UNIDAPT Group (UNIversal aDAPTation) pioneers and promotes radically new practical techniques to overcome the complexity of current architectures, compilers, operating systems and programming environments, and to automate and improve architecture and compiler design (particularly for future heterogeneous multi-core systems), optimize and parallelize systems to improve performance, power consumption, size and fault-tolerance, reduce cost and time to market based on machine learning, artificial intelligence, statistical methods and biologically inspired techniques. This is critical to be able to continue innovation in science and industry (bioinformatics, medicine, physics, chemistry, finances, gaming, etc) that demands ever-increasing computing resources while placing strict requirements on systems.

UNIDAPT group is created by Dr. Grigori Fursin and has close collaboration with the groups of Prof.Michael O'Boyle (University of Edinburgh), Prof.Olivier Temam (INRIA) and colleagues from IBM.

We are leading the following research & developments:

• MILEPOST GCC (thanks to the fruitful collaboration with all the MILEPOST partners) - machine learning based research compiler adpatable to different architectures, programs and run-time behavior.
  
  [GCC Summit'08 paper and bib; CGO'06 paper and bib];
  
  
• Continuous Collective Compilation (and Parallelization) Framework - framework to reuse the optimization knowledge among different programs and architectures using static and dynamic (hardware counters and reactions) program features based on MILEPOST GCC and Global Optimization Database .
  
  [GCC Summit'08 paper and bib];
  
  
• Interactive Compilation Interface - interface for GCC and Open64 to enable practical iterative interactive optimizations and make future compilers modular and extensible with automatically tuned optimization heuristic easily adaptable to different architectures and constraints .
  
  [GCC Summit'08 paper and bib];
  
  
• MiDataSets - multiple datasets to enable realistic program and architecture optimizations, run-time adaptation and performance evaluation .
  
  [HiPEAC'07 paper and bib, GCC Summit'08 paper and bib];
  
  
• UNIDAPT Framework - techniques to enable run-time adaptation and adaptive parallelization for statically compiled programs on heterogeneous multi-core architectures .
  
  [HiPEAC'05 paper and bib, GCC Summit'07 paper and bib, NPC'07 paper and bib];
  
  
• Reactions to optimizations (with the University of Edinburgh) - we introduced techniques to reuse knowledge and speedup program optimizations and architecture design space exploration based on reactions to optimizations.
  
  [CASES'06 paper and bib];

More information about UNIDAPT research and software.

If you would like to discuss research, collaboration, visits, internships, grant proposals, funding, consulting services and knowledge transfer, startups, provide sponsorship, etc, please don't hesitate to contact Grigori Fursin (UNIDAPT leader).

We have a fruitful collaboration with both academia and industry:

• INRIA/IRISA, France - long term collaboration with the group of Prof. Olivier Temam
• University of Edinburgh, UK - long term collaboration with the group of Prof. Michael O'Boyle
• Imperial College, UK
• UPC, Spain
• UIUC, USA
• University of Delaware, USA
• Ghent University, Belgium
• ICT, China
• IBM
• ARC, UK
• CAPS Enterprise, France
• AMD, USA
• STMicro
• NXP (Philips), Netherlands
• ARM, UK

We are currently involved in the following projects:

• MILEPOST
• SARC
• OMP
• GGCC
• ACOTES
• HiPEAC Project: Context-aware optimization and run-time adaptation of sequential libraries for multi-core systems
• HiPEAC Clusters: Adaptive Compilation, Multi-core architecture, Compilation platform, Programming models and operating systems, Simulation platform
• various HiPEAC activities
• organizing SMART workshop series

We would like to thank the following colleagues for fruitful research discussions, collaboration, help with workshop organizations, developments or other activities:

Matthew Arnold (IBM), Elton Ashton (ARC), Phil Barnard (ARC), Sandro Bartolini (University of Siena), David Bernstain (IBM), Francois Bodin (CAPS Enterprise), Edwin Bonilla (University of Edinburgh), Calin Cascaval (IBM), John Cavazos (University of Delaware), Zbigniew Chamski (Płock Industrial and Technological Park), Marcelo Cintra (University of Edinburgh), Albert Cohen (INRIA), Marco Cornero (STMicro), Eric Courtois (CAPS Enterprise), Koen De Bosschere (Ghent University), Marc Duranton (NXP), Veerle Desmet (Ghent University), David Edelsohn (IBM), Lieven Eeckhout (Ghent University), Bjoern Franke (University of Edinburgh), Ari Freund (IBM), Marisa Gil (UPC), Sylvian Girbal (INRIA), Kenneth Hoste (Ghent University), Wen-Mei Hwu (UIUC), Tim Jones (University of Edinburgh), David Kaeli (Northeastern University), Paul Kelly (Imperial College), Hugh Leather (University of Edinburgh), Anton Lokhmotov (Imperial College), Shun Long (Jinan University), Sally McKee (Cornell University), Bilha Mendelson (IBM), Eliot Moss (University of Massachusetts), Nacho Navarro (UPC), Diego Novillo (Google), Michael O'Boyle (University of Edinburgh), David Padua (UIUC), Basile Starynkevitch (CEA-LIST), Kevin Stoodley (IBM), Devika Subramanian (Rice University), Olivier Temam (INRIA), John Thomson (University of Edinburgh), Nigel Topham (University of Edinburgh/ARC), Matthew J. Thazhuthaveetil (Indian Institute of Science), Sid Touati (University of Versailles), Nicolas Vasilache (Reservoir Labs), Richard Vuduc (Georgia Tech), Ian Watson (University of Manchester), David Whalley (Florida State University), Chris Williams (University of Edinburgh), Sami Yehia (Thales), Elad Yom-Tov (IBM), Ayal Zaks (IBM) and many others

• We started preparing the 3rd Workshop on Statistical and Machine learning approaches to ARchitectures and compilaTion (SMART'09). CFP will be available soon. Here is the webpage of the previous SMART'08.
  
  
• Here is IBM's press release at CNN about MILEPOST project and MILEPOST GCC.
  
  
• We are preparing the release of the MILEPOST GCC v1.0 , Continuous Collective Compilation (and Parallelization) Framework v1.0 and MiDataSets v1.3a .
  
  
• We presented a poster about MILEPOST GCC (first machine learning based research compiler) and Continuous Collective Compilation (and Parallelization) Framework (to continuously reuse optimization knowledge among different programs and architectures) at the ACACES Summer school in L'Aquila, Italy on the 16th of July, 2008.
  
  
• Information about positions, internships, virtual collaborations and developments in the UNIDAPT Group at INRIA Saclay.
  

If you would like to receive UNIDAPT news about releases of our tools, positions and grants available, etc, immediately by email, please subscribe here.

2008.June.23

We uploaded the CCC Framework and MILEPOST GCC beta pre-release versions to the CVS to be able to continue collaboration with our colleagues. We hope to release the final stable versions in summer 2008. We also released GCC ICI v1.0.

2008.June

Grigori Fursin presented a paper about MILEPOST GCC (first machine learning based research compiler) and Continuous Collective Compilation (and Parallelization) Framework (to continuously reuse optimization knowledge among different programs and architectures) at the GCC Summit in Ottawa, Canada on the 19th of June, 2008.

2008.May

Anton Lokhmotov from Imperial College (UK) gave a talk at INRIA Saclay - "Programming accelerated systems".

2008.April.11-17

We successfully passed MILEPOST review demonstrating MILEPOST GCC with a new program feature extractor and Interactive Compilation Interface to manipulate GCC optimization passes, and a new version of the Continuous Collective Compilation Framework to continuously reuse optimization knowledge among different programs and architectures.

2008.March

Ewdin Bonilla from the University of Edinburgh (UK) gave a talk at INRIA Saclay - "Machine Learning for Compiler Optimization: Principles, Techniques and Challenges".

2008.February

We received HiPEAC funding for the project "Context-aware optimization and run-time adaptation of sequential libraries for multi-core systems".

2008.January.27

SMART'08 workshop co-located with HiPEAC'08 conference took place in Goteborg. Thanks to everyone who participated in this event!

• We would like to thank HiPEAC network of excellence, EU FP7 program and MILEPOST project for generous support of UNIDAPT research activities and software developments!
  
  
• If you would like to support our research and software developments, use our tools in commercial projects, interested in our consulting services, startups, etc, please don't hesitate to contact us.