Analyse pire cas pour processeur multi-cœurs disposant de caches partagés
نویسنده
چکیده
Hard real-time systems are subject to timing constraints and failure to respectthem can cause economic, ecological or human disasters. The validation processwhich guarantees the safety of such software, by ensuring the respect of theseconstraints in all situations including the worst case, is based on the knowledgeof the worst case execution time of each task. However, determining the worst caseexecution time is a difficult problem for modern architectures because of complexhardware mechanisms that could cause significant execution time variability.This document focuses on the analysis of the worst case timing behavior ofcache hierarchies, to determine their contribution to the worst case executiontime. Several approaches are proposed to predict and improve the worst caseexecution time of tasks running on multicore processors with a cache hierarchyin which some cache levels are shared between cores.
منابع مشابه
WCET analysis of multi-level set-associative instruction caches
With the advent of increasingly complex hardware in real-time embedded systems (processors with performance enhancing features such as pipelines, cache hierarchy, multiple cores), many processors now have a set-associative L2 cache. Thus, there is a need for considering cache hierarchies when validating the temporal behavior of real-time systems, in particular when estimating tasks’ worst-case ...
متن کاملPerformance analysis and models for collocated VMs running on multi-core physical machines
Next generation high performance computers will massively use virtualization as a way to share hardware resources between multiple applications and to provide flexible mechanisms for fault tolerance and energy optimisation. In this context, understanding the performance behavior of virtual machines and the interference between them is a major scientific challenge that will allow a more efficien...
متن کاملReplacement policies for shared caches on symmetric multicores : a programmer - centric point of view Pierre Michaud
The presence of shared caches in current multicore processors may generate a lot of performance variability when several applications execute simultaneously. For the programmer of an application with quality-of-service goals, this performance variability may lead to a very pessimistic tuning. To solve this problem, there must be a way for the programmer to define a reasonable performance target...
متن کاملThe impact of cache misses on the performance of matrix product algorithms on multicore platforms
The multicore revolution is underway, bringing new chips introducing more complex memory architectures. Classical algorithms must be revisited in order to take the hierarchical memory layout into account. In this paper, we aim at designing cache-aware algorithms that minimize the number of cache misses paid during the execution of the matrix product kernel on a multicore processor. We analytica...
متن کاملCompilation pour cibles hétérogènes: automatisation des analyses, transformations et décisions nécessaires
Résumé Les accélérateurs matériels, telles les cartes FPGA ou les cartes graphiques, apportent une alternative ou un complément intéressant aux processeurs multi-cœurs classiques pour de nombreuses applications scientifiques. Il est cependant coûteux et difficile d’y porter des applications existantes ; et les compilateurs standards, traditionnellement portés sur la génération de code pour proc...
متن کامل