An approach for code generation in the Sparse Polyhedral Framework
Name:
Strout-PARCO-March-2016.pdf
Size:
979.8Kb
Format:
PDF
Description:
Final Accepted Manuscript
Author
Strout, Michelle MillsLaMielle, Alan

Carter, Larry
Ferrante, Jeanne
Kreaseck, Barbara
Olschanowsky, Catherine
Affiliation
Computer Science Department, University of ArizonaIssue Date
2016-04
Metadata
Show full item recordPublisher
ELSEVIER SCIENCE BVCitation
An approach for code generation in the Sparse Polyhedral Framework 2016, 53:32 Parallel ComputingJournal
Parallel ComputingRights
Copyright © 2016 Elsevier B.V. All rights reserved.Collection Information
This item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at repository@u.library.arizona.edu.Abstract
Applications that manipulate sparse data structures contain memory reference patterns that are unknown at compile time due to indirect accesses such as A[B[i]]. To exploit parallelism and improve locality in such applications, prior work has developed a number of Run-Time Reordering Transformations (RTRTs). This paper presents the Sparse Polyhedral Framework (SPF) for specifying RTRTs and compositions thereof and algorithms for automatically generating efficient inspector and executor code to implement such transformations. Experimental results indicate that the performance of automatically generated inspectors and executors competes with the performance of hand-written ones when further optimization is done.Note
Available online 4 March 2016. 24 month embargo.ISSN
0167-8191Version
Final accepted manuscriptSponsors
We thank Jon Roelofs for his implementation of the IEGenCC tool, which converts C programs into the specification format IEGen expects as input. We thank Christopher Krieger, Andrew Stone, Tomofumi Yuki, and anonymous reviewers for their careful reading and suggestions. This work was sponsored by NSF CAREER Grant CCF-0746693, DOE Early Career Grant DE-SC3956, the CSCAPES Institute DOE Grant 7F-00323, and the CACHE project DOE Grant DE-SC04030.Additional Links
http://linkinghub.elsevier.com/retrieve/pii/S0167819116000557ae974a485f413a2113503eed53cd6c53
10.1016/j.parco.2016.02.004