Objective: End-to-end parallel programming solutions for high-performance interactive computing with provable performances.

1
The MOAIS team-project

• Objective End-to-end parallel programming
  solutionsfor high-performance interactive
  computing with provable performances.
• optimization computational
  steering, VR embedded
• Performance is multi-objective
• Adaptive to the platform

Option pricing on gridLIG Moais / LJK MathFi
INRIA Grimage platform LIG/LJK MOAIS,
PERCEPTION, EVASION
Streaming on MPSoCs ST
2
Evolution of parallel programming and
Moaissoftwares
AWS / Kaapi-C

• Multi-objective Adaptive Performance
• Applications are time-consuming but essential to
  validate scientific approach

3
Highlights Kaapi/Taktuk

• 1st prize Plugtest Nov. 2007 and Nov 2008
• KAAPI winner of V GRID PLUGTEST 2008 Super Quant
  Monte Carlo Challenge

• Nov. 2007 1st prize
• Nqueens(23) in 2107s with 3654 cores
• Nov 2008 1st Prize
• Portolio Pricing (LIG/Moais LJK/MathFi)

• Parallelization of a code for pricing high
  dimensional European options (Monte Carlo method)
• 1st rank 988 options priced in the 1 hour
  allocated computation time on a grid with 3609
  cores (machines in Japan and France)
• Kaapi/Taktuk programming/deployment tool for
  GCM standardfault tolerance

4
Highlights FlowVR / Grimagehttp//grimage.inrialp
es.fr/

• GRIMAGE platform SIGGRAPH Aug. 2007 Emerging
  Tech. Demo
• fête de la science / Grand Palais Paris Nov.
  2008
• Valorization start-up

4000 visitors
5
Highlight HPC Kaapi SOFA http//www.sofa-frame
work.org/
6
Contacts

• Moais Jean-Louis.Roch_at_imag.fr
• Kaapi Thierry.Gautier_at_imag.fr
• FlowVR / Grimage Bruno.Raffin_at_imag.fr