Performance potential of optimization phase selection during dynamic JIT compilation

  title={Performance potential of optimization phase selection during dynamic JIT compilation},
  author={Michael R. Jantz and P. Kulkarni},
  booktitle={VEE '13},
  • Michael R. Jantz, P. Kulkarni
  • Published in VEE '13 2013
  • Computer Science
  • Phase selection is the process of customizing the applied set of compiler optimization phases for individual functions or programs to improve performance of generated code. Researchers have recently developed novel feature-vector based heuristic techniques to perform phase selection during online JIT compilation. While these heuristics improve program startup speed, steady-state performance was not seen to benefit over the default fixed single sequence baseline. Unfortunately, it is still not… CONTINUE READING
    19 Citations
    A graph-based iterative compiler pass selection and phase ordering approach
    • 22
    Efficient Target and Application Specific Selection and Ordering of Compiler Passes
    Finding Good Compilation Plans: A Strategy to Enhance an Adaptive Optimization System
    A clustering-based approach for exploring sequences of compiler optimizations
    • 5
    The use of different strategies of search space reduction in mitigation of optimization selection problem
    An Optimization-Driven Incremental Inline Substitution Algorithm for Just-in-Time Compilers
    • 5
    • PDF


    Optimizing general purpose compiler optimization
    • 35
    • Highly Influential
    Cole: compiler optimization level exploration
    • 137
    • Highly Influential
    • PDF
    Method-specific dynamic compilation using logistic regression
    • 109
    • Highly Influential
    • PDF
    Milepost GCC: Machine Learning Enabled Self-tuning Compiler
    • 178
    • Highly Influential
    • PDF
    Using machines to learn method-specific compilation strategies
    • 24
    • Highly Influential
    • PDF
    The Java HotSpotTM Server Compiler
    • 98
    • Highly Influential
    • PDF
    • J. Lackie
    • The Dictionary of Cell & Molecular Biology
    • 2013
    • 130,442
    • Highly Influential
    • PDF
    • 148,341
    • Highly Influential
    • PDF
    • 3,997
    • Highly Influential
    The Java hotspottm se rver compiler
    • Proceedings of the Symposium on JavaTM Virtual Machine Research and Technology Symposium , pages 1–12, Berkeley, CA, USA
    • 2001