Data

Tools

Indexes

Applications

Experiments

Open Source Science

An Open-Source Tool for Generating Domain-Specific Accelerators for Resource-Constrained Computing

An Open-Source Tool for Generating Domain-Specific Accelerators for Resource-Constrained Computing - Published in JOSS (2024)

https://github.com/dtkerns/d2

Science Score: 95.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ✓
    DOI references
    Found 4 DOI reference(s) in README and JOSS metadata
  • ✓
    Academic publication links
    Links to: joss.theoj.org
  • ✓
    Committers with academic emails
    2 of 3 committers (66.7%) from academic institutions
  • ○
    Institutional organization owner
  • ✓
    JOSS paper metadata
    Published in Journal of Open Source Software
Last synced: 6 months ago · JSON representation

Repository

OSS implementation of DOSAGE

Basic Info
  • Host: GitHub
  • Owner: dtkerns
  • License: mit
  • Language: C++
  • Default Branch: main
  • Size: 2.9 MB
Statistics
  • Stars: 1
  • Watchers: 2
  • Forks: 0
  • Open Issues: 1
  • Releases: 2
Created almost 3 years ago · Last pushed over 1 year ago
Metadata Files
Readme License

readme.md

Statement of need

D2 enables computer architects and researchers to easily generate domain-specific accelerators (DSAs) given a set of input workloads based on the super block (SB) granularity. To the best of our knowledge, no such automation tool currently exists. D2 is a shell script “driver” and expects to be run from the top directory of the user’s source code to be analyzed. It takes one argument, the working directory for D2 to store its intermediate data.

Installation Instructions

D2 is intended run on a Unix/Linux system. To build D2, cd to the src directory and type make, then add the D2 bin directory to your PATH

Dependencies

D2 requires an LLVM compiler (e.g. clang) and opt. (I have only tested with clang) Some LLVM installations install opt as opt-VV where VV is a version number. Please resolve with a symbolic link or modify the d2 script as appropriate. It is not required to switch your application to use clang, it is only used for SB intentification. Other dependencies: awk bash cpio cut find grep python3 sed sort and a c++ compiler, either g++ or clang++

Example usage

  • modify your project Makefiles to generate .ll and .llg files (see below)
  • run the D2 software $ PATH=$D2BASE/bin:$PATH $ cd $MYPRJBASE $ d2 /tmp/myproj
  • results are found in the file sbselected in the output directory
  • if you modify the constraints file, you can re-run without remaking the .ll/.llg files with the -m option

Community guidelines

To contribute to D2, please submit your request on github or contact the authors/maintainers

Functionality documentation

see step-by-step.md

Benchmarks

Quantitative analysis is available as an appendix. see paper/appendix.md

DOI

Owner

  • Name: dtkerns
  • Login: dtkerns
  • Kind: user
  • Location: Arizona

nothing interesting here...

JOSS Publication

An Open-Source Tool for Generating Domain-Specific Accelerators for Resource-Constrained Computing
Published
October 13, 2024
DOI
Volume 9, Issue 102, Page 6825
Authors
David T. Kerns
Department of Electrical & Computer Engineering, University of Arizona, Tucson, Arizona, USA
Tosiron Adegbija
Department of Electrical & Computer Engineering, University of Arizona, Tucson, Arizona, USA
Editor
George K. Thiruvathukal ORCID
Tags
Domain-specific accelerators computer-aided design resource-constrained systems energy efficient
View PDF Review Thread Software Archive

GitHub Events

Total
Last Year

Committers

Last synced: 7 months ago

All Time
  • Total Commits: 148
  • Total Committers: 3
  • Avg Commits per committer: 49.333
  • Development Distribution Score (DDS): 0.061
Past Year
  • Commits: 14
  • Committers: 2
  • Avg Commits per committer: 7.0
  • Development Distribution Score (DDS): 0.071
Top Committers
Name Email Commits
dtkerns d****s@a****u 139
tosiron 1****n 8
Daniel S. Katz d****z@i****g 1
Committer Domains (Top 20 + Academic)
ieee.org: 1 arizona.edu: 1

Issues and Pull Requests

Last synced: 6 months ago

All Time
  • Total issues: 7
  • Total pull requests: 3
  • Average time to close issues: about 2 months
  • Average time to close pull requests: 31 minutes
  • Total issue authors: 3
  • Total pull request authors: 2
  • Average comments per issue: 5.14
  • Average comments per pull request: 0.67
  • Merged pull requests: 2
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 1
  • Pull requests: 1
  • Average time to close issues: about 2 hours
  • Average time to close pull requests: about 2 hours
  • Issue authors: 1
  • Pull request authors: 1
  • Average comments per issue: 9.0
  • Average comments per pull request: 1.0
  • Merged pull requests: 1
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • abhishektiwari (3)
  • manuel-g-castro (3)
  • dtkerns (1)
Pull Request Authors
  • danielskatz (2)
  • dtkerns (1)
Top Labels
Issue Labels
Pull Request Labels

Dependencies

.github/workflows/draft-pdf.yml actions
  • actions/checkout v4 composite
  • actions/upload-artifact v1 composite
  • openjournals/openjournals-draft-action master composite