文章基本信息

标题：Integrating Profiling Into MDE Compilers
本地全文：下载
作者：Vincent Aranega ; A. Wendell O. Rodrigues ; Anne Etien 等
期刊名称：International Journal of Software Engineering & Applications (IJSEA)
印刷版ISSN：0976-2221
电子版ISSN：0975-9018
出版年度：2014
卷号：5
期号：4
页码：1
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Scientific computation requires more and more performance in its algorithms. New massively parallelarchitectures suit well to these algorithms. They are known for offering high performance and powerefficiency. Unfortunately, as parallel programming for these architectures requires a complex distributionof tasks and data, developers find difficult to implement their applications effectively. Although approachesbased on source-to-source intends to provide a low learning curve for parallel programming and takeadvantage of architecture features to create optimized applications, programming remains difficult forneophytes. This work aims at improving performance by returning to the high-level models, specificexecution data from a profiling tool enhanced by smart advices computed by an analysis engine. In order tokeep the link between execution and model, the process is based on a traceability mechanism. Once themodel is automatically annotated, it can be re-factored aiming better performances on the re-generatedcode. Hence, this work allows keeping coherence between model and code without forgetting to harness thepower of parallel architectures. To illustrate and clarify key points of this approach, we provide anexperimental example in GPUs context. The example uses a transformation chain from UML-MARTEmodels to OpenCL code.
关键词：Profiling – MDE – Traceability – Performance Analysis