Non quia difficilia sunt non audemus, sed quia non audemus difficilia sunt
Home -> Publications
Home
  Publications
    
all years
    2020
    2019
    2018
    2017
    2016
    2015
    2014
    2013
    2012
    2011
    2010
    2009
    2008
    2007
    2006
    2005
    2004
    theses
    techreports
    presentations
    conferences
    edited volumes
  Awards
  Research
  Teaching
  BLOG
  Miscellaneous
  Full CV [pdf]






  Events








  Past Events





Publications of Torsten Hoefler
Copyright Notice:

The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Tobias Gysi, Tobias Grosser, Torsten Hoefler:

 Absinthe: Learning an Analytical Performance Model to Fuse and Tile Stencil Codes in One Shot

(In Proceedings of the 28th International Conference on Parallel Architectures and Compilation Techniques (PACT), presented in Seattle, WA, USA, IEEE, Sep. 2019)

Abstract

Expensive data movement makes the optimal targetspecific selection of data-locality transformations essential. Loop fusion and tiling are the most important data-locality transformations. Their optimal selection is hard since good tile size choices inherently depend on the fusion choices and vice versa. Existing approaches avoid this difficulty by optimizing independent analytical models or by reverting to heuristics. Absinthe formulates the first unified linear optimization problem to derive single shot fusion and tile size decisions for stencil codes. At the core of our optimization problem, we place a learned analytic performance model that captures the characteristics of the target system. The tuned application kernels demonstrate excellent performance within 10% of exhaustively auto-tuned versions and up to 74% faster than the results of independent optimization with max fusion heuristic and Absinthe tile size selection. While the full search space is non-linear, bounding it to relevant solutions enables the efficient exploration of the exponential search space using linear solvers. As a result, the tuning of our application kernels takes less than one minute. Our approach thus establishes the foundations for next-generation compilers, which exploit empirical information to guide targetspecific code transformations.

Documents

download article:
download slides:
 

BibTeX

@inproceedings{,
  author={Tobias Gysi and Tobias Grosser and Torsten Hoefler},
  title={{Absinthe: Learning an Analytical Performance Model to Fuse and Tile Stencil Codes in One Shot}},
  year={2019},
  month={Sep.},
  booktitle={Proceedings of the 28th International Conference on Parallel Architectures and Compilation Techniques (PACT)},
  location={Seattle, WA, USA},
  publisher={IEEE},
  source={http://www.unixer.de/~htor/publications/},
}

serving: 18.205.60.226:36454© Torsten Hoefler