Fast and Accurate Simulation of Multithreaded Sparse Linear Algebra Solvers

Luka Stanisic 1, 2 Emmanuel Agullo 3, 4 Alfredo Buttari 5, 6 Abdou Guermouche 4 Arnaud Legrand 1, 2 Florent Lopez 7, 6 Brice Videau 8, 1
1 LIG - Laboratoire d'Informatique de Grenoble
2 MESCAL - Middleware efficiently scalable
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
3 HiePACS - High-End Parallel Algorithms for Challenging Numerical Simulations
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
4 LaBRI - Laboratoire Bordelais de Recherche en Informatique
5 CNRS - Centre National de la Recherche Scientifique
6 IRIT - Institut de recherche en informatique de Toulouse
7 UPS - Université Toulouse III - Paul Sabatier
8 CORSE - Compiler Optimization and Run-time Systems
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
Abstract : The ever growing complexity and scale of parallel architectures imposes to rewrite classical monolithic HPC scientific applications and libraries as their portability and performance optimization only comes at a prohibitive cost. There is thus a recent and general trend in using instead a modular approach where numerical algorithms are written at a high level independently of the hardware architecture as Directed Acyclic Graphs (DAG) of tasks. A task-based runtime system then dynamically schedules the resulting DAG on the different computing resources, automatically taking care of data movement and taking into account the possible speed heterogeneity and variability. Evaluating the performance of such complex and dynamic systems is extremely challenging especially for irregular codes. In this article, we explain how we crafted a faithful simulation, both in terms of performance and memory usage, of the behavior of qr_mumps, a fully-featured sparse linear algebra library, on multi-core architectures. In our approach, the target high-end machines are calibrated only once to derive sound performance models. These models can then be used at will to quickly predict and study in a reproducible way the performance of such irregular and resource-demanding applications using solely a commodity laptop.
Keywords : Sparse Linear Algebra Mumps Starpu-simgrid HPC Simgrid Runtime
Document type :
Conference papers
Complete list of metadatas

Cited literature [32 references]  Display  Hide  Download
https://hal.inria.fr/hal-01180272
Contributor : Luka Stanisic <>
Submitted on : Tuesday, January 26, 2016 - 11:25:30 AM
Last modification on : Friday, June 14, 2019 - 6:31:09 PM
Long-term archiving on : Wednesday, April 27, 2016 - 1:19:07 PM

File

QRMSTARSG_article.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01180272, version 2

Collections

LIG | INRIA-BORDEAUX | UNIV-BORDEAUX | LABRI | INRIA | SMS | UNIV-TLSE2 | UGA | PLAFRIM | LIG-SRCPR-CORSE | LIG_SRCPR | UNIV-TLSE3

Citation

Luka Stanisic, Emmanuel Agullo, Alfredo Buttari, Abdou Guermouche, Arnaud Legrand, et al.. Fast and Accurate Simulation of Multithreaded Sparse Linear Algebra Solvers. The 21st IEEE International Conference on Parallel and Distributed Systems, Dec 2015, Melbourne, Australia. ⟨hal-01180272v2⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

1100

Files downloads

895

Contact


archive-ouverte@inria.fr