Narrowing the Search Space of Applications Mapping on Hierarchical Topologies

Autor: Nicolas Denoyelle, Swann Perarnau, Brice Videau, Pete Beckman, Emmanuel Jeannot

Publikováno v: PMBS21 Workshop-12th IEEE International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems, to be held in conjunction with SC21
PMBS21 Workshop-12th IEEE International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems, to be held in conjunction with SC21, Nov 2021, Saint-Louis, United States
PMBS21 Workshop, to be held in conjunction with SC21
PMBS21 Workshop, to be held in conjunction with SC21, Nov 2021, Saint-Louis, United States

To be held in conjunction with SC21; International audience; Processor architectures at exascale and beyond are expected to continue to suffer from nonuniform access issues to in-die and node-wide shared resources. Mapping applications onto these res

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5eabb8b078a105638b71cb459a8ae313
https://hal.inria.fr/hal-03364531

Explicit Data Layout Management for Autotuning Exploration on Complex Memory Topologies

Autor: Kamil Iskra, Nicolas Denoyelle, Pete Beckman, Brice Videau, Florence Monna, Swann Perarnau

Publikováno v: MCHPC@SC

The memory topology of high-performance com- puting platforms is becoming more complex. Future exascale platforms in particular are expected to feature multiple types of memory technologies, and multiple accelerator devices per compute node. In this

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::61916659f15ea9511004adfdbe58c9e0
https://doi.org/10.1109/mchpc49590.2019.00015

Modeling Non-Uniform Memory Access on Large Compute Nodes with the Cache-Aware Roofline Model

Autor: Aleksandar Ilic, Brice Goglin, Leonel Sousa, Emmanuel Jeannot, Nicolas Denoyelle

Publikováno v: IEEE Transactions on Parallel and Distributed Systems
IEEE Transactions on Parallel and Distributed Systems, 2019, 30 (6), pp.1374--1389. ⟨10.1109/TPDS.2018.2883056⟩
IEEE Transactions on Parallel and Distributed Systems, Institute of Electrical and Electronics Engineers, 2019, 30 (6), pp.1374--1389. ⟨10.1109/TPDS.2018.2883056⟩

International audience; NUMA platforms, emerging memory architectures with on-package high bandwidth memories bring new opportunities and challenges to bridge the gap between computing power and memory performance. Heterogeneous memory machines featu

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::27ae4cc694c365ac7fe1bddecaf99ef5
https://inria.hal.science/hal-01924951/document

A Topology-Aware Performance Monitoring Tool for Shared Resource Management in Multicore Systems

Autor: Nicolas Denoyelle, Brice Goglin, Emmanuel Jeannot

Publikováno v: Proceedings of Euro-Par 2015: Parallel Processing Workshops
Proceedings of Euro-Par 2015: Parallel Processing Workshops, Aug 2015, Vienna, Austria. ⟨10.1007/978-3-319-27308-2_57⟩
Euro-Par 2015: Parallel Processing Workshops ISBN: 9783319273075
Euro-Par Workshops

International audience; Nowadays, performance optimization involves careful data and task placement to deal with parallel application needs with respect to the underlying hardware topology. Monitoring the application behavior provides useful informat

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::55e16473ccac895837acdc31509ff72a
https://hal.inria.fr/hal-01183083/file/ROME-workshop-camera-ready.pdf