Change search
ReferencesLink to record
Permanent link

Direct link
Improving Cloud Service Resilience using Brownout-Aware Load-Balancing
Umeå University, Faculty of Science and Technology, Department of Computing Science. (DS)ORCID iD: 0000-0003-0106-3049
Lund University, Sweden.ORCID iD: 0000-0002-1364-8127
Lund University, Sweden.
Lund University, Sweden.
Show others and affiliations
2014 (English)Conference paper (Refereed)
Abstract [en]

We focus on improving resilience of cloud services (e.g., e-commerce website), when correlated or cascading failures lead to computing capacity shortage. We study how to extend the classical cloud service architecture composed of a load-balancer and replicas with a recently proposed self-adaptive paradigm called brownout. Such services are able to reduce their capacity requirements by degrading user experience (e.g., disabling recommendations).

Combining resilience with the brownout paradigm is to date an open practical problem. The issue is to ensure that replica self-adaptivity would not confuse the load-balancing algorithm, overloading replicas that are already struggling with capacity shortage. For example, load-balancing strategies based on response times are not able to decide which replicas should be selected, since the response times are already controlled by the brownout paradigm.

In this paper we propose two novel brownout-aware load-balancing algorithms. To test their practical applicability, we extended the popular lighttpd web server and load-balancer, thus obtaining a production-ready implementation. Experimental evaluation shows that the approach enables cloud services to remain responsive despite cascading failures. Moreover, when compared to Shortest Queue First (SQF), believed to be near-optimal in the non-adaptive case, our algorithms improve user experience by 5%, with high statistical significance, while preserving response time predictability.

Place, publisher, year, edition, pages
IEEE Computer Society, 2014.
Keyword [en]
cloud, load-balancing, self-adaptation, control theory, statistical evaluation
National Category
Computer Systems Control Engineering
Research subject
Computer Science
URN: urn:nbn:se:umu:diva-91327OAI: diva2:735567
33rd IEEE Symposium on Reliable Distributed Systems (SRDS 2014)
Cloud Control
Swedish Research CouncileLLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationsLinnaeus research environment CADICS
Available from: 2014-07-29 Created: 2014-07-29 Last updated: 2014-08-08Bibliographically approved

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Klein, CristianPapadopoulos, Alessandro VittorioMaggio, MartinaHernández-Rodriguez, FranciscoElmroth, Erik
By organisation
Department of Computing Science
Computer SystemsControl Engineering

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 67 hits
ReferencesLink to record
Permanent link

Direct link