Tile Size Selection Using Cache Organization and Data Layout

@inproceedings{Coleman1995TileSS,
  title={Tile Size Selection Using Cache Organization and Data Layout},
  author={Stephanie Coleman and Kathryn S. McKinley},
  booktitle={PLDI},
  year={1995}
}
When dense matrix computations are too large to fit in cache, previous research proposes tiling to reduce or eliminate capacity misses. This paper presents a new algorithm for choosing problem-size dependent tile sizes based on the cache size and cache line size for a direct-mapped cache. The algorithm eliminates both capacity and self-interference misses and reduces cross-interference misses. We measured simulated miss rates and execution times for our algorithm and two others on a variety of… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 323 CITATIONS, ESTIMATED 24% COVERAGE

Compiler Construction

  • Lecture Notes in Computer Science
  • 2012
VIEW 18 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

A Comparison of Compiler Tiling Algorithms

VIEW 10 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Software Methods for Avoiding Cache Conflicts

VIEW 13 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Reactive tiling

  • 2015 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)
  • 2015
VIEW 5 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Defensive loop tiling for shared cache

  • Proceedings of the 2013 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)
  • 2013
VIEW 5 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Optimal Tile Size Selection Problem Using Machine Learning

  • 2012 11th International Conference on Machine Learning and Applications
  • 2012
VIEW 5 EXCERPTS
CITES METHODS & RESULTS
HIGHLY INFLUENCED

A case for a working-set-based memory hierarchy

  • Conf. Computing Frontiers
  • 2005
VIEW 11 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Reduction transformations for optimization parameter selection

  • Eighth International Conference on High-Performance Computing in Asia-Pacific Region (HPCASIA'05)
  • 2005
VIEW 5 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Languages and Compilers for Parallel Computing

  • Lecture Notes in Computer Science
  • 2001
VIEW 11 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

1996
2018

CITATION STATISTICS

  • 53 Highly Influenced Citations

  • Averaged 7 Citations per year over the last 3 years

Similar Papers

Loading similar papers…