High Order Stencil Transformation Suite

From HPCRL Wiki
(Difference between revisions)
Jump to: navigation, search
(Overview: based on abstract)
(Downloads: temporary gdrive link for polyhedral code)
 
(One intermediate revision by one user not shown)
Line 20: Line 20:
  
 
'''A Framework for Enhancing Data Reuse via Associative Reordering'''
 
'''A Framework for Enhancing Data Reuse via Associative Reordering'''
<!--[LINK PDF] <br /> -->
+
[http://www.cse.ohio-state.edu/~stockk/pldi14.pdf PDF] <br />
 
Kevin Stock, Martin Kong, Tobias Grosser, Louis-Noël Pouchet, Fabrice Rastello, J. Ramanujam, and  P. Sadayappan <br/>
 
Kevin Stock, Martin Kong, Tobias Grosser, Louis-Noël Pouchet, Fabrice Rastello, J. Ramanujam, and  P. Sadayappan <br/>
 
ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'14), June 2014.
 
ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'14), June 2014.
Line 27: Line 27:
  
 
The recent version of the code can be downloaded from the following location: <br />
 
The recent version of the code can be downloaded from the following location: <br />
* [http://www.cse.ohio-state.edu/~stockk/fixme.zip Polyhedral Retiming]
+
* [https://drive.google.com/file/d/0BxZpzD9-0H5bRE5PeGVDbTFNc0E/edit?usp=sharing Polyhedral Retiming]
 
* [http://www.cse.ohio-state.edu/~stockk/stencil_gen.zip Stencil Prototype]
 
* [http://www.cse.ohio-state.edu/~stockk/stencil_gen.zip Stencil Prototype]
 
Please see the included README file for installation instructions. For questions or comments, contact [mailto:stockk@cse.ohio-state.edu Kevin Stock].
 
Please see the included README file for installation instructions. For questions or comments, contact [mailto:stockk@cse.ohio-state.edu Kevin Stock].

Latest revision as of 17:10, 7 June 2014

Contents

Overview

These tools utilize associativity and commutativity of operations in regular loop computations to enhance stencil computations. By aggressively reordering accumulations to the output of a stencil using multi-dimensional retiming, better register reuse is achieved and the total load/store operations required per pixel is significantly reduced.

People

Publications

A Framework for Enhancing Data Reuse via Associative Reordering PDF
Kevin Stock, Martin Kong, Tobias Grosser, Louis-Noël Pouchet, Fabrice Rastello, J. Ramanujam, and P. Sadayappan
ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'14), June 2014.

Downloads

The recent version of the code can be downloaded from the following location:

Please see the included README file for installation instructions. For questions or comments, contact Kevin Stock.

Acknowledgements

This work was supported in part by the U.S. National Science Foundation through awards 0811457, 0904549, 0926127, 1059417 and 1321147, by the U.S. Department of Energy through award DE-SC0008844, and a Google European Fellowship in Efficient Computing.

Personal tools