DDGVec

From HPCRL Wiki
(Difference between revisions)
Jump to: navigation, search
(Download)
 
Line 30: Line 30:
 
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/~elangov/download/DDGVec_1_1.tar.gz DDGVec v1.1] <br />
 
* [http://www.cse.ohio-state.edu/~elangov/download/DDGVec_1_1.tar.gz DDGVec v1.1] <br />
Please see the included README file for installation instructions. For questions or comments, contact Venmugil Elango (elango [dot] 4 [at] buckeyemail [dot] osu [dot] edu).
+
Please see the included README file for installation instructions. For questions or comments, contact Venmugil Elango (elango [dot] 4 [at] osu [dot] edu).
  
 
== Acknowledgements ==
 
== Acknowledgements ==

Latest revision as of 22:28, 17 July 2014

Contents

Overview

DDGVec is a dynamic trace-based analysis tool used to detect potential SIMD parallelism in the programs that may otherwise be missed by the conservative compile-time analyses. The tool takes an automatic approach to characterize the inherent vectorizability potential of existing applications by analyzing information about run-time dependences and memory access patterns.

People

Faculty

Students

Publications

Dynamic Trace-Based Analysis of Vectorization Potential of Applications PDF
Justin Holewinski, Ragavendar Ramamurthi, Mahesh Ravishankan, Naznin Fauzia, Louis-Noel Pouchet, Atanas Rountev, and P. Sadayappan.
ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI'12), June 2012.

Download

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 Venmugil Elango (elango [dot] 4 [at] osu [dot] edu).

Acknowledgements

This software is based upon work supported by the National Science Foundation under grants CCF-0811781, CCF-0926127, OCI-0904549, CCF-1017204, and by the Department of Energy's Office of Advanced Scientific Computing under grant DE-SC0005033. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, the Department of Energy, or The Ohio State University.

Personal tools