Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Energy-Efficient Stable and Balanced Task Scheduling in Data Centers

Safavi, Mohammadhassan LU and Landfeldt, Björn LU (2021) In IEEE Transactions on Sustainable Computing 6(2).
Abstract
It is well known that load balancing in data centers can lead to unnecessary energy usage if all servers are kept active. Using
dynamic server provisioning, the number of servers that serve requests can be reduced by turning off idle servers and thereby saving
energy. However, such a scheme, usually increases the risk of instability of server queues. In this work, we analyze the trade-off
between energy usage and stability of servers in a data center when we balance the load by dispatching arriving jobs. We propose
algorithms to solve a stability and energy objective stochastic optimization problem with a high degree of flexibility to handle the trade-off
between these two objectives. We consider variable size jobs to... (More)
It is well known that load balancing in data centers can lead to unnecessary energy usage if all servers are kept active. Using
dynamic server provisioning, the number of servers that serve requests can be reduced by turning off idle servers and thereby saving
energy. However, such a scheme, usually increases the risk of instability of server queues. In this work, we analyze the trade-off
between energy usage and stability of servers in a data center when we balance the load by dispatching arriving jobs. We propose
algorithms to solve a stability and energy objective stochastic optimization problem with a high degree of flexibility to handle the trade-off
between these two objectives. We consider variable size jobs to apply load balancing on selected active servers and find that the optimal
solution is an NP-hard problem. We therefore develop two computationally efficient greedy and randomized approximation schemes to
achieve the trade-off between these objectives. We investigate the performance of our proposed algorithms in minimizing the risk of
queue length growth as well as the number of active servers needed to serve jobs, and compare it with several metrics in heterogeneous
load scenarios. (Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
IEEE Transactions on Sustainable Computing
volume
6
issue
2
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
external identifiers
  • scopus:85086721627
ISSN
2377-3782
DOI
10.1109/TSUSC.2020.2999717
language
English
LU publication?
yes
id
0ff51566-929b-4359-b55b-5f64548d8064
date added to LUP
2020-05-24 19:09:24
date last changed
2022-05-04 17:53:28
@article{0ff51566-929b-4359-b55b-5f64548d8064,
  abstract     = {{It is well known that load balancing in data centers can lead to unnecessary energy usage if all servers are kept active. Using<br/>dynamic server provisioning, the number of servers that serve requests can be reduced by turning off idle servers and thereby saving<br/>energy. However, such a scheme, usually increases the risk of instability of server queues. In this work, we analyze the trade-off<br/>between energy usage and stability of servers in a data center when we balance the load by dispatching arriving jobs. We propose<br/>algorithms to solve a stability and energy objective stochastic optimization problem with a high degree of flexibility to handle the trade-off<br/>between these two objectives. We consider variable size jobs to apply load balancing on selected active servers and find that the optimal<br/>solution is an NP-hard problem. We therefore develop two computationally efficient greedy and randomized approximation schemes to<br/>achieve the trade-off between these objectives. We investigate the performance of our proposed algorithms in minimizing the risk of<br/>queue length growth as well as the number of active servers needed to serve jobs, and compare it with several metrics in heterogeneous<br/>load scenarios.}},
  author       = {{Safavi, Mohammadhassan and Landfeldt, Björn}},
  issn         = {{2377-3782}},
  language     = {{eng}},
  number       = {{2}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  series       = {{IEEE Transactions on Sustainable Computing}},
  title        = {{Energy-Efficient Stable and Balanced Task Scheduling in Data Centers}},
  url          = {{http://dx.doi.org/10.1109/TSUSC.2020.2999717}},
  doi          = {{10.1109/TSUSC.2020.2999717}},
  volume       = {{6}},
  year         = {{2021}},
}