2012•Unpublished venueRequires access

Optimization of aspect-instantiation strategies in the JIT-compiler

Martin Zandberg

Open publisher page 1 citations

Abstract

Aspect-instantiation policies define the rules how aspect instances are, looked-up and instantiated. Current aspect-oriented languages implementations are restricted to generating bytecode for realizing the semantics of aspect instantiation. Often however, bytecode is not expressive enough to support the most efficient optimizations. The ALIA4J framework for implementing execution environments with support for aspect-oriented languages allows implementing the semantics of language features in terms of bytecode generation and JIT-compilation. In this thesis a JIT compilation strategy for aspect-instantiation is developed that supports a variety of aspect-instantiation policies, ranging from singleton and per-object policies to per-multiple-object policies. The compilation strategy is benchmarked and its performance is compared to state-of-the-art aspect-oriented language implementations.

About this research paper

What this paper is about

Aspect-instantiation policies define the rules how aspect instances are, looked-up and instantiated. Current aspect-oriented languages implementations are restricted to generating bytecode for realizing the semantics of aspect instantiation. Often however, bytecode is not expressive enough to support the most efficient optimizations. The ALIA4J framework for implementing execution environments with support for aspect-oriented languages allows implementing the semantics of language features in terms of bytecode generation and JIT-compilation. In this thesis a JIT compilation strategy for aspect-instantiation is developed that supports a variety of aspect-instantiation policies, ranging from singleton and per-object policies to per-multiple-object policies. The compilation strategy is benchmarked and its performance is compared to state-of-the-art aspect-oriented language implementations.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Aspect-instantiation policies define the rules how aspect instances are, looked-up and instantiated. Current aspect-oriented languages implementations are restricted to generating bytecode for realizing the semantics of aspect instantiation. Often however, bytecode is not expressive enough to support the most efficient optimizations. The ALIA4J framework for implementing execution environments with support for aspect-oriented languages allows implementing the semantics of language features in terms of bytecode generation and JIT-compilation. In this thesis a JIT compilation strategy for aspect-instantiation is developed that supports a variety of aspect-instantiation policies, ranging from singleton and per-object policies to per-multiple-object policies. The compilation strategy is benchmarked and its performance is compared to state-of-the-art aspect-oriented language implementations.

Key concepts: Bytecode, Programming language, Computer science, Aspect-oriented programming, Compiler, Implementation, Semantics (computer science), Variety (cybernetics)

Related papers

Back to paper searchBrowse research topicsOriginal source
Optimization of aspect-instantiation strategies in the JIT-compiler — Research Paper | ScholarLens