Search Results - "Jeannot, Emmanuel"

An introspection monitoring library to improve MPI communication time by Jeannot, Emmanuel, Sartori, Richard

“…In this paper, we describe how to improve communication time of MPI parallel applications with the use of a library that enables to monitor MPI applications…”
Get full text

IO-aware Job-Scheduling: Exploiting the Impacts of Workload Characterizations to select the Mapping Strategy by Jeannot, Emmanuel, Pallez, Guillaume, Vidal, Nicolas

“…In high performance, computing concurrent applications are sharing the same file system. However, the bandwidth which provides access to the storage is…”
Get full text

Scheduling periodic I/O access with bi-colored chains: models and algorithms by Jeannot, Emmanuel, Pallez, Guillaume, Vidal, Nicolas

“…Observations show that some HPC applications periodically alternate between (i) operations (computations, local data accesses) executed on the compute nodes,…”
Get full text

Evaluation and Optimization of the Robustness of DAG Schedules in Heterogeneous Environments by Canon, L.-C., Jeannot, E.

“…A schedule is said to be robust if it is able to absorb some degree of uncertainty in task or communication durations while maintaining a stable solution. This…”
Get full text

Study on progress threads placement and dedicated cores for overlapping MPI nonblocking collectives on manycore processor by Denis, Alexandre, Jaeger, Julien, Jeannot, Emmanuel, Pérache, Marc, Taboada, Hugo

“…To amortize the cost of MPI collective operations, nonblocking collectives have been proposed so as to allow communications to be overlapped with computation…”
Get full text

TAPIOCA: An I/O Library for Optimized Topology-Aware Data Aggregation on Large-Scale Supercomputers by Tessier, Francois, Vishwanath, Venkatram, Jeannot, Emmanuel

“…Reading and writing data efficiently from storage system is necessary for most scientific simulations to achieve good performance at scale. Many software…”
Get full text

Topology-aware job mapping by Georgiou, Yiannis, Jeannot, Emmanuel, Mercier, Guillaume, Villiermet, Adèle

“…A Resource and Job Management System (RJMS) is a crucial system software part of the HPC stack. It is responsible for efficiently delivering computing power to…”
Get full text

Grid'5000: A Large Scale And Highly Reconfigurable Experimental Grid Testbed by Bolze, Raphaël, Cappello, Franck, Caron, Eddy, Daydé, Michel, Desprez, Frédéric, Jeannot, Emmanuel, Jégou, Yvon, Lanteri, Stephane, Leduc, Julien, Melab, Noredine, Mornet, Guillaume, Namyst, Raymond, Primet, Pascale, Quetier, Benjamin, Richard, Olivier, Talbi, El-Ghazali, Touche, Iréa

“…Large scale distributed systems such as Grids are difficult to study from theoretical models and simulators only. Most Grids deployed at large scale are…”
Get full text

Process mapping on any topology with TopoMatch by Jeannot, Emmanuel

“…•We present TopoMatch.•TopoMatch is a tool and algorithm to perform process and thread mapping.•We Show that TopoMatch can use any type of topologies.•We show…”
Get full text

Process Placement in Multicore Clusters:Algorithmic Issues and Practical Techniques by Jeannot, Emmanuel, Mercier, Guillaume, Tessier, Francois

“…Current generations of NUMA node clusters feature multicore or manycore processors. Programming such architectures efficiently is a challenge because numerous…”
Get full text

Adding topology and memory awareness in data aggregation algorithms by Tessier, François, Vishwanath, Venkatram, Jeannot, Emmanuel

“…With the growing gap between computing power and the ability of large-scale systems to ingest data, I/O is becoming the bottleneck for many scientific…”
Get full text

Symbolic mapping and allocation for the Cholesky factorization on NUMA machines: Results and optimizations by Jeannot, Emmanuel

“…We discuss some performance issues of the tiled Cholesky factorization on non-uniform memory access-time (NUMA) shared memory machines. We show how to optimize…”
Get full text

Tracing task‐based runtime systems: Feedbacks from the StarPU case by Denis, Alexandre, Jeannot, Emmanuel, Swartvagher, Philippe, Thibault, Samuel

“…Summary Given the complexity of current supercomputers and applications, being able to trace application executions to understand their behavior is not a…”
Get full text

Modeling Non-Uniform Memory Access on Large Compute Nodes with the Cache-Aware Roofline Model by Denoyelle, Nicolas, Goglin, Brice, Ilic, Aleksandar, Jeannot, Emmanuel, Sousa, Leonel

“…NUMA platforms, emerging memory architectures with on-package high bandwidth memories bring new opportunities and challenges to bridge the gap between…”
Get full text

Experimenting task-based runtimes on a legacy Computational Fluid Dynamics code with unstructured meshes by Jeannot, Emmanuel, Fournier, Yvan, Lorendeau, Benjamin

“…•Discussion on porting a Legacy CFD code onto task-based runtime system.•Porting gradient reconstruction onto StarPU and PARSEC.•Comparing the PARSEC and…”
Get full text

A methodology for assessing computation/communication overlap of MPI nonblocking collectives by Denis, Alexandre, Jaeger, Julien, Jeannot, Emmanuel, Reynier, Florian

“…Summary By allowing computation/communication overlap, MPI nonblocking collectives (NBC) are supposed to improve application scalability and performance…”
Get full text

H2M: Exploiting Heterogeneous Shared Memory Architectures by Klinkenberg, Jannis, Kozhokanova, Anara, Terboven, Christian, Foyer, Clément, Goglin, Brice, Jeannot, Emmanuel

“…Over the past decades, the performance gap between the memory subsystem and compute capabilities continued to spread. However, scientific applications and…”
Get full text

Foreword to the Special Issue of the Twenty Sixth International Heterogeneity in Computing Workshop (HCW) and to the Fifteenth International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Platforms (HeteroPar) by Barbosa, Jorge G., Jeannot, Emmanuel

“…Heterogeneity is emerging as one of the most profound and challenging characteristics of today's parallel environments. As most modern computing systems are…”
Get full text

READYS: A Reinforcement Learning Based Strategy for Heterogeneous Dynamic Scheduling by Grinsztajn, Nathan, Beaumont, Olivier, Jeannot, Emmanuel, Preux, Philippe

“…In this paper, we propose READYS, a reinforcement learning algorithm for the dynamic scheduling of computations modeled as a Directed Acyclic Graph (DAGs). Our…”
Get full text

Topology-Aware Data Aggregation for Intensive I/O on Large-Scale Supercomputers by Tessier, Francois, Malakar, Preeti, Vishwanath, Venkatram, Jeannot, Emmanuel, Isaila, Florin

“…Reading and writing data efficiently from storage systems is critical for high performance data-centric applications. These I/O systems are being increasingly…”
Get full text