Options
Multiobjective compiler-based optimizations for hard real-time systems
Citation Link: https://doi.org/10.15480/882.4799
Publikationstyp
Thesis
Thesis Type
Doctoral Thesis
Publikationsdatum
2022-12
Sprache
English
Author
Advisor
Referee
Title Granting Institution
Technische Universität Hamburg
Place of Title Granting Institution
Hamburg
Examination Date
2022-11-02
Institut
Citation
Technische Universität Hamburg (2022)
The thesis presents multiobjective compiler-based optimizations to find trade-offs between the worst-case execution time, energy consumption, and code size of hard real-time systems. It introduces a novel compiler-based compression technique guaranteeing that a compiled program satisfies its timing constraints. The thesis tackles timing issues caused by solving multiobjective problems by evolutionary algorithms at compile time. The thesis introduces a search space reduction technique and a prediction model based on machine learning to speed up the solution process.
Schlagworte
Multiobjective optimization
Compiler
Hard real-time system
Compression
Prediction models
Search space reduction
DDC Class
004: Informatik
600: Technik
620: Ingenieurwissenschaften