SlideShare uma empresa Scribd logo
1 de 11
Baixar para ler offline
COLDTM: High Performance with Scilab
Boosting your workflow – Boosting your Performances
Béatrice	
  Creusillet	
  &	
  Thomas	
  Izard	
  
ScilabTec	
  2015,	
  Paris	
  
2 - ScilabTec 2015
Development	
  workflow	
  
3 - ScilabTec 2015
Development	
  workflow	
  
4 - ScilabTec 2015
Development	
  workflow	
  
5 - ScilabTec 2015
From	
  code	
  to	
  binaries	
  or	
  libraries	
  
6 - ScilabTec 2015
▶  CFD	
  applicaAon	
  based	
  on	
  LaDce	
  Boltzmann	
  method1	
  
▶  IniAalizaAon	
  step:	
  cfd_init.sce	
  
▶  Time	
  loop:	
  
–  computaAon	
  kernel:	
  cfd_iteraAon.sce	
  
–  visualizaAon	
  
–  10000	
  cycles	
  
	
  	
  	
  	
  	
  (400000	
  originally)	
  
	
  
	
  
	
  
	
  
	
  1	
  From	
  wiki.palabos.org/numerics:codes	
  
Copyright	
  ©	
  2006-­‐2008	
  by	
  Jonas	
  LaW	
  
A	
  prac7cal	
  case	
  
7 - ScilabTec 2015
Performances	
  
	
  
Scilab	
  
Scilab	
  
+	
  
COLD	
  
(kernel)	
  
Scilab	
  
+	
  
COLD	
  
(main	
  script)	
  
	
  
COLD	
  
	
  
(executable)	
  
average	
  
kernel	
  
duraAon	
  
390	
  ms	
   24	
  ms	
   12	
  ms	
   12	
  ms	
  
total	
  Ame	
  
for	
  10000	
  
cycles	
  
1	
  h	
   3	
  min	
  39	
  s	
   1	
  min	
  56	
  s	
   1	
  min	
  52	
  s	
  
speed-­‐up	
   x	
  16	
   x	
  31	
   x	
  32	
  
Experimental	
  condiAons:	
  
•  Intel	
  Core	
  i7-­‐3770	
  @	
  3.40GHz,	
  16Go,	
  Ubuntu	
  14.04	
  
•  Scilab	
  5.5.2,	
  COLD	
  2.1,	
  gcc	
  4.8.2	
  
8 - ScilabTec 2015
Some	
  more	
  performances…	
  
▶  Series	
  of	
  benchmarks	
  from	
  an	
  industrial	
  partner	
  (Richelieu	
  Project)	
  
▶  Raw	
  code,	
  no	
  calls	
  to	
  Scilab	
  library	
  funcAons	
  or	
  vector	
  operaAons	
  
Scilab	
  5.4.1	
   COLDTM	
  1.6.3	
   Speed-­‐up	
  
Mat_mul	
  (100)	
  
Mat_vec_mul	
  (1000)	
  
Mat_mul2	
  (100)	
  
2.38	
  s	
  
2.13	
  s	
  
2.08	
  s	
  
0.0011	
  s	
  
0.0016	
  s	
  
0.0016	
  s	
  
x2000	
  
x1300	
  
x1300	
  
Mymesh	
  
Skyline	
  
Matrice	
  
0.64	
  s	
  
14.55	
  s	
  
11.96	
  s	
  
0.0014	
  s	
  
0.035	
  s	
  
0.054	
  s	
  
x450	
  
x400	
  
x200	
  
Mymesh	
  	
  triangular	
  mesh	
  generaAon	
  
Skyline	
  	
  	
  	
  	
  	
  morse	
  matrix	
  indexing	
  building	
  
Matrice 	
  morse	
  matrix	
  building	
  
9 - ScilabTec 2015
COLD:	
  a	
  Scilab-­‐to-­‐C++	
  compiler	
  
Inputs	
  
▶  Scripts	
  and/or	
  funcAons	
  
▶  Large	
  subset	
  of	
  language	
  basics	
  
–  scalar	
  types	
  (int,	
  real,	
  complex),	
  N-­‐D	
  arrays,	
  sparse	
  matrices,	
  structures	
  
–  all	
  control	
  flow	
  structures	
  
–  user	
  funcAons…	
  
▶  Large	
  subset	
  of	
  Scilab	
  funcAons	
  (>	
  200)	
  
	
  
Outputs	
  
▶  C++11	
  high-­‐performance	
  code	
  
▶  Standalone	
  executables	
  from	
  scripts	
  
▶  Scilab	
  libraries	
  
▶  Libraries	
  with	
  specific	
  interfaces	
  (for	
  integraAon	
  in	
  user	
  libraries)	
  
▶  Human-­‐readable	
  or	
  obfuscated	
  code	
  
	
  
10 - ScilabTec 2015
COLD:	
  specific	
  features	
  
High-­‐performance	
  C++	
  code	
  
▶  Strong	
  type	
  inference	
  mechanism:	
  no	
  dynamic	
  run-­‐Ame	
  overhead	
  
▶  CompilaAon	
  opAmizaAons	
  
▶  MathemaAcal	
  opAmizaAons	
  
▶  ParallelizaAon	
  
▶  Extends	
  input	
  data-­‐size	
  /	
  parameter	
  values	
  coverage	
  
User	
  friendliness	
  
▶  Windows	
  /	
  Linux	
  
▶  Compliance	
  analysis	
  to	
  detect	
  unsupported	
  features	
  and	
  opAmizaAon	
  
obstacles	
  
▶  Fast	
  compilaAon	
  (5000	
  lines	
  of	
  Scilab:	
  <	
  0.5s	
  to	
  generate	
  C++	
  code)	
  
11 - ScilabTec 2015
COLDTM	
  …	
  
…	
  generates	
  high-­‐performance	
  code	
  from	
  Scilab	
  codes	
  …	
  
▶  stand-­‐alone	
  applica7ons	
  
▶  op7mized	
  libraries,	
  re-­‐usable	
  from	
  Scilab	
  environment	
  
	
  
…	
  offers	
  …	
  
▶  reducAon	
  of	
  Ame-­‐to-­‐demo	
  or	
  Ame-­‐to-­‐product	
  
▶  code	
  obfuscaAon,	
  intellectual	
  property	
  protec7on	
  
…	
  evolves	
  con?nuously	
  and	
  benefits	
  from	
  …	
  
▶  industrial	
  partnerships	
  (real	
  use-­‐cases)	
  
▶  collaboraAve	
  research	
  projects	
  (FUI	
  Richelieu,	
  FUI	
  Similan,	
  ITEA2	
  Mach)	
  
	
  
…	
  is	
  a	
  scalable	
  product	
  …	
  
▶  on-­‐demand	
  customiza7on	
  for	
  specific	
  needs	
  
	
  
Contact	
  us	
  on	
  cold-­‐support@silkan.com	
  

Mais conteúdo relacionado

Mais procurados

Classification of Compiler
Classification of CompilerClassification of Compiler
Classification of Compiler
maharajdey
 
Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...
Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...
Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...
Thanos Zolotas
 

Mais procurados (20)

scilab
scilabscilab
scilab
 
Scilab for very beginners
Scilab for very beginnersScilab for very beginners
Scilab for very beginners
 
Incubating Apache Hivemall
Incubating Apache HivemallIncubating Apache Hivemall
Incubating Apache Hivemall
 
Hivemall meets Digdag @Hackertackle 2018-02-17
Hivemall meets Digdag @Hackertackle 2018-02-17Hivemall meets Digdag @Hackertackle 2018-02-17
Hivemall meets Digdag @Hackertackle 2018-02-17
 
Local Optimizations in Eclipse QVTc and QVTr using the Micro-Mapping Model of...
Local Optimizations in Eclipse QVTc and QVTr using the Micro-Mapping Model of...Local Optimizations in Eclipse QVTc and QVTr using the Micro-Mapping Model of...
Local Optimizations in Eclipse QVTc and QVTr using the Micro-Mapping Model of...
 
Optimized declarative transformation First Eclipse QVTc results
Optimized declarative transformation First Eclipse QVTc resultsOptimized declarative transformation First Eclipse QVTc results
Optimized declarative transformation First Eclipse QVTc results
 
Simulating Large-scale Aggregate MASs with Alchemist and Scala
Simulating Large-scale Aggregate MASs with Alchemist and ScalaSimulating Large-scale Aggregate MASs with Alchemist and Scala
Simulating Large-scale Aggregate MASs with Alchemist and Scala
 
Functional programming-advantages
Functional programming-advantagesFunctional programming-advantages
Functional programming-advantages
 
Experiments & Experiences with Scilab in Undergraduate Education
Experiments & Experiences with Scilab in Undergraduate Education Experiments & Experiences with Scilab in Undergraduate Education
Experiments & Experiences with Scilab in Undergraduate Education
 
Scilab
ScilabScilab
Scilab
 
QVT Traceability: What does it really mean?
QVT Traceability: What does it really mean?QVT Traceability: What does it really mean?
QVT Traceability: What does it really mean?
 
Classification of Compiler
Classification of CompilerClassification of Compiler
Classification of Compiler
 
Commutative Short Circuit Operators
Commutative Short Circuit OperatorsCommutative Short Circuit Operators
Commutative Short Circuit Operators
 
R4ML: An R Based Scalable Machine Learning Framework
R4ML: An R Based Scalable Machine Learning FrameworkR4ML: An R Based Scalable Machine Learning Framework
R4ML: An R Based Scalable Machine Learning Framework
 
TMPA-2017: Modeling of PLC-programs by High-level Coloured Petri Nets
TMPA-2017: Modeling of PLC-programs by High-level Coloured Petri NetsTMPA-2017: Modeling of PLC-programs by High-level Coloured Petri Nets
TMPA-2017: Modeling of PLC-programs by High-level Coloured Petri Nets
 
MATLAB/Simulink test02
MATLAB/Simulink test02MATLAB/Simulink test02
MATLAB/Simulink test02
 
Microcontrollers 80 Marks Sample Question Paper
Microcontrollers   80 Marks Sample Question PaperMicrocontrollers   80 Marks Sample Question Paper
Microcontrollers 80 Marks Sample Question Paper
 
Verilog data types -For beginners
Verilog data types -For beginnersVerilog data types -For beginners
Verilog data types -For beginners
 
Image Caption Generation: Intro to Distributed Tensorflow and Distributed Sco...
Image Caption Generation: Intro to Distributed Tensorflow and Distributed Sco...Image Caption Generation: Intro to Distributed Tensorflow and Distributed Sco...
Image Caption Generation: Intro to Distributed Tensorflow and Distributed Sco...
 
Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...
Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...
Bridging Proprietary Modelling and Open-Source Model Management Tools: The Ca...
 

Semelhante a ScilabTEC 2015 - Silkan

Cray XT Porting, Scaling, and Optimization Best Practices
Cray XT Porting, Scaling, and Optimization Best PracticesCray XT Porting, Scaling, and Optimization Best Practices
Cray XT Porting, Scaling, and Optimization Best Practices
Jeff Larkin
 
May2010 hex-core-opt
May2010 hex-core-optMay2010 hex-core-opt
May2010 hex-core-opt
Jeff Larkin
 

Semelhante a ScilabTEC 2015 - Silkan (20)

Software Frameworks for Deep Learning (D1L7 2017 UPC Deep Learning for Comput...
Software Frameworks for Deep Learning (D1L7 2017 UPC Deep Learning for Comput...Software Frameworks for Deep Learning (D1L7 2017 UPC Deep Learning for Comput...
Software Frameworks for Deep Learning (D1L7 2017 UPC Deep Learning for Comput...
 
Scilab Technical Talk at NTU, TP and HCMUT (Dr Claude Gomez)
Scilab Technical Talk at NTU, TP and HCMUT (Dr Claude Gomez)Scilab Technical Talk at NTU, TP and HCMUT (Dr Claude Gomez)
Scilab Technical Talk at NTU, TP and HCMUT (Dr Claude Gomez)
 
Onnc intro
Onnc introOnnc intro
Onnc intro
 
Cray XT Porting, Scaling, and Optimization Best Practices
Cray XT Porting, Scaling, and Optimization Best PracticesCray XT Porting, Scaling, and Optimization Best Practices
Cray XT Porting, Scaling, and Optimization Best Practices
 
Bounded Model Checking for C Programs in an Enterprise Environment
Bounded Model Checking for C Programs in an Enterprise EnvironmentBounded Model Checking for C Programs in an Enterprise Environment
Bounded Model Checking for C Programs in an Enterprise Environment
 
What’s New in ScyllaDB Open Source 5.0
What’s New in ScyllaDB Open Source 5.0What’s New in ScyllaDB Open Source 5.0
What’s New in ScyllaDB Open Source 5.0
 
Eugene Burmako
Eugene BurmakoEugene Burmako
Eugene Burmako
 
Scikit-Learn: Machine Learning in Python
Scikit-Learn: Machine Learning in PythonScikit-Learn: Machine Learning in Python
Scikit-Learn: Machine Learning in Python
 
Pitfalls of machine learning in production
Pitfalls of machine learning in productionPitfalls of machine learning in production
Pitfalls of machine learning in production
 
TensorFlow meetup: Keras - Pytorch - TensorFlow.js
TensorFlow meetup: Keras - Pytorch - TensorFlow.jsTensorFlow meetup: Keras - Pytorch - TensorFlow.js
TensorFlow meetup: Keras - Pytorch - TensorFlow.js
 
H2O Design and Infrastructure with Matt Dowle
H2O Design and Infrastructure with Matt DowleH2O Design and Infrastructure with Matt Dowle
H2O Design and Infrastructure with Matt Dowle
 
It's always sunny with OpenJ9
It's always sunny with OpenJ9It's always sunny with OpenJ9
It's always sunny with OpenJ9
 
Porting a Streaming Pipeline from Scala to Rust
Porting a Streaming Pipeline from Scala to RustPorting a Streaming Pipeline from Scala to Rust
Porting a Streaming Pipeline from Scala to Rust
 
Deep Learning for Computer Vision: Software Frameworks (UPC 2016)
Deep Learning for Computer Vision: Software Frameworks (UPC 2016)Deep Learning for Computer Vision: Software Frameworks (UPC 2016)
Deep Learning for Computer Vision: Software Frameworks (UPC 2016)
 
Kubecon seattle 2018 workshop slides
Kubecon seattle 2018 workshop slidesKubecon seattle 2018 workshop slides
Kubecon seattle 2018 workshop slides
 
May2010 hex-core-opt
May2010 hex-core-optMay2010 hex-core-opt
May2010 hex-core-opt
 
XConf 2022 - Code As Data: How data insights on legacy codebases can fill the...
XConf 2022 - Code As Data: How data insights on legacy codebases can fill the...XConf 2022 - Code As Data: How data insights on legacy codebases can fill the...
XConf 2022 - Code As Data: How data insights on legacy codebases can fill the...
 
Webinar: Começando seus trabalhos com Machine Learning utilizando ferramentas...
Webinar: Começando seus trabalhos com Machine Learning utilizando ferramentas...Webinar: Começando seus trabalhos com Machine Learning utilizando ferramentas...
Webinar: Começando seus trabalhos com Machine Learning utilizando ferramentas...
 
Cling the llvm based interpreter
Cling the llvm based interpreterCling the llvm based interpreter
Cling the llvm based interpreter
 
Clipper: A Low-Latency Online Prediction Serving System
Clipper: A Low-Latency Online Prediction Serving SystemClipper: A Low-Latency Online Prediction Serving System
Clipper: A Low-Latency Online Prediction Serving System
 

Mais de Scilab

Mais de Scilab (20)

Statistical Analysis for Robust Design
Statistical Analysis for Robust DesignStatistical Analysis for Robust Design
Statistical Analysis for Robust Design
 
Electric motor optimization
Electric motor optimizationElectric motor optimization
Electric motor optimization
 
Asteroidlanding - Scilab conference 2019 Keynote
Asteroidlanding - Scilab conference 2019 KeynoteAsteroidlanding - Scilab conference 2019 Keynote
Asteroidlanding - Scilab conference 2019 Keynote
 
Faster Time to Market using Scilab/XCOS/X2C for motor control algorithm devel...
Faster Time to Market using Scilab/XCOS/X2C for motor control algorithm devel...Faster Time to Market using Scilab/XCOS/X2C for motor control algorithm devel...
Faster Time to Market using Scilab/XCOS/X2C for motor control algorithm devel...
 
Scilab and Xcos for Very Low Earth Orbits satellites modelling
Scilab and Xcos for Very Low Earth Orbits satellites modellingScilab and Xcos for Very Low Earth Orbits satellites modelling
Scilab and Xcos for Very Low Earth Orbits satellites modelling
 
X2C -a tool for model-based control development and automated code generation...
X2C -a tool for model-based control development and automated code generation...X2C -a tool for model-based control development and automated code generation...
X2C -a tool for model-based control development and automated code generation...
 
A Real-Time Interface for Xcos – an illustrative demonstration using a batter...
A Real-Time Interface for Xcos – an illustrative demonstration using a batter...A Real-Time Interface for Xcos – an illustrative demonstration using a batter...
A Real-Time Interface for Xcos – an illustrative demonstration using a batter...
 
Aircraft Simulation Model and Flight Control Laws Design Using Scilab and XCos
Aircraft Simulation Model and Flight Control Laws Design Using Scilab and XCosAircraft Simulation Model and Flight Control Laws Design Using Scilab and XCos
Aircraft Simulation Model and Flight Control Laws Design Using Scilab and XCos
 
Scilab for real dummies j.heikell - part3
Scilab for real dummies j.heikell - part3Scilab for real dummies j.heikell - part3
Scilab for real dummies j.heikell - part3
 
Scilab for real dummies j.heikell - part 2
Scilab for real dummies j.heikell - part 2Scilab for real dummies j.heikell - part 2
Scilab for real dummies j.heikell - part 2
 
Scilab for real dummies j.heikell - part 1
Scilab for real dummies j.heikell - part 1Scilab for real dummies j.heikell - part 1
Scilab for real dummies j.heikell - part 1
 
Multiobjective optimization and Genetic algorithms in Scilab
Multiobjective optimization and Genetic algorithms in ScilabMultiobjective optimization and Genetic algorithms in Scilab
Multiobjective optimization and Genetic algorithms in Scilab
 
Scilab optimization workshop
Scilab optimization workshop Scilab optimization workshop
Scilab optimization workshop
 
INRA @ Scilab Conference 2018
INRA @ Scilab Conference 2018INRA @ Scilab Conference 2018
INRA @ Scilab Conference 2018
 
Qualcomm @ Scilab Conference 2018
Qualcomm @ Scilab Conference 2018Qualcomm @ Scilab Conference 2018
Qualcomm @ Scilab Conference 2018
 
Sanofi @ Scilab Conference 2018
Sanofi @ Scilab Conference 2018Sanofi @ Scilab Conference 2018
Sanofi @ Scilab Conference 2018
 
University of Applied Science Esslingen @ Scilab Conference 2018
University of Applied Science Esslingen @ Scilab Conference 2018University of Applied Science Esslingen @ Scilab Conference 2018
University of Applied Science Esslingen @ Scilab Conference 2018
 
DLR @ Scilab Conference 2018
DLR @ Scilab Conference 2018DLR @ Scilab Conference 2018
DLR @ Scilab Conference 2018
 
Fraunhofer IIS @ Scilab Conference 2018
Fraunhofer IIS @ Scilab Conference 2018Fraunhofer IIS @ Scilab Conference 2018
Fraunhofer IIS @ Scilab Conference 2018
 
Arcelormittal @ Scilab Conference 2018
Arcelormittal @ Scilab Conference 2018Arcelormittal @ Scilab Conference 2018
Arcelormittal @ Scilab Conference 2018
 

Último

Uncommon Grace The Autobiography of Isaac Folorunso
Uncommon Grace The Autobiography of Isaac FolorunsoUncommon Grace The Autobiography of Isaac Folorunso
Uncommon Grace The Autobiography of Isaac Folorunso
Kayode Fayemi
 
Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...
Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...
Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...
amilabibi1
 
If this Giant Must Walk: A Manifesto for a New Nigeria
If this Giant Must Walk: A Manifesto for a New NigeriaIf this Giant Must Walk: A Manifesto for a New Nigeria
If this Giant Must Walk: A Manifesto for a New Nigeria
Kayode Fayemi
 
Chiulli_Aurora_Oman_Raffaele_Beowulf.pptx
Chiulli_Aurora_Oman_Raffaele_Beowulf.pptxChiulli_Aurora_Oman_Raffaele_Beowulf.pptx
Chiulli_Aurora_Oman_Raffaele_Beowulf.pptx
raffaeleoman
 

Último (18)

Dreaming Music Video Treatment _ Project & Portfolio III
Dreaming Music Video Treatment _ Project & Portfolio IIIDreaming Music Video Treatment _ Project & Portfolio III
Dreaming Music Video Treatment _ Project & Portfolio III
 
My Presentation "In Your Hands" by Halle Bailey
My Presentation "In Your Hands" by Halle BaileyMy Presentation "In Your Hands" by Halle Bailey
My Presentation "In Your Hands" by Halle Bailey
 
Digital collaboration with Microsoft 365 as extension of Drupal
Digital collaboration with Microsoft 365 as extension of DrupalDigital collaboration with Microsoft 365 as extension of Drupal
Digital collaboration with Microsoft 365 as extension of Drupal
 
Uncommon Grace The Autobiography of Isaac Folorunso
Uncommon Grace The Autobiography of Isaac FolorunsoUncommon Grace The Autobiography of Isaac Folorunso
Uncommon Grace The Autobiography of Isaac Folorunso
 
The workplace ecosystem of the future 24.4.2024 Fabritius_share ii.pdf
The workplace ecosystem of the future 24.4.2024 Fabritius_share ii.pdfThe workplace ecosystem of the future 24.4.2024 Fabritius_share ii.pdf
The workplace ecosystem of the future 24.4.2024 Fabritius_share ii.pdf
 
Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...
Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...
Bring back lost lover in USA, Canada ,Uk ,Australia ,London Lost Love Spell C...
 
If this Giant Must Walk: A Manifesto for a New Nigeria
If this Giant Must Walk: A Manifesto for a New NigeriaIf this Giant Must Walk: A Manifesto for a New Nigeria
If this Giant Must Walk: A Manifesto for a New Nigeria
 
Thirunelveli call girls Tamil escorts 7877702510
Thirunelveli call girls Tamil escorts 7877702510Thirunelveli call girls Tamil escorts 7877702510
Thirunelveli call girls Tamil escorts 7877702510
 
Causes of poverty in France presentation.pptx
Causes of poverty in France presentation.pptxCauses of poverty in France presentation.pptx
Causes of poverty in France presentation.pptx
 
Report Writing Webinar Training
Report Writing Webinar TrainingReport Writing Webinar Training
Report Writing Webinar Training
 
Busty Desi⚡Call Girls in Sector 51 Noida Escorts >༒8448380779 Escort Service-...
Busty Desi⚡Call Girls in Sector 51 Noida Escorts >༒8448380779 Escort Service-...Busty Desi⚡Call Girls in Sector 51 Noida Escorts >༒8448380779 Escort Service-...
Busty Desi⚡Call Girls in Sector 51 Noida Escorts >༒8448380779 Escort Service-...
 
Aesthetic Colaba Mumbai Cst Call girls 📞 7738631006 Grant road Call Girls ❤️-...
Aesthetic Colaba Mumbai Cst Call girls 📞 7738631006 Grant road Call Girls ❤️-...Aesthetic Colaba Mumbai Cst Call girls 📞 7738631006 Grant road Call Girls ❤️-...
Aesthetic Colaba Mumbai Cst Call girls 📞 7738631006 Grant road Call Girls ❤️-...
 
Dreaming Marissa Sánchez Music Video Treatment
Dreaming Marissa Sánchez Music Video TreatmentDreaming Marissa Sánchez Music Video Treatment
Dreaming Marissa Sánchez Music Video Treatment
 
Chiulli_Aurora_Oman_Raffaele_Beowulf.pptx
Chiulli_Aurora_Oman_Raffaele_Beowulf.pptxChiulli_Aurora_Oman_Raffaele_Beowulf.pptx
Chiulli_Aurora_Oman_Raffaele_Beowulf.pptx
 
ICT role in 21st century education and it's challenges.pdf
ICT role in 21st century education and it's challenges.pdfICT role in 21st century education and it's challenges.pdf
ICT role in 21st century education and it's challenges.pdf
 
AWS Data Engineer Associate (DEA-C01) Exam Dumps 2024.pdf
AWS Data Engineer Associate (DEA-C01) Exam Dumps 2024.pdfAWS Data Engineer Associate (DEA-C01) Exam Dumps 2024.pdf
AWS Data Engineer Associate (DEA-C01) Exam Dumps 2024.pdf
 
lONG QUESTION ANSWER PAKISTAN STUDIES10.
lONG QUESTION ANSWER PAKISTAN STUDIES10.lONG QUESTION ANSWER PAKISTAN STUDIES10.
lONG QUESTION ANSWER PAKISTAN STUDIES10.
 
Sector 62, Noida Call girls :8448380779 Noida Escorts | 100% verified
Sector 62, Noida Call girls :8448380779 Noida Escorts | 100% verifiedSector 62, Noida Call girls :8448380779 Noida Escorts | 100% verified
Sector 62, Noida Call girls :8448380779 Noida Escorts | 100% verified
 

ScilabTEC 2015 - Silkan

  • 1. COLDTM: High Performance with Scilab Boosting your workflow – Boosting your Performances Béatrice  Creusillet  &  Thomas  Izard   ScilabTec  2015,  Paris  
  • 2. 2 - ScilabTec 2015 Development  workflow  
  • 3. 3 - ScilabTec 2015 Development  workflow  
  • 4. 4 - ScilabTec 2015 Development  workflow  
  • 5. 5 - ScilabTec 2015 From  code  to  binaries  or  libraries  
  • 6. 6 - ScilabTec 2015 ▶  CFD  applicaAon  based  on  LaDce  Boltzmann  method1   ▶  IniAalizaAon  step:  cfd_init.sce   ▶  Time  loop:   –  computaAon  kernel:  cfd_iteraAon.sce   –  visualizaAon   –  10000  cycles            (400000  originally)            1  From  wiki.palabos.org/numerics:codes   Copyright  ©  2006-­‐2008  by  Jonas  LaW   A  prac7cal  case  
  • 7. 7 - ScilabTec 2015 Performances     Scilab   Scilab   +   COLD   (kernel)   Scilab   +   COLD   (main  script)     COLD     (executable)   average   kernel   duraAon   390  ms   24  ms   12  ms   12  ms   total  Ame   for  10000   cycles   1  h   3  min  39  s   1  min  56  s   1  min  52  s   speed-­‐up   x  16   x  31   x  32   Experimental  condiAons:   •  Intel  Core  i7-­‐3770  @  3.40GHz,  16Go,  Ubuntu  14.04   •  Scilab  5.5.2,  COLD  2.1,  gcc  4.8.2  
  • 8. 8 - ScilabTec 2015 Some  more  performances…   ▶  Series  of  benchmarks  from  an  industrial  partner  (Richelieu  Project)   ▶  Raw  code,  no  calls  to  Scilab  library  funcAons  or  vector  operaAons   Scilab  5.4.1   COLDTM  1.6.3   Speed-­‐up   Mat_mul  (100)   Mat_vec_mul  (1000)   Mat_mul2  (100)   2.38  s   2.13  s   2.08  s   0.0011  s   0.0016  s   0.0016  s   x2000   x1300   x1300   Mymesh   Skyline   Matrice   0.64  s   14.55  s   11.96  s   0.0014  s   0.035  s   0.054  s   x450   x400   x200   Mymesh    triangular  mesh  generaAon   Skyline            morse  matrix  indexing  building   Matrice  morse  matrix  building  
  • 9. 9 - ScilabTec 2015 COLD:  a  Scilab-­‐to-­‐C++  compiler   Inputs   ▶  Scripts  and/or  funcAons   ▶  Large  subset  of  language  basics   –  scalar  types  (int,  real,  complex),  N-­‐D  arrays,  sparse  matrices,  structures   –  all  control  flow  structures   –  user  funcAons…   ▶  Large  subset  of  Scilab  funcAons  (>  200)     Outputs   ▶  C++11  high-­‐performance  code   ▶  Standalone  executables  from  scripts   ▶  Scilab  libraries   ▶  Libraries  with  specific  interfaces  (for  integraAon  in  user  libraries)   ▶  Human-­‐readable  or  obfuscated  code    
  • 10. 10 - ScilabTec 2015 COLD:  specific  features   High-­‐performance  C++  code   ▶  Strong  type  inference  mechanism:  no  dynamic  run-­‐Ame  overhead   ▶  CompilaAon  opAmizaAons   ▶  MathemaAcal  opAmizaAons   ▶  ParallelizaAon   ▶  Extends  input  data-­‐size  /  parameter  values  coverage   User  friendliness   ▶  Windows  /  Linux   ▶  Compliance  analysis  to  detect  unsupported  features  and  opAmizaAon   obstacles   ▶  Fast  compilaAon  (5000  lines  of  Scilab:  <  0.5s  to  generate  C++  code)  
  • 11. 11 - ScilabTec 2015 COLDTM  …   …  generates  high-­‐performance  code  from  Scilab  codes  …   ▶  stand-­‐alone  applica7ons   ▶  op7mized  libraries,  re-­‐usable  from  Scilab  environment     …  offers  …   ▶  reducAon  of  Ame-­‐to-­‐demo  or  Ame-­‐to-­‐product   ▶  code  obfuscaAon,  intellectual  property  protec7on   …  evolves  con?nuously  and  benefits  from  …   ▶  industrial  partnerships  (real  use-­‐cases)   ▶  collaboraAve  research  projects  (FUI  Richelieu,  FUI  Similan,  ITEA2  Mach)     …  is  a  scalable  product  …   ▶  on-­‐demand  customiza7on  for  specific  needs     Contact  us  on  cold-­‐support@silkan.com