Use this link to cite:
http://hdl.handle.net/2183/27091 Banco de pruebas poliédrico en Python
Loading...
Identifiers
Publication date
Authors
Abella González, Miguel Ángel
Advisors
Other responsabilities
Enxeñaría informática, Grao en
Journal Title
Bibliographic citation
Type of academic work
Academic degree
Abstract
[Resumen]
En este trabajo se estudia una aplicación de los métodos de optimización poliédrica, que
se basa en aplicar métodos matemáticos sobre estructuras de código afines que se caracterizan
por utilizar bucles regulares de gran tamaño en donde el control y los datos dependen únicamente
de las variables de inducción del bucle y constantes mediante funciones afines. Estas
regiones, que se suelen llamar Static Control Parts (SCoPs), se modelan y optimizan usando
técnicas de compilación poliédrica.
El objetivo principal de este trabajo consiste en portar las aplicaciones de PolyBench/C,
que conforman un conjunto de pruebas de rendimiento (benchmarks) empleadas para el desarrollo
y validación de técnicas de optimización poliédrica en el lenguaje de programación C,
al lenguaje de programación Python para de forma similar conformar un banco de pruebas
estándar de cara al futuro desarrollo de optimizaciones poliédricas en este lenguaje.
[Abstract] This paper explores an application of polihedral optimization, which consists on using mathematical methods on affine code structures which are characterized by using large regular loops where control and data depend solely on loop induction variables and constants using affine functions. These regions, often called Static Control Parts (SCoPs), are modeled and optimized using polyhedral compilation. The main objective of this work is porting the applications present in PolyBench/C, which form a set of performance tests (benchmarks) used for the development and validation of polyhedric optimization techniques for the C programming language, to the Python programming language to form in a similar manner a standard test bench for future development of polihedral optimizations on this language.
[Abstract] This paper explores an application of polihedral optimization, which consists on using mathematical methods on affine code structures which are characterized by using large regular loops where control and data depend solely on loop induction variables and constants using affine functions. These regions, often called Static Control Parts (SCoPs), are modeled and optimized using polyhedral compilation. The main objective of this work is porting the applications present in PolyBench/C, which form a set of performance tests (benchmarks) used for the development and validation of polyhedric optimization techniques for the C programming language, to the Python programming language to form in a similar manner a standard test bench for future development of polihedral optimizations on this language.
Description
Editor version
Rights
Atribución-NoComercial-SinDerivadas 3.0 España







