POET: Parameterized Optimizations for Empirical Tuning
POET: Parameterized Optimizations for Empirical Tuning
复制标题
POET：经验调优的参数化优化
DOI：
10.1109/ipdps.2007.370637
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
D. Quinlan
中科院分区：
文献类型：
--
作者：
Qing Yi;Keith Seymour;Haihang You;R. Vuduc;D. Quinlan
机器体系结构和应用程序的过度复杂性使得编译器难以静态建模和预测应用程序行为。这一观察激发了最近对使用经验技术进行性能调优的兴趣。我们提出了一种新的嵌入式脚本语言，POET（参数化优化经验调整），参数化复杂的代码转换，使他们可以根据经验调整。POET语言旨在显著提高现有经验调优系统的通用性、灵活性和效率。我们已经使用该语言来参数化和经验调整三个循环优化-交换，阻塞和展开-两个线性代数内核。我们的实验表明，调整这些优化使用POET，不需要任何程序分析，所需的时间显着短于使用一个完整的基于编译器的源代码优化器，执行复杂的程序分析和优化。
The excessive complexity of both machine architectures and applications have made it difficult for compilers to statically model and predict application behavior. This observation motivates the recent interest in performance tuning using empirical techniques. We present a new embedded scripting language, POET (parameterized optimization for empirical tuning), for parameterizing complex code transformations so that they can be empirically tuned. The POET language aims to significantly improve the generality, flexibility, and efficiency of existing empirical tuning systems. We have used the language to parameterize and to empirically tune three loop optimizations - interchange, blocking, and unrolling - for two linear algebra kernels. We show experimentally that the time required to tune these optimizations using POET, which does not require any program analysis, is significantly shorter than that when using a full compiler-based source-code optimizer which performs sophisticated program analysis and optimizations.