Publication: Soporte de sistema operativo para ahorro de energía en plataformas móviles con procesadores multicore asimétricos
Loading...
Official URL
Full text at PDC
Publication Date
2016
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
Los procesadores multicore asimétricos con repertorio común de instrucciones (AMPsAsymmetric Multicore Processors) han sido propuestos recientemente como alternativa de bajo consumo a los procesadores multicore simétricos convencionales. Los AMPs combinan, en un mismo chip, cores rápidos de alto rendimiento, con cores más lentos y sencillos de consumo reducido. Uno de los ejemplos más destacados de procesador multicore asimétrico es el procesador big.LITTLE de ARM, que incorporan algunos modelos de teléfonos móviles y tablets disponibles en la actualidad. Trabajos previos han demostrado que para explotar los beneficios potenciales de los procesadores multicore asimétricos, el sistema operativo debe tener en cuenta el beneficio relativo (speedup) que cada aplicación experimenta al ejecutar en un core rápido frente a un core lento. Actualmente, los planificadores por defecto de los sistemas operativos de propósito general no tienen en cuenta la diversidad de speedups entre aplicaciones que puede estar presente en una carga de trabajo multiprogramada. En consecuencia, la asignación de aplicaciones a cores que hacen estos planificadores no extrae el máximo rendimiento por vatio de la plataforma. Recientemente se han realizado extensiones en el kernel Linux para ofrecer un mejor soporte de planificación en multicore asimétricos. Sin embargo, estas extensiones del planificador, utilizadas fundamentalmente en dispositivos móviles con el sistema operativo Android, tampoco tienen en cuenta la diversidad de speedups en las aplicaciones de la carga de trabajo. Por lo tanto estas extensiones no constituyen una aproximación robusta desde el punto de vista de la eficiencia energética. En este proyecto se lleva a cabo la evaluación exhaustiva de distintos algoritmos de planificación para multicore asimétricos sobre una plataforma provista de un procesador ARM big.LITTLE. El principal objetivo del estudio es cuantificar el grado de eficiencia energética y el rendimiento global proporcionado por implementaciones de estos algoritmos en el kernel Linux sobre hardware multicore asimétrico real.
Asymmetric single-ISA (instruction set architecture) multicore processors (AMPs) have been recently proposed as a more power-efficient alternative to conventional symmetric multicore processors. AMPs combine high-performance big cores with power-efficient small cores on a single chip. The big.LITTLE ARM processor, which is integrated in many commercial off-the-shelf mobile devices, constitutes a remarkable example of asymmetric multicore processor. In order to fully tap into the potential of AMPs, the operating system scheduler must factor in, when making scheduling decisions, the relative benefit (aka. speedup) that each application derives when running on a big core relative to a small one. Notably, current default schedulers in general-purpose operating systems do not take into account the diversity in relative speedups that may be present in a multi-program workload. As a result, these schedulers do not always guarantee the maximum performance-per-watt on asymmetric multicore platforms. Recently, scheduling extensions have been created for the Linux kernel to offer a better support on asymmetric multicore systems. However, these scheduling extensions (used primarily on mobile devices running the Android operating system) do not take into consideration the relative speedup of each application when making scheduling decisions. Consequently, these extensions do not constitute a robust approach from the energy efficiency standpoint. In this project, we carry out a comprehensive evaluation of different asymmetryaware scheduling algorithms on a system featuring an ARM big.LITTLE processor. The main goal of the study is to assess the degree of energy efficiency and overall system performance of implementations of these algorithms in the Linux kernel running on real asymmetric hardware.
Asymmetric single-ISA (instruction set architecture) multicore processors (AMPs) have been recently proposed as a more power-efficient alternative to conventional symmetric multicore processors. AMPs combine high-performance big cores with power-efficient small cores on a single chip. The big.LITTLE ARM processor, which is integrated in many commercial off-the-shelf mobile devices, constitutes a remarkable example of asymmetric multicore processor. In order to fully tap into the potential of AMPs, the operating system scheduler must factor in, when making scheduling decisions, the relative benefit (aka. speedup) that each application derives when running on a big core relative to a small one. Notably, current default schedulers in general-purpose operating systems do not take into account the diversity in relative speedups that may be present in a multi-program workload. As a result, these schedulers do not always guarantee the maximum performance-per-watt on asymmetric multicore platforms. Recently, scheduling extensions have been created for the Linux kernel to offer a better support on asymmetric multicore systems. However, these scheduling extensions (used primarily on mobile devices running the Android operating system) do not take into consideration the relative speedup of each application when making scheduling decisions. Consequently, these extensions do not constitute a robust approach from the energy efficiency standpoint. In this project, we carry out a comprehensive evaluation of different asymmetryaware scheduling algorithms on a system featuring an ARM big.LITTLE processor. The main goal of the study is to assess the degree of energy efficiency and overall system performance of implementations of these algorithms in the Linux kernel running on real asymmetric hardware.
Description
Trabajo de Fin de Grado en Ingeniería Informática y en Ingeniería de Computadores (Universidad Complutense, Facultad de Informática, curso 2015/2016)