License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/DagSemProc.08161.6
URN: urn:nbn:de:0030-drops-15745
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2008/1574/
Go to the corresponding Portal

Gesellensetter, Lars

Scalable Analysis via Machine Learning: Predicting Memory Dependencies Precisely

pdf-format:
08161.GesellensetterLars.ExtAbstract.1574.pdf (0.09 MB)

Abstract

Using Machine Learning to yield Scalable Program Analyses

Program Analysis tackles the problem of predicting the behavior or
certain properties of the considered program code. The challenge lies in
determining the dynamic runtime behavior statically at compile time.
While in rare cases it is possible to determine exact dynamic properties
already statically, in many cases, e.g., in analyzing memory dependencies,
one can only find imprecise information. To overcome this, we apply
Machine Learning (ML) techniques which are particularly suited for this
task. They yield highly scalable predictors and are safely applicable when
erroneous predictions merely have an impact on program optimality but
not on correctness.

In this talk, I present our approach to mitigate the impact of the memory
gap. Over the last decade, computer performance is often dominated
by memory speed, which did not manage to keep pace with the ever
increasing cpu rates. We consider novel speculative optimization
techniques of memory accesses to reduce their effective latency.
We trained predictors to learn the memory dependencies of a given pair
of accesses, and use the result in our optimization do decide about the
profitability of a given optimization step.


BibTeX - Entry

@InProceedings{gesellensetter:DagSemProc.08161.6,
  author =	{Gesellensetter, Lars},
  title =	{{Scalable Analysis via Machine Learning: Predicting Memory Dependencies Precisely}},
  booktitle =	{Scalable Program Analysis},
  pages =	{1--3},
  series =	{Dagstuhl Seminar Proceedings (DagSemProc)},
  ISSN =	{1862-4405},
  year =	{2008},
  volume =	{8161},
  editor =	{Florian Martin and Hanne Riis Nielson and Claudio Riva and Markus Schordan},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2008/1574},
  URN =		{urn:nbn:de:0030-drops-15745},
  doi =		{10.4230/DagSemProc.08161.6},
  annote =	{Keywords: Program Analysis, Alias Analysis, Memory Depdencies, Speculative Optimizations, Machine Learning}
}

Keywords: Program Analysis, Alias Analysis, Memory Depdencies, Speculative Optimizations, Machine Learning
Collection: 08161 - Scalable Program Analysis
Issue Date: 2008
Date of publication: 28.08.2008

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI