2018 Ninth International Green and Sustainable Computing Conference (IGSC) 2018
DOI: 10.1109/igcc.2018.8752150
|View full text |Cite
|
Sign up to set email alerts
|

Holistic Approaches to HPC Power and Workflow Management

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2021
2021
2021
2021

Publication Types

Select...
1

Relationship

1
0

Authors

Journals

citations
Cited by 1 publication
(1 citation statement)
references
References 16 publications
0
1
0
Order By: Relevance
“…In that work, application-level power usage of a production system was characterized for the Peregrine supercomputer and potential methods for predicting power usage were explored based on a priori and in situ characteristics about application job type. Other related work has shown the effects of different approaches to power management, such as power throttling strategies like core parking, which for certain applications are optimal in minimizing performance losses while maximizing power savings leading to further reductions in PUE/ERE (Purkayastha et al, 2018). Other strategies like frequency reduction have also shown benefits in reducing power consumption without affecting performance for a different class of applications.…”
Section: Application Power Usagementioning
confidence: 99%
“…In that work, application-level power usage of a production system was characterized for the Peregrine supercomputer and potential methods for predicting power usage were explored based on a priori and in situ characteristics about application job type. Other related work has shown the effects of different approaches to power management, such as power throttling strategies like core parking, which for certain applications are optimal in minimizing performance losses while maximizing power savings leading to further reductions in PUE/ERE (Purkayastha et al, 2018). Other strategies like frequency reduction have also shown benefits in reducing power consumption without affecting performance for a different class of applications.…”
Section: Application Power Usagementioning
confidence: 99%