skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: POET: Parameterized Optimization for Empirical Tuning

Abstract

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.

Authors:
; ; ; ;
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
908082
Report Number(s):
UCRL-PROC-227558
TRN: US200722%%408
DOE Contract Number:
W-7405-ENG-48
Resource Type:
Conference
Resource Relation:
Conference: Presented at: IPDPS Workshop on Performance Optimization of High-Level Languages and Libraries, Long Beach, CA, United States, Mar 30 - Mar 30, 2007
Country of Publication:
United States
Language:
English
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; ALGEBRA; EFFICIENCY; FLEXIBILITY; KERNELS; OPTIMIZATION; PERFORMANCE; TRANSFORMATIONS; TUNING; COMPUTER ARCHITECTURE

Citation Formats

Yi, Q, Seymour, K, You, H, Vuduc, R, and Quinlan, D. POET: Parameterized Optimization for Empirical Tuning. United States: N. p., 2007. Web.
Yi, Q, Seymour, K, You, H, Vuduc, R, & Quinlan, D. POET: Parameterized Optimization for Empirical Tuning. United States.
Yi, Q, Seymour, K, You, H, Vuduc, R, and Quinlan, D. Mon . "POET: Parameterized Optimization for Empirical Tuning". United States. doi:. https://www.osti.gov/servlets/purl/908082.
@article{osti_908082,
title = {POET: Parameterized Optimization for Empirical Tuning},
author = {Yi, Q and Seymour, K and You, H and Vuduc, R and Quinlan, D},
abstractNote = {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.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Mon Jan 29 00:00:00 EST 2007},
month = {Mon Jan 29 00:00:00 EST 2007}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: