Compilação Dinâmica com Seleção Inteligente de Otimizações

Systems based on dynamic compilation generate target code at run time. Thus, the compilation time is included in the system runtime and it is crucial to build a low-cost system, which can at the same time generate good quality code and have low compilation cost. In this paper, we present a novel approach to use machine learning to perform an intelligent selection of optimizations to apply for each region dynamic compiled. Such a system was tested in a dynamic binary translator, OI-DBT, whose performance was increased by 26.32% when using this smart optimization approach.

[1]  Michael F. P. O'Boyle,et al.  Method-specific dynamic compilation using logistic regression , 2006, OOPSLA '06.

[2]  Michael D. Smith,et al.  Improving region selection in dynamic optimization systems , 2005, 38th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO'05).

[3]  José Nelson Amaral,et al.  Using machines to learn method-specific compilation strategies , 2011, International Symposium on Code Generation and Optimization (CGO 2011).

[4]  Guilherme Ottoni,et al.  HHVM JIT: a profile-guided, region-based compiler for PHP and Hack , 2018, PLDI.

[5]  Toshiaki Yasue,et al.  Effectiveness of cross-platform optimizations for a java just-in-time compiler , 2003, OOPSLA '03.

[6]  E. Duesterwald,et al.  Software profiling for hot path prediction: less is more , 2000, SIGP.

[7]  Radhakishan Yadav,et al.  Performance Analysis for Android Runtime Environment , 2015, 2015 Fifth International Conference on Communication Systems and Network Technologies.

[8]  Christus,et al.  A General Method Applicable to the Search for Similarities in the Amino Acid Sequence of Two Proteins , 2022 .