Gold, Li, Mahdavi Protocols Alba, Clark, Jacob, Troya Component sel Baker, Skaliotis, Steinhofel, Yoo Agent Oriented Haas, Peysakhov, Sinclair, Shami, Mancoridis

  41. SBSE Applications in which SEBASE is active Transformation Cooper, Ryan, Schielke, Subramanian, Fatiregun, Williams Requirements Bagnall, Mansouri, Zhang Effort prediction Aguilar-Ruiz, Burgess, Dolado, Lefley, Shepperd Management Alba, Antoniol, Chicano, Di Pentam Greer, Ruhe Heap allocation Cohen, Kooi, Srisa-an Regression test Li, Yoo, Elbaum, Rothermel, Walcott, Soffa, Kampfhamer SOA Canfora, Di Penta, Esposito, Villani Refactoring Antoniol, Briand, Cinneide, O’Keeffe, Merlo, Seng, Tratt Test Generation Alba, Binkley, Bottaci, Briand, Chicano, Clark, Cohen, Gutjahr, Harrold, Holcombe, Jones, Korel, Pargass, Reformat, Roper, McMinn, Michael, Sthamer, Tracy, Tonella,Xanthakis, Xiao, Wegener, Wilkins Maintenance Antoniol, Lutz, Di Penta, Madhavi, Mancoridis, Mitchell, Swift Model checking Alba, Chicano, Godefroid Probe dist’ion Cohen, Elbaum UIOs Derderian, Guo, Hierons Comprehension Gold, Li, Mahdavi Protocols Alba, Clark, Jacob, Troya Component sel Baker, Skaliotis, Steinhofel, Yoo Agent Oriented Haas, Peysakhov, Sinclair, Shami, Mancoridis

  42. Regression Companies have their act together We have large pools of regression test data They are getting too large We need to Select and Prioritize

  43. Regression Selection

  44. Regression Selection

  45. C D A B D C A B Regression Prioritization The optimal test case orderings:

  46. Regression Prioritization

  47. Requirements Optimization Most expensive mistakes and misunderstandings Many customer Many requirements Many releases Fairness, Approximation

  48. Requirements optimization

  49. Dependence Analysis Some case studies in dependence analysis Work we are doing with Prof. Binkley for DaimlerChrysler IBM Motorola Case studies are based on open source Prof. Binkley’s visit is supported by EPSRC

  50. Bubble charts replace prepro

More Related