Sites Inria

Version française

International conference

Inria participates to the annual international high performance computing conference

Logo SC13

Inria participates to Supercomputing 2013, the annual international meeting of all the scientific and industrial actors in High Performance Computing, Networking, Storage and Analysis. 

  • Date : 18/11/2013 to 21/11/2013
  • Place : Denver,
  • Guest(s) : Marie Granier (SysFera), Lavanya Ramakrishnan (LBNL) , Serge Petiton and Mitsuhisa Sato (Japanese-­‐French FP3C project), Guido Juckeland (TU Dresden / ZIH), Jack Dongarra (University of Tennessee), Christine Morin (Inria), Yves Robert (ENS Lyon), Jeff Squyres (Cisco), Christophe Calvin (CEA/DEN/DANS/DM2S)

Several Inria teams will present the last results of their works, among the scientific works selected by the organisation but also on Inria's booth (#2116).
On Inria's booth, visitors will discover a demo of Rubens, a purely mechanical Turing machine using LEGOTM bricks. They will also meet some Inria's teams which will be honoring their work on software tools and platforms to simplify the execution of parallel application, but also on algorithmic aspects:

TEAM or PROJECT

WORK

Alpines

Algorithms and parallel tools for integrated numerical simulations

Avalon

Green HPC Clouds – OpenStack based Energy Efficient HPC as a Service
iExec: A Scalable, Secure, Low-Cost and On-Demand Service Providing High Performance Computing to SMEs
SEED4C – Security Embedded Element and Data privacy for Cloud

Grid’5000

A Production-grade Testbed for Experiment-driven Computer Science on HPC and Clouds

Inria Project Lab C2S@Exa (équipes HiePACS, Runtime, SAGE)

Parallel Numerical Solvers on top of Runtime Systems

KerData

Efficient In Situ Visualization of Large Scale HPC Simulations with Damaris

Mescal

OAR: a versatile resource and task manager for HPC clusters and other computing infrastructures

Myriads

Decentralizing Scientific Workflows with HOCL
VEP: Virtual Execution Platform

Runtime

StarPU: A Runtime System for Hybrid Computing

SimGrid for HPC

 

Still on Inria's booth, some experts who, for mainly of us, work with Inria's teams, will propose various talks on many themes:  from algorithms to software and applications, network accelerators and cloud (see beside the program to download)

Among the selected scientific papers Inria's team have contributed to do, mention may be made of: 

  • in "Fault Tolerance and Migration in the Cloud" session: 

Optimization of Cloud Task Processing with Checkpoint-Restart Mechanism , which involves Roma team

  • in "Preconditioners and Unstructured Meshes" session : 

Scalable Domain Decomposition Preconditioners For Heterogeneous Elliptic Problems which involves Alpines team, registered as finalist in the "best paper" category
Parallel Design and Performance of Nested Filtering Factorization Preconditioner , which involves Alpines team, registered as finalist the "best student paper" category

Keywords: Simulation Cloud computing Network Runtime Middleware High performance computing Grid'5000

Top