Parallel computing, benchmarking and ATLAS software on ARM
Master Thesis
2015
Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Department
Faculty
License
Series
Abstract
This thesis explores the use of the ARM architecture in high energy physics computing. ARM processors are predominantly found in smartphones and mobile tablets. Results from benchmarks which were performed on the armv7l architecture are presented. These provide qualitative data as well as confirmation that specialized high energy physics software does run on ARM. This thesis presents the first ever port of the ATLAS software stack to the ARM architecture, as well as the issues that ensued. A new framework, ANA, is introduced which facilitates the compilation of the ATLAS software stack on ARM.
Description
Includes bibliographical references,
Keywords
Reference:
Smith, J. 2015. Parallel computing, benchmarking and ATLAS software on ARM. University of Cape Town.