Grigori Fursin

Last updated

Grigori Fursin
Grigori Fursin (2019).jpg
Alma mater
Known for MILEPOST GCC, cTuning foundation, Collective Knowledge framework, Collective Mind, Artifact Evaluation at ACM and IEEE conferences
Awards
  • Test of Time Award (IEEE/ACM International Symposium on Code Generation and Optimization, 2017) [1]
Scientific career
Fields Computer engineering
Machine learning
Institutions
Thesis Iterative Compilation and Performance Prediction for Numerical Applications  (2004;20 years ago (2004))
Website fursin.net

Grigori Fursin is a British [2] computer scientist, president of the non-profit CTuning foundation, founding member of MLCommons, [3] and co-chair of the MLCommons Task Force on Automation and Reproducibility. [4] His research group created open-source machine learning based self-optimizing compiler, MILEPOST GCC, considered to be the first in the world. [5] At the end of the MILEPOST project he established cTuning foundation to crowdsource program optimisation and machine learning across diverse devices provided by volunteers. His foundation also developed Collective Knowledge Framework and Collective Mind [6] to support open research. Since 2015 Fursin leads Artifact Evaluation at several ACM and IEEE computer systems conferences. He is also a founding member of the ACM taskforce on Data, Software, and Reproducibility in Publication. [7] [8] [9]

Contents

Education

Fursin completed his PhD in computer science at the University of Edinburgh in 2005. While in Edinburgh, he worked on foundations of practical program autotuning and performance prediction. [10]

Notable projects

Related Research Articles

In computing, just-in-time (JIT) compilation is compilation during execution of a program rather than before execution. This may consist of source code translation but is more commonly bytecode translation to machine code, which is then executed directly. A system implementing a JIT compiler typically continuously analyses the code being executed and identifies parts of the code where the speedup gained from compilation or recompilation would outweigh the overhead of compiling that code.

The Programming Language Design and Implementation (PLDI) conference is an annual computer science conference organized by the Association for Computing Machinery (ACM) which focuses on the study of algorithms, programming languages and compilers. It is sponsored by the SIGPLAN special interest group on programming languages.

In software engineering, profiling is a form of dynamic program analysis that measures, for example, the space (memory) or time complexity of a program, the usage of particular instructions, or the frequency and duration of function calls. Most commonly, profiling information serves to aid program optimization, and more specifically, performance engineering.

<span class="mw-page-title-main">Robert Harper (computer scientist)</span> Computer scientist

Robert William Harper, Jr. is a computer science professor at Carnegie Mellon University who works in programming language research. Prior to his position at Carnegie Mellon, Harper was a research fellow at the University of Edinburgh.

Superoptimization is the process where a compiler automatically finds the optimal sequence for a loop-free sequence of instructions. Real-world compilers generally cannot produce genuinely optimal code, and while most standard compiler optimizations only improve code partly, a superoptimizer's goal is to find the optimal sequence, the canonical form. Superoptimizers can be used to improve conventional optimizers by highlighting missed opportunities so a human can write additional rules.

Thread Level Speculation (TLS), also known as Speculative Multi-threading, or Speculative Parallelization, is a technique to speculatively execute a section of computer code that is anticipated to be executed later in parallel with the normal execution on a separate independent thread. Such a speculative thread may need to make assumptions about the values of input variables. If these prove to be invalid, then the portions of the speculative thread that rely on these input variables will need to be discarded and squashed. If the assumptions are correct the program can complete in a shorter time provided the thread was able to be scheduled efficiently.

Matthew Flatt is an American computer scientist and professor at the University of Utah School of Computing in Salt Lake City. He is also the leader of the core development team for the Racket programming language.

In computing, compiler correctness is the branch of computer science that deals with trying to show that a compiler behaves according to its language specification. Techniques include developing the compiler using formal methods and using rigorous testing on an existing compiler.

The Interactive Compilation Interface (ICI) is a plugin system with a high-level compiler-independent and low-level compiler-dependent API to transform production compilers into interactive research toolsets. It was developed by Grigori Fursin during the MILEPOST project. The ICI framework acts as a "middleware" interface between the compiler and the user-definable plugins. It opens up and reuses the production-quality compiler infrastructure to enable program analysis and instrumentation, fine-grain program optimizations, simple prototyping of new development and research ideas while avoiding building new compilation tools from scratch. For example, it is used in MILEPOST GCC to automate compiler and architecture design and program optimizations based on statistical analysis and machine learning, and predict profitable optimization to improve program execution time, code size and compilation time.

MILEPOST GCC is a free, community-driven, open-source, adaptive, self-tuning compiler that combines stable production-quality GCC, Interactive Compilation Interface and machine learning plugins to adapt to any given architecture and program automatically and predict profitable optimizations to improve program execution time, code size and compilation time. It is currently used and supported by academia and industry and is intended to open up research opportunities to automate compiler and architecture design and optimization.

The Collective Tuning Initiative is a community-driven initiative started by Grigori Fursin to develop free and open-source research tools with a unified API for collaborative characterization, optimization and co-design of computer systems. They enable sharing of benchmarks, data sets and optimization cases from the community in the Collective Optimization Database through unified web services to predict better optimizations or architecture designs. Using common research-and-development tools should help to improve the quality and reproducibility of computer systems' research and development and accelerate innovation in this area. This approach helped establish Reproducibility Initiatives and Artifact Evaluation at several ACM-sponsored conferences to encourage sharing of artifacts and validation of experimental results from accepted papers.

<span class="mw-page-title-main">Kathryn S. McKinley</span> American computer scientist

Kathryn S. McKinley is an American computer scientist noted for her research on compilers, runtime systems, and computer architecture. She is also known for her leadership in broadening participation in computing. McKinley was co-chair of CRA-W from 2011 to 2014.

Susan Beth Horwitz was an American computer scientist noted for her research on programming languages and software engineering, and in particular on program slicing and dataflow-analysis. She had several best paper and an impact paper award mentioned below under awards.

The Collective Knowledge (CK) project is an open-source framework and repository to enable collaborative, reproducible and sustainable research and development of complex computational systems. CK is a small, portable, customizable and decentralized infrastructure helping researchers and practitioners:

David Bacon is an American computer programmer.

The cTuning Foundation is a global non-profit organization developing a common methodology and open-source tools to support sustainable, collaborative and reproducible research in Computer science and organize and automate artifact evaluation and reproducibility inititiaves at machine learning and systems conferences and journals.

Ilya Sergey is a Russian computer scientist and an Associate Professor at the School of Computing at the National University of Singapore, where he leads the Verified Systems Engineering lab. Sergey does research in programming language design and implementation, software verification, distributed systems, program synthesis, and program repair. He is known for designing the Scilla programming language for smart contracts. He is the author of the free online book Programs and Proofs: Mechanizing Mathematics with Dependent Types, Lecture notes with exercises, which introduce the basic concepts of mechanized reasoning and interactive theorem proving using Coq.

A sea of nodes is a graph representation of single-static assignment (SSA) representation of a program that combines data flow and control flow, and relaxes the control flow from a total order to a partial order, keeping only the orderings required by data flow. It is similar to a value dependency graph (VDG).

References

  1. HiPEAC info 50 (page 8), April 2017
  2. Companies House profile, June 2015
  3. MLCommons press-release, December 2020
  4. MLCommons Task Force on Automation and Reproducibility, June 2022
  5. World's First Intelligent, Open Source Compiler Provides Automated Advice on Software Code Optimization, IBM press-release, June 2009 (link)
  6. 1 2 Fursin, Grigori (June 2024). "Enabling more efficient and cost-effective AI/ML systems with Collective Mind, virtualized MLOps, MLPerf, Collective Knowledge Playground and reproducible optimization tournaments". arXiv: 2406.16791 [cs.LG].
  7. "The ACM Task Force on Data, Software, and Reproducibility in Publication" . Retrieved 5 December 2017.
  8. Fursin, Grigori; Bruce Childers; Alex K. Jones; Daniel Mosse (June 2014). TRUST'14. Proceedings of the 1st ACM SIGPLAN Workshop on Reproducible Research Methodologies and New Publication Models in Computer Engineering at PLDI'14. doi:10.1145/2618137.
  9. "ACM TechTalk "Reproducing 150 Research Papers and Testing Them in the Real World: Challenges and Solutions with Grigori Fursin"" . Retrieved 11 February 2021.
  10. Grigori Fursin (July 2004). "PhD thesis" . Retrieved 21 May 2017.
  11. Fursin, Grigori (June 2023). Toward a common language to facilitate reproducible research and technology transfer: challenges and solutions. keynote at the 1st ACM Conference on Reproducibility and Replicability. doi:10.5281/zenodo.8105339.
  12. Online catalog of automation recipes developed by MLCommons
  13. HPCWire: MLPerf Releases Latest Inference Results and New Storage Benchmark, September 2023
  14. Fursin, Grigori (October 2020). Collective Knowledge: organizing research projects as a database of reusable components and portable workflows with common interfaces. Philosophical Transactions of the Royal_Society. arXiv: 2011.01149 . doi:10.1098/rsta.2020.0211 . Retrieved 22 October 2020.
  15. Fursin, Grigori; Bruce Childers; Alex K. Jones; Daniel Mosse (June 2014). TRUST'14. Proceedings of the 1st ACM SIGPLAN Workshop on Reproducible Research Methodologies and New Publication Models in Computer Engineering at PLDI'14. doi:10.1145/2618137.
  16. Fursin, Grigori; Christophe Dubach (June 2014). Community-driven reviewing and validation of publications. Proceedings of TRUST'14 at PLDI'14. arXiv: 1406.4020 . doi:10.1145/2618137.2618142.
  17. Childers, Bruce R; Grigori Fursin; Shriram Krishnamurthi; Andreas Zeller (March 2016). Artifact evaluation for publications. Dagstuhl Perspectives Workshop 15452. doi: 10.4230/DagRep.5.11.29 .