Dynamic Server Provisioning with Queue Stabilization in Data Centres

Mohammadhassan Safavi, Mehmet Karaca, Saeed Bastani, Björn Landfeldt

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

1 Citation (SciVal)

Abstract

When load balancing data centres with all servers
active to serve jobs, the result can be excessive energy usage. On
the other hand, using dynamic server provisioning, the number
of servers that serve requests can be reduced by turning off idle
servers and thereby save 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 an algorithm, SEOL, 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 show the performance of our proposed algorithm
in minimizing the risk of queue length growth as well as the
number of active servers needed to serve jobs, and compare it
with a number of well-established commercial load balancing
mechanisms.
Original languageEnglish
Title of host publicationIEEE International Conference on Cloud Networking
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)978-1-5090-4026-1
ISBN (Print)978-1-5090-4027-8
DOIs
Publication statusPublished - 2017
Event6th IEEE International Conference on Cloud Networking (IEEE CloudNet 2017) - Prague, Czech Republic
Duration: 2017 Sep 252017 Sep 27

Conference

Conference6th IEEE International Conference on Cloud Networking (IEEE CloudNet 2017)
Country/TerritoryCzech Republic
CityPrague
Period2017/09/252017/09/27

Subject classification (UKÄ)

  • Communication Systems

Keywords

  • Energy Usage
  • Data centers
  • Stability

Fingerprint

Dive into the research topics of 'Dynamic Server Provisioning with Queue Stabilization in Data Centres'. Together they form a unique fingerprint.

Cite this