Collective Mind: Towards Practical and Collaborative Auto-Tuning

Autor: Davide Del Vento, Marc Baboulin, Michael Gerndt, Diego Novillo, Allen D. Malony, Renato Miceli, Zbigniew Chamski, Anton Lokhmotov, Grigori Fursin
Přispěvatelé: Performance Optimization by Software Transformation and Algorithms & Librairies Enhancement (POSTALE), Laboratoire de Recherche en Informatique (LRI), Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Sud - Paris 11 (UP11)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), cTuning foundation, Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES), ICHEC, ARM Research [Cambridge], Technische Universität Munchen - Université Technique de Munich [Munich, Allemagne] (TUM), Computer Science Department [Oregon], University of Oregon [Eugene], Infrasoft IT Solutions, Google Inc, Research at Google, National Center for Atmospheric Research [Boulder] (NCAR), Université de Rennes (UR)
Rok vydání: 2014
Předmět:
code and data sharing
model driven optimization
Computer science
performance tracking buildbot
systematic auto-tuning
Multi-objective optimization
feature selection
Software
[MATH.MATH-ST]Mathematics [math]/Statistics [math.ST]
Computer software
systematic benchmarking
agile development
modeling of computer behavior
performance regression buildbot
collaborative knowledge management
plugin-based auto-tuning
Collective intelligence
collective intelligence
open access publication model
Benchmarking
Predictive analytics
performance prediction
Computer Science Applications
predictive analytics
machine learning
[INFO.INFO-ES]Computer Science [cs]/Embedded Systems
The Internet
[INFO.INFO-DC]Computer Science [cs]/Distributed
Parallel
and Cluster Computing [cs.DC]
Agile software development
NoSQL repository
reproducible research
QA76.75-76.765
[INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG]
[INFO.INFO-DL]Computer Science [cs]/Digital Libraries [cs.DL]
big data driven optimization
collaborative experimentation
business.industry
data mining
[STAT.TH]Statistics [stat]/Statistics Theory [stat.TH]
Modular design
[INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation
Data science
performance tuning buildbot
[INFO.INFO-PF]Computer Science [cs]/Performance [cs.PF]
multi-objective optimization
specification sharing
[INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR]
High performance computing
business
Zdroj: Scientific Programming
Scientific Programming, IOS Press, 2014, Automatic Application Tuning for HPC Architectures, 22 (4), pp.309-329. ⟨10.3233/SPR-140396⟩
Scientific Programming, Vol 22, Iss 4, Pp 309-329 (2014)
Scientific Programming, 2014, Automatic Application Tuning for HPC Architectures, 22 (4), pp.309-329. ⟨10.3233/SPR-140396⟩
ISSN: 1875-919X
1058-9244
Popis: Empirical auto-tuning and machine learning techniques have been showing high potential to improve execution time, power consumption, code size, reliability and other important metrics of various applications for more than two decades. However, they are still far from widespread production use due to lack of native support for auto-tuning in an ever changing and complex software and hardware stack, large and multi-dimensional optimization spaces, excessively long exploration times, and lack of unified mechanisms for preserving and sharing of optimization knowledge and research material. We present a possible collaborative approach to solve above problems using Collective Mind knowledge management system. In contrast with previous cTuning framework, this modular infrastructure allows to preserve and share through the Internet the whole auto-tuning setups with all related artifacts and their software and hardware dependencies besides just performance data. It also allows to gradually structure, systematize and describe all available research material including tools, benchmarks, data sets, search strategies and machine learning models. Researchers can take advantage of shared components and data with extensible meta-description to quickly and collaboratively validate and improve existing auto-tuning and benchmarking techniques or prototype new ones. The community can now gradually learn and improve complex behavior of all existing computer systems while exposing behavior anomalies or model mispredictions to an interdisciplinary community in a reproducible way for further analysis. We present several practical, collaborative and model-driven auto-tuning scenarios. We also decided to release all material atc-mind.org/repoto set up an example for a collaborative and reproducible research as well as our new publication model in computer engineering where experimental results are continuously shared and validated by the community.
Databáze: OpenAIRE
Externí odkaz: