8665 articles  [english version]

Vous pouvez consulter et rechercher ici l'ensemble des rapports de recherche et rapports techniques de l'Inria déposées sur HAL ou sur HAL-Inria.
Pour déposer sur Hal-Inria: http://hal.inria.fr


Contact : archive-ouverte@inria.fr


Derniers rapports des équipes Inria




Publications HAL de la collection INRIA-RRRT

2014

Titre
FrameSoC Workbench: Facilitating Trace Analysis through a Consistent User Interface
Auteurs
Generoso Pagano; Vania Marangonzova-Martin
Détail
[Technical Report], 2014, pp. 26. RT-0447
Début du résumé
Execution traces are a precious instrument for debugging or performance evaluation of complex applications. They are used in multiple domains, including high performance systems and embedded systems. Given the constantly increasing trace size, powerful tools for trace analysis become essential. FrameSoC, the SoC-Trace project trace management infrastructure, provides generic solutions to facilitate trace analysis. This document describes the graphical environment provided by FrameSoC for execution trace analysis. We detail the functionalities provided to the user, as well as the mechanisms employed for the design of an intuitive, flexible and configurable environment. .....
Accès au texte intégral et bibtex
RT-447.pdf BibTex
Titre
An approximate analysis of heterogeneous and general cache networks
Auteurs
Nicaise Eric Choungmo Fofack url; Don Towsley url; Misha Badov; Mostafa Dehghan; Dennis L. Goeckel
Détail
[Research Report], 2014, pp. 36. RR-8516
Début du résumé
In this paper, we propose approximate models to assess the performance of a cache network with arbitrary topology where nodes run the Least Recently Used (LRU), First-In First-Out (FIFO), or Random (RND) replacement policies on arbitrary size caches. Our model takes advantage of the notions of cache characteristic time and Time-To-Live (TTL)-based cache to develop a unified framework for approximating metrics of interest of interconnected caches. Our approach is validated through event-driven simulations; and when possible, compared to the existing a-NET model 23. .....
Accès au texte intégral et bibtex
RR-8516.pdf BibTex
Titre
Task-based FMM for heterogeneous architectures
Auteurs
Emmanuel Agullo; Bérenger Bramas; Olivier Coulaud url; Eric Darve; Matthias Messner; Toru Takahashi
Détail
[Research Report], 2014, pp. 29. RR-8513
Début du résumé
High performance \FMM is crucial for the numerical simulation of many physical problems. In a previous study~\cite{Agullo2013}, we have shown that task-based \FMM provides the flexibility required to process a wide spectrum of particle distributions efficiently on multicore architectures. In this paper, we now show how such an approach can be extended to fully exploit heterogeneous platforms. For that, we design highly tuned GPU versions of the two dominant operators (P2P and M2L) as well as a scheduling strategy that dynamically decides which proportion of subsequent tasks are processed on regular CPU cores and on GPU accelerators. We assess our .....
Accès au texte intégral et bibtex
RR-8513.pdf BibTex
Titre
SPARQL Template: A Transformation Language for RDF
Auteurs
Olivier Corby url; Catherine Faron-Zucker url; Fabien Gandon url
Détail
[Research Report], 2014, pp. 22. RR-8514
Début du résumé
RDF can be viewed as a meta-model to represent on the Web other languages and models, and in particular their abstract graph structure. The general research question addressed in this document is "How to transform RDF into other languages" and, in particular, how to generate the concrete syntax of expressions of a given language from their RDF representation. We show how SPARQL can be used as a generic RDF transformation rule language, independant from the output language. We define an RDF transformer as a set of transformation rules processed by a generic transformation rule engine. We present a lightweight syntactic .....
Accès au texte intégral et bibtex
RR-8514.pdf BibTex
Titre
Deterministic Partial Replay for MPSoC Debugging
Auteurs
Kirill Georgiev; Vania Marangozova-Martin
Détail
[Research Report], 2014, pp. 30. RR-8515
Début du résumé
This work reports on a debugging methodology for MPSoC based on deterministic record-replay. It defines a general model of MPSoC, identifies the major sources for non determinism and selects a set of adapted algorithms for the record and replay of non deterministic errors. The contribution of this work the definition of a debugging cycle targeting errors by applying temporal and spatial selection criteria. is the proposal of and . The idea behind spatial and temporal selection is to consider not the entire execution of the whole application but replay a part of the application during a specific execution interval. The .....
Accès au texte intégral et bibtex
RR_8515.pdf BibTex