.png){kind=icon}
.png){kind=icon}
Return
Energy-aware performance analysis methodologies for HPC architectures-An exploratory study
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jnca.2012.08.003.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Performance analysis is a crucial step in HPC architectures including clouds. Traditional performance analysis methodologies that were proposed, implemented, and enacted are functional with the objective of identifying bottlenecks or issues related to memory, programming languages, hardware, and virtualization aspects. However, the need for energy efficient architectures in highly scalable computing environments, such as. Grid or Cloud, has widened the research thrust on developing performance analysis methodologies that analyze the energy inefficiency of HPC applications or their associated hardware. This paper surveys the performance analysis methodologies that investigates into the available energy monitoring and energy awareness mechanisms for HPC architectures. In addition, the paper validates the existing tools in terms of overhead, portability, and user-friendly parameters by conducting experiments at HPCCLoud Research Laboratory at our premise. This research work will promote HPC application developers to select an apt monitoring mechanism and HPC tool developers to augment required energy monitoring mechanisms which fit well with their basic monitoring infrastructures. (C) 2012 Elsevier Ltd. All rights reserved.
Keywords:
HPC
Performance analysis
Tools
Energy monitoring
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
8
Papers:
3.6K
Citations:
1.1W
Organization
No organization information available