Whittle Index Approach to Energy-Aware Dispatching
S. Aalto, and P. Lassila. 30th International Teletraffic Congress (ITC 30), Vienna, Austria, (2018)
Abstract
A data center can be modeled as a set of parallel queues, and the dispatcher decides to which queue the arriving jobs are routed. We consider an energy-aware dispatching system in a Markovian setting, where each server upon becoming empty enters a sleep mode to save energy and to activate the server after sleep incurs an additional setup delay cost. We seek to optimize the performance-energy trade-off by applying the so-called Whittle index approach. As our main result, we rigorously prove, under a certain technical assumption, that the problem is indexable, and derive the explicit form of the Whittle index. Our numerical examples demonstrate that the resulting energy-aware Whittle index policy is able to perform very close to the numerically obtained optimal policy.
%0 Conference Paper
%1 Aalto18ITC30
%A Aalto, Samuli
%A Lassila, Pasi
%B 30th International Teletraffic Congress (ITC 30)
%C Vienna, Austria
%D 2018
%K Session_1:_Performance_Evaluation itc itc30
%T Whittle Index Approach to Energy-Aware Dispatching
%U https://gitlab2.informatik.uni-wuerzburg.de/itc-conference/itc-conference-public/-/raw/master/itc30/Aalto18ITC30.pdf?inline=true
%X A data center can be modeled as a set of parallel queues, and the dispatcher decides to which queue the arriving jobs are routed. We consider an energy-aware dispatching system in a Markovian setting, where each server upon becoming empty enters a sleep mode to save energy and to activate the server after sleep incurs an additional setup delay cost. We seek to optimize the performance-energy trade-off by applying the so-called Whittle index approach. As our main result, we rigorously prove, under a certain technical assumption, that the problem is indexable, and derive the explicit form of the Whittle index. Our numerical examples demonstrate that the resulting energy-aware Whittle index policy is able to perform very close to the numerically obtained optimal policy.
@inproceedings{Aalto18ITC30,
abstract = {A data center can be modeled as a set of parallel queues, and the dispatcher decides to which queue the arriving jobs are routed. We consider an energy-aware dispatching system in a Markovian setting, where each server upon becoming empty enters a sleep mode to save energy and to activate the server after sleep incurs an additional setup delay cost. We seek to optimize the performance-energy trade-off by applying the so-called Whittle index approach. As our main result, we rigorously prove, under a certain technical assumption, that the problem is indexable, and derive the explicit form of the Whittle index. Our numerical examples demonstrate that the resulting energy-aware Whittle index policy is able to perform very close to the numerically obtained optimal policy.},
added-at = {2018-09-12T17:41:00.000+0200},
address = {Vienna, Austria},
author = {Aalto, Samuli and Lassila, Pasi},
biburl = {https://www.bibsonomy.org/bibtex/2bd74c7e45a441812378342f8949d0682/itc},
booktitle = {30th International Teletraffic Congress (ITC 30)},
interhash = {af32010d3b1123fa0603086684676db5},
intrahash = {bd74c7e45a441812378342f8949d0682},
keywords = {Session_1:_Performance_Evaluation itc itc30},
timestamp = {2020-05-24T20:14:34.000+0200},
title = {Whittle Index Approach to Energy-Aware Dispatching},
url = {https://gitlab2.informatik.uni-wuerzburg.de/itc-conference/itc-conference-public/-/raw/master/itc30/Aalto18ITC30.pdf?inline=true},
year = 2018
}