skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Roofline: an insightful visual performance model for multicore architectures

Abstract

We propose an easy-to-understand, visual performance model that offers insights to programmers and architects on improving parallel software and hardware for floating point computations.

Authors:
 [1];  [1];  [1]
  1. Univ. of California, Berkeley, CA (United States). Parallel Computing Lab.
Publication Date:
Research Org.:
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1407073
Grant/Contract Number:  
AC02-05CH11231
Resource Type:
Accepted Manuscript
Journal Name:
Communications of the ACM
Additional Journal Information:
Journal Volume: 52; Journal Issue: 4; Journal ID: ISSN 0001-0782
Publisher:
Association for Computing Machinery
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING

Citation Formats

Williams, Samuel, Waterman, Andrew, and Patterson, David. Roofline: an insightful visual performance model for multicore architectures. United States: N. p., 2009. Web. doi:10.1145/1498765.1498785.
Williams, Samuel, Waterman, Andrew, & Patterson, David. Roofline: an insightful visual performance model for multicore architectures. United States. doi:10.1145/1498765.1498785.
Williams, Samuel, Waterman, Andrew, and Patterson, David. Sat . "Roofline: an insightful visual performance model for multicore architectures". United States. doi:10.1145/1498765.1498785. https://www.osti.gov/servlets/purl/1407073.
@article{osti_1407073,
title = {Roofline: an insightful visual performance model for multicore architectures},
author = {Williams, Samuel and Waterman, Andrew and Patterson, David},
abstractNote = {We propose an easy-to-understand, visual performance model that offers insights to programmers and architects on improving parallel software and hardware for floating point computations.},
doi = {10.1145/1498765.1498785},
journal = {Communications of the ACM},
number = 4,
volume = 52,
place = {United States},
year = {2009},
month = {4}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 352 works
Citation information provided by
Web of Science

Save / Share: