
Louis-Noël Pouchet
Dept. of Computer Science and Engineering
2015 Neil Avenue
Columbus, OH 43210
tel: (+1) 614-859-5115
email: pouchet@cse.ohio-state.edu
office: DL-589
|
|
[News] [Position] [Software]
[Lectures] [Publications]
News
- 07/31/2012: Last day at OSU. I have moved to UCLA. Please now use my new webpage: http://www.cs.ucla.edu/~pouchet for updated news. My OSU website is not maintained anymore, and will be closed at an undetermined time.
- 04/23/2012: Publication list updated (3 new papers).
- 03/29/2012: PolyBench/GPU
1.0 is released. Contributed by John Cavazos group at University of Delaware.
- 03/28/2012: PolyBench/Fortran 1.0 is released. Port of PolyBench/C 3.2, written by Mohanish Narayan.
- 03/12/2012: PolyOpt/C 0.2.1 is released. PolyOpt now uses the latest PoCC version (1.1). Numerous bug fixes and enhancements.
- 03/12/2012: PoCC 1.1 is released. Numerous bug fixes and many new features, now uses the PAST back-end by default.
- 03/12/2012: PolyBench/C 3.2 is released. Minor bug fixes.
- 01/27/2012: PolyOpt/C source code is available.
- 01/27/2012: Publication list updated (2 new papers, 3 new presentations).
Position
I am currently a post-doctoral researcher at Ohio State University,
working with
Prof. P. Sadayappan. I
work on a variety of topics for high-performance computing, and in
particular I develop compiler technologies based on the polyhedral
framework. I am a member of
the Center for Domain-Specific
Computing (NSF). I was also an active member of
the Platform-Aware Compilation
Environment project (DARPA), where I contributed a polyhedral
compilation engine
(PolyOpt) for
the ROSE compiler.
I received a Ph.D in Computer Science in Jan. 2010, for my work at
the ALCHEMY group,
INRIA Saclay. I was advised
by Albert Cohen
and Cédric
Bastoul. In 2006, I graduated
from EPITA (a French Engineer school
specialized in Computer Science) and received a Master's degree in
Computer Science (minor in Cognitive Sciences) from
the University of Paris-Sud
XI.
I am working on several areas of high-performance computing:
- polyhedral compilation
- iterative and adaptive compilation / auto-tuning compilers
- performance portability
- machine learning
- performance-oriented domain-specific languages
- hardware-software co-design
- energy-aware program optimization
- program transformations for high-level synthesis
- programming languages
- I/O complexity models
Software
Compilation
-
PoCC: the Polyhedral Compiler Collection
A flexible source-to-source iterative
and model-driven compiler, embedding most of the state-of-the-art
tools for polyhedral compilation.
Status: On development, version 1.1 available.
-
PolyOpt/C: A Polyhedral Optimizer for the ROSE compiler
A complete polyhedral framework (data dependence analysis, annotation, program transformations) accessible in the ROSE compiler. Automatic detection and optimization of polyhedral regions.
Status: On development, version 0.2.1 available.
-
LeTSeE: the LEgal Transformation SpacE Explorator
Tool to build and explore
systematically or heuristically the legal transformation space of a
program, in the polyhedral model.
Status: Version 0.2.0 available.
-
FM: the Fourier-Motzkin library
A library for manipulating
Q-polyhedra. It comprises a fast and redundancy-controlled C
implementation of the Fourier-Motzkin projection algorithm.
Status: Version 0.5.0 available.
Benchmarking
-
PolyBench/C
A set of reference C files containing Static Control Parts.
Status: Version 3.2 available.
-
PolyBench/Fortran
A set of reference Fortran files containing Static Control Parts.
Status: Version 1.0 available.
Miscellaneous
-
AutoSkelet
A tunable project skeleton generator using Autoconf, Automake and Libtools.
Status: Version 0.1 available.
-
LoopbackPortForwarding
A small Java app to forward ports
between your public IP address and the loopback.
Proven to be very useful to allow Internet users to connect
to Google Android apps
running on your PC!
Status: Version 0.1
available.
LetSee, PACE, AutoSkelet
and LoopbackPortForwarding are
released under the GNU GPL (v2) license. FM is released under the GNU Lesser GPL (v3) license starting version 0.5.0, and GPL v2 for previous versions.
Lectures
Part I: Theory
- 888.11 - Polyhedral Compilation Foundations, lecture I.1: Iteration domains and polyhedra
|
[slides] [notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture I.2: Data dependences
|
[slides] [notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture I.3: Scheduling and Optimization - I
|
[slides] [notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture I.4: Scheduling and Optimization - II
|
[slides] [notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture I.5: Scheduling and Optimization - III
|
[slides] [notes]
|
Part II: Practice
- 888.11 - Polyhedral Compilation Foundations, lecture II.1: Dependence polyhedra
|
[notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture II.2: Dependence polyhedra, again
|
[notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture II.3: Dependence polyhedra and program Legality
|
[notes]
|
- 888.11 - Polyhedral Compilation Foundations, lecture II.4: Scanning Polyhedra
|
[notes]
|
Part III: Communication management
- 888.11 - Polyhedral Compilation Foundations, exercise III.1: Data space
|
[notes]
|
- 888.11 - Polyhedral Compilation Foundations, exercise III.2: Tiling
|
[notes]
|
- 888.11 - Polyhedral Compilation Foundations, exercise III.3: Communication generation
|
[notes]
|
Methodology and problem solving
- 888.11 - Iterator Live-out value, lecture 1: Formulating the problem
|
[slides] [notes]
|
- 888.11 - Iterator Live-out value, lecture 2: Writing algorithms
|
[slides] [notes]
|
- 888.11 - Iterator Live-out value, lecture 3: Solving the problem, part I
|
[slides] [notes]
|
- 888.11 - Iterator Live-out value, lecture 4: Solving the problem, part II
|
[slides] [notes]
|
- 888.11 - Iterator Live-out value, lecture 5: General algorithm for the Live-out Iterator Problem
|
[slides] [notes]
|
- 888.11 - Iterator Live-out value, lecture 6: "Proving" your algorithm
|
[slides] [notes]
|
Randomly selected topics
- 888.11 - Vectorization in the Polyhedral Model, lecture 1: Pre-transformations for vectorizability
|
[slides] [notes]
|
- 888.11 - Vectorization in the Polyhedral Model, lecture 2: Transformation Selection for Good Vectorization
|
[slides] [notes]
|
Publications
In conferences and journals
-
Justin Holewinski, Louis-Noël Pouchet and P. Sadayappan.
High-Performance Code Generation for Stencil Computations on GPU Architectures.
In ACM International Conference on Supercomputing (ICS'12).
ACM Press, Venice, Italy, June 2012. To appear.
[bibtex-entry] [pdf] [slides]
-
Justin Holewinski, Ragavendar Ramamurthi, Mahesh Ravishankar, Naznin Fauzia, Louis-Noël Pouchet, Atanas Rountev and P. Sadayappan.
Dynamic Trace-Based Analysis of Vectorization Potential of Applications.
In ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'12).
ACM Press, Beijing, China, June 2012. To appear.
[bibtex-entry] [pdf] [slides]
-
Jun Shirako, Kamal Sharma, Naznin Fauzia, Louis-Noël Pouchet,
J. Ramanujam,P. Sadayappan and Vivek Sarkar.
Analytical Bounds for Optimal Tile Size Selection.
In ETAPS International Conference on Compiler Construction (CC'2012).
Springer Verlag, Tallinn, Estonia, March 2012. To appear.
[bibtex-entry] [pdf] [slides]
-
Kevin Stock, Louis-Noël Pouchet and P. Sadayappan.
Using Machine Learning to Improve Automatic Vectorization.
In ACM Transactions on Architecture and Code Optimization
(TACO).
Special issue on high-performance and embedded
architectures and compilers, Volume 8 Issue 4, January 2012.
[bibtex-entry] [pdf] [[slides]
-
Sanket Tavarageri, Louis-Noël Pouchet, Ramanujam, Atanas Rountev and P. Sadayappan.
Dynamic Selection of Tile Sizes.
In 18th IEEE International Conference on High Performance Computing (HiPC'11).
Bangalore, India, December 2011.
[bibtex-entry] [pdf] [slides]
-
Naser Sedaghati, Renji Thomas, Louis-Noël Pouchet, Radu Teodorescu and P. Sadayappan.
StVEC: A Vector Instruction Extension for High Performance Stencil Computation.
In 20th IEEE International Conference on Parallel Architecture and Compilation Techniques (PACT'11).
Galveston Island, Texas, October 2011.
[bibtex-entry] [pdf] [slides]
-
Eunjung Park, Louis-Noël Pouchet, John Cavazos, Albert Cohen and P. Sadayappan.
Predictive Modeling in a Polyhedral Optimization Space.
In 9th IEEE/ACM International Symposium on Code Generation and Optimization (CGO'11).
pp 119--129, Chamonix, France, April 2011.
[bibtex-entry] [pdf] [slides]
-
Tom Henretty, Kevin Stock, Louis-Noël Pouchet, Franz Franchetti, J. Ramanujam and P. Sadayappan.
Data Layout Transformation for Stencil Computations on Short-Vector SIMD
Architectures.
In ETAPS International Conference on Compiler Construction (CC'2011).
pp 225--245, Springer Verlag, Saarbrucken, Germany, March 2011.
[bibtex-entry] [pdf] [slides]
-
Louis-Noël Pouchet, Uday Bondhugula, Cédric Bastoul, Albert Cohen, J. Ramanujam, P. Sadayappan and Nicolas Vasilache.
Loop Transformations: Convexity, Pruning and Optimization.
In 38th ACM SIGACT-SIGPLAN Symposium on Principles of Programming Languages (POPL'11).
pp 549--562, Austin, TX, January 2011. Also appeared in ACM SIGPLAN Notices 46(1):549--562.
[bibtex-entry] [pdf] [slides]
-
H. Munk, E. Ayguadé, C. Bastoul, P. Carpenter, Z. Chamski, A. Cohen,
M. Cornero, P. Dumont, M. Duranton, M. Fellahi, R. Ferrer,
R. Ladelsky, M. Lindwer, X. Martorell, C. Miranda, D. Nuzman,
A. Ornstein, A. Pop, S. Pop, L.-N. Pouchet, A. Ramirez, D. Rodenas,
E. Rohou, I. Rosen, U. Shvadron, K. Trifunovic, and A. Zaks.
ACOTES
Project: Advanced Compiler Technologies for Embedded
Streaming.
In International Journal of Parallel Programming
(IJPP).
2011. Special issue on European HiPEAC network of
excellence member's projects.
[bibtex-entry] [pdf]
-
Louis-Noël Pouchet, Uday Bondhugula, Cédric Bastoul, Albert Cohen, J. Ramanujam and P. Sadayappan.
Combined Iterative and Model-driven Optimization in an
Automatic Parallelization Framework.
In IEEE Conference on Supercomputing (SC'2010).
New Orleans, LA, November 2010.
[bibtex-entry] [pdf] [slides]
-
Mohamed-Walid Benabderrahmane,
Louis-Noël Pouchet, Albert Cohen, Cédric Bastoul.
The Polyhedral Model Is More Widely Applicable Than You Think.
In ETAPS International Conference on Compiler Construction (CC'2010).
pp 283--303, Springer Verlag, Paphos, Cyprus, March 2010.
[bibtex-entry] [pdf] [slides]
-
Louis-Noël Pouchet, Cédric
Bastoul, Albert Cohen and John Cavazos.
Iterative Pptimization
in the Polyhedral Model: Part II, Multidimensional Time.
In ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'08).
pp 90--100, ACM Press, Tucson, Arizona, June 2008. Also appeared in ACM SIGPLAN Notices 43(6):90--100.
[bibtex-entry] [pdf] [slides]
-
Nicolas Vasilache, Albert Cohen and Louis-Noël Pouchet.
Automatic Correction of Loop Transformations.
In 16th IEEE International Conference on Parallel Architecture and Compilation Techniques (PACT'07).
pp 292--304, IEEE Computer Society Press, Brasov, Roumania, September 2007.
[bibtex-entry] [pdf]
-
Louis-Noël Pouchet, Cédric
Bastoul, Albert Cohen and Nicolas Vasilache.
Iterative Optimization
in the Polyhedral Model: Part I, One-dimensional Time.
In 5th IEEE/ACM International Symposium on Code Generation and Optimization (CGO'07).
pp 144--156, IEEE Computer Society Press, San Jose, California, March 2007.
[bibtex-entry] [pdf] [slides]
-
Thomas Claveirole, Sylvain Lombardy,
Sarah O'Connor, Louis-Noël Pouchet and Jacques
Sakarovitch.
Inside Vaucanson.
In 10th International Conference
on Implementation and Application of Automata (CIAA'05).
pp. 117--128, Lecture Notes in Computer Science Series, Vol. 3845,
Springer-Verlag, Sophia-Antipolis, France, June 2005.
[bibtex-entry] [pdf] [slides]
In workshops
-
Kevin Stock, Louis-Noël Pouchet and P. Sadayappan.
Automatic Transformations for Effective Parallel Execution on Intel Many Integrated Core.
In TACC-Intel Highly Parallel Computing Symposium.
Austin, TX,
April 2012.
[bibtex-entry] [pdf] [slides]
-
Tobias Grosser, Hongbin Zheng, Raghesh A, Andreas Simburger, Armin
Grosslinger and Louis-Noël Pouchet.
Polly - Polyhedral
Optimization in LLVM.
In First International Workshop on
Polyhedral Compilation Techniques (IMPACT'11).
Chamonix, France,
April 2011.
[bibtex-entry] [pdf] [slides]
-
Sanket Tavarageri, Albert Hartono, Muthu Baskaran, Louis-Noël Pouchet, J. Ramanujam and P. Sadayappan.
Parametric Tiling of Affine Loop Nests.
In
15th Workshop on Compilers for Parallel Computing (CPC'10).
Vienna, Austria, July 2010.
[bibtex-entry] [pdf]
-
Riyadh Baghdadi, Albert Cohen, Cédric Bastoul, Louis-Noël Pouchet and Lawrence Rauchwerger.
The Potential of Synergistic Static, Dynamic and Speculative Loop Nest Optimizations for Automatic Parallelization.
In 3rd Workshop on
Parallel Execution of Sequential Programs on Multi-core Architectures
(PESPMA'10).
Saint-Malo, France, June 2010. Position paper.
[bibtex-entry] [pdf]
-
Mohammed Rahman, Louis-Noël Pouchet and P. Sadayappan.
Neural
Network Assisted Tile Size Selection.
In 5th International
Workshop on Automatic Performance Tuning (iWAPT'10).
Berkeley,
CA, USA, June 2010.
[bibtex-entry] [pdf] [slides]
-
K. Ibrahim, J. Jaeger, Z. Liu, L.N. Pouchet, P. Lesnicki, L. Djoudi,
D.Barthou, F. Bodin, C. Eisenbeis, G. Grosdidier, O. Pene and
P. Roudeau.
Simulation of the Lattice QCD and Technological Trends in
Computation.
In 14th Workshop on Compilers for Parallel Computing
(CPC'09).
Zurich, Switzerland, January 2009.
[bibtex-entry] [pdf]
-
Louis-Noël Pouchet, Cédric
Bastoul, John Cavazos and Albert Cohen.
A Note on the Performance Distribution of Affine Schedules.
In 2nd Workshop on Statistical and Machine learning approaches to ARchitectures and compilaTion (SMART'08).
Gotenborg, Sweden, January 2008.
[bibtex-entry] [pdf] [slides]
-
Louis-Noël Pouchet, Cédric
Bastoul and Albert Cohen.
Iterative Optimization in the Polyhedral
Model: the One-dimensional Affine Scheduling Case.
In 2nd HiPEAC
Industrial Workshop.
Eindhoven, the Netherlands, October 2006.
[bibtex-entry] [pdf] [slides]
Miscellaneous
-
Louis-Noël Pouchet, Uday
Bondhugula, Cédric Bastoul, Albert Cohen, J. Ramanujam and
P. Sadayappan.
Hybrid Iterative and Model-Driven Optimization in the
Polyhedral Model.
In INRIA Research Report 6962.
June
2009.
[bibtex-entry] [pdf]
-
Mohamed-Walid Benabderrahmane,
Cédric Bastoul, Louis-Noël Pouchet and Albert Cohen.
A
Conservative Approach to Handle Full Functions in the Polyhedral
Model.
In INRIA Research Report 6814.
January 2009.
[bibtex-entry] [pdf]
-
Louis-Noël Pouchet, Cédric
Bastoul and Albert Cohen.
LeTSeE: the LEgal Transformation SpacE
Explorator.
In Third International Summer School on Advanced
Computer Architecture and Compilation for Embedded Systems.
pp
247--251, L'Aquila, Italia, July
2007. Extended abstract.
[bibtex-entry] [pdf] [poster]
-
Louis-Noël Pouchet.
An Interpreter for Vaucanson.
In EPITA
Research and Development Laboratory Seminar.
Le Kremlin-Bicetre,
France, June 2004.
[bibtex-entry] [pdf] [slides]
Thesis
-
Louis-Noël Pouchet.
Iterative Optimization in the Polyhedral Model.
Ph.D (University of Paris-Sud XI).
Orsay,
France, January 2010.
[bibtex-entry] [pdf] [slides]
-
Louis-Noël Pouchet.
When
Iterative Optimization Meets the Polyhedral Model: One-dimensional
Date.
Master thesis (University of Paris-Sud XI).
Orsay,
France, October 2006.
[bibtex-entry] [pdf] [slides]
Copyright notice This material is presented to ensure timely dissemination
of scholarly and technical work. Copyright and all rights therein are
retained by authors or by other copyright holders. All persons copying
this information are expected to adhere to the terms and constraints
invoked by each author's copyright. In most cases, these works may not
be reposted without the explicit permission of the copyright
holder.