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

Title: Understanding the Impact of Dynamic Power Capping on Application Progress

Abstract

Electrical power has become an important design constraint in high-performance computing (HPC) systems. On future HPC machines, power is likely to be a budgeted resource and thus managed dynamically. Power management software needs to reliably measure application performance at runtime in order to respond effectively to changes in application behavior. Execution time tells us little about how the science in the application is progressing toward an application-defined end goal. To the best of our knowledge, no study has defined or categorized online application progress in the context of power management. Based on semi-structured interviews with HPC application-specialists, we define an online notion of progress—an application-specific metric that can be monitored at runtime to provide a sense of the rate at which application science is being performed. Using instrumentation, we characterize and categorize the progress of various production scientific applications and benchmarks. We propose a model of the impact of dynamic power capping on application progress. By experimental evaluation, we show that our model accurately captures the general behavior of the progress of different classes of applications under a power cap. We believe that such a model is an important first step toward the design of more dynamic power management policiesmore » for HPC systems.« less

Authors:
; ; ; ;
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC); USDOE National Nuclear Security Administration (NNSA); USDOE Office of Science - Office of Advanced Scientific Computing Research
OSTI Identifier:
1515188
DOE Contract Number:  
AC02-06CH11357
Resource Type:
Conference
Resource Relation:
Conference: 33rd IEEE International Parallel and Distributed Processing Symposium, 05/20/19 - 05/24/19, Rio de Janeiro, BR
Country of Publication:
United States
Language:
English
Subject:
application progress; dynamic power management; online performance characterization; power capping

Citation Formats

Ramesh, Srinivasan, Perarnau, Swann, Bhalachandra, Sridutt, Malony, Allen D., and Beckman, Pete. Understanding the Impact of Dynamic Power Capping on Application Progress. United States: N. p., 2019. Web.
Ramesh, Srinivasan, Perarnau, Swann, Bhalachandra, Sridutt, Malony, Allen D., & Beckman, Pete. Understanding the Impact of Dynamic Power Capping on Application Progress. United States.
Ramesh, Srinivasan, Perarnau, Swann, Bhalachandra, Sridutt, Malony, Allen D., and Beckman, Pete. Mon . "Understanding the Impact of Dynamic Power Capping on Application Progress". United States.
@article{osti_1515188,
title = {Understanding the Impact of Dynamic Power Capping on Application Progress},
author = {Ramesh, Srinivasan and Perarnau, Swann and Bhalachandra, Sridutt and Malony, Allen D. and Beckman, Pete},
abstractNote = {Electrical power has become an important design constraint in high-performance computing (HPC) systems. On future HPC machines, power is likely to be a budgeted resource and thus managed dynamically. Power management software needs to reliably measure application performance at runtime in order to respond effectively to changes in application behavior. Execution time tells us little about how the science in the application is progressing toward an application-defined end goal. To the best of our knowledge, no study has defined or categorized online application progress in the context of power management. Based on semi-structured interviews with HPC application-specialists, we define an online notion of progress—an application-specific metric that can be monitored at runtime to provide a sense of the rate at which application science is being performed. Using instrumentation, we characterize and categorize the progress of various production scientific applications and benchmarks. We propose a model of the impact of dynamic power capping on application progress. By experimental evaluation, we show that our model accurately captures the general behavior of the progress of different classes of applications under a power cap. We believe that such a model is an important first step toward the design of more dynamic power management policies for HPC systems.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2019},
month = {5}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: