umu.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Publications (7 of 7) Show all publications
Papadopoulos, A. V., Klein, C., Maggio, M., Dürango, J., Dellkrantz, M., Hernández-Rodriguez, F., . . . Årzén, K.-E. (2016). Control-based load-balancing techniques: Analysis and performance evaluation via a randomized optimization approach. Control Engineering Practice, 52, 24-34
Open this publication in new window or tab >>Control-based load-balancing techniques: Analysis and performance evaluation via a randomized optimization approach
Show others...
2016 (English)In: Control Engineering Practice, ISSN 0967-0661, E-ISSN 1873-6939, Vol. 52, p. 24-34Article in journal (Refereed) Published
Abstract [en]

Cloud applications are often subject to unexpected events like flashcrowds and hardware failures. Users that expect a predictable behavior may abandon an unresponsive application when these events occur. Researchers and engineers addressed this problem on two separate fronts: first, they introduced replicas - copies of the application with the same functionality - for redundancy and scalability; second, they added a self-adaptive feature called brownout inside cloud applications to bound response times by modulating user experience. The presence of multiple replicas requires a dedicated component to direct incoming traffic: a load-balancer. Existing load-balancing strategies based on response times interfere with the response time controller developed for brownout-compliant applications. In fact, the brownout approach bounds response times using a control action. Hence, the response time, that was used to aid load-balancing decision, is not a good indicator of how well a replica is performing. To fix this issue, this paper reviews some proposal for brownout-aware load-balancing and provides a comprehensive experimental evaluation that compares them. To provide formal guarantees on the load balancing performance, we use a randomized optimization approach and apply the scenario theory. We perform an extensive set of experiments on a real machine, extending the popular lighttpd web server and load-balancer, and obtaining a production-ready implementation. Experimental results show an improvement of the user experience over Shortest Queue First (SQF)-believed to be near-optimal in the non-adaptive case. The improved user experience is obtained preserving the response time predictability.

Place, publisher, year, edition, pages
Elsevier, 2016
Keywords
Load-balancing, Randomized optimization, Cloud control
National Category
Computer Sciences
Research subject
Computer Science
Identifiers
urn:nbn:se:umu:diva-119368 (URN)10.1016/j.conengprac.2016.03.020 (DOI)000377740300003 ()
Funder
Swedish Research Council, Cloud ControlSwedish Research Council, Power and temperature control for large-scale computing infrastructuresELLIIT - The Linköping‐Lund Initiative on IT and Mobile Communications
Available from: 2016-04-18 Created: 2016-04-18 Last updated: 2018-06-07Bibliographically approved
Klein, C., Maggio, M., Årzén, K.-E. & Hernández-Rodriguez, F. (2014). Brownout: Building More Robust Cloud Applications. In: 36th International Conference on Software Engineering (ICSE 2014): . Paper presented at 36th International Conference on Software Engineering ICSE 2014, Hyderabad, India, May 31-June 7 2014 (pp. 700-711). ACM Digital Library
Open this publication in new window or tab >>Brownout: Building More Robust Cloud Applications
2014 (English)In: 36th International Conference on Software Engineering (ICSE 2014), ACM Digital Library, 2014, p. 700-711Conference paper, Published paper (Refereed)
Abstract [en]

Self-adaptation is a first class concern for cloud applications, which should be able to withstand diverse runtime changes. Variations are simultaneously happening both at the cloud infrastructure level - for example hardware failures - and at the user workload level - flash crowds. However, robustly withstanding extreme variability, requires costly hardware over-provisioning. In this paper, we introduce a self-adaptation programming paradigm called brownout. Using this paradigm, applications can be designed to robustly withstand unpredictable runtime variations, without over-provisioning. The paradigm is based on optional code that can be dynamically deactivated through decisions based on control theory. We modified two popular web application prototypes - RUBiS and RUBBoS - with less than 170 lines of code, to make them brownout-compliant. Experiments show that brownout self-adaptation dramatically improves the ability to withstand flash-crowds and hardware failures.

Place, publisher, year, edition, pages
ACM Digital Library, 2014
Keywords
adaptive Software, control theory, brownout, cloud
National Category
Computer Systems Control Engineering
Research subject
Automatic Control; Computer Science
Identifiers
urn:nbn:se:umu:diva-84212 (URN)10.1145/2568225.2568227 (DOI)000387829200062 ()
Conference
36th International Conference on Software Engineering ICSE 2014, Hyderabad, India, May 31-June 7 2014
Projects
Cloud Control
Funder
eSSENCE - An eScience CollaborationEU, FP7, Seventh Framework Programme, 257019Linnaeus research environment CADICSELLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationsSwedish Research Council, C0590801
Note

accepted

Available from: 2013-12-18 Created: 2013-12-18 Last updated: 2019-06-19Bibliographically approved
Maggio, M., Klein, C. & Årzén, K.-E. (2014). Control strategies for predictable brownouts in cloud applications. In: IFAC PAPERSONLINE: . Paper presented at 19th World Congress of the International-Federation-of-Automatic-Control (IFAC), AUG 24-29, 2014, Cape Town, SOUTH AFRICA (pp. 689-694). , 47
Open this publication in new window or tab >>Control strategies for predictable brownouts in cloud applications
2014 (English)In: IFAC PAPERSONLINE, 2014, Vol. 47, p. 689-694Conference paper, Published paper (Refereed)
Abstract [en]

Cloud computing is an application hosting model providing the illusion of infinite computing power. However, even the largest datacenters have finite computing capacity, thus cloud infrastructures have experienced overload due to overbooking or transient failures. The topic of this paper is the comparison of different control strategies to mitigate overload for datacenters, that assume that the running cloud applications are cooperative and help the infrastructure in recovering from critical events. Specifically, the paper investigates the behavior of different controllers when they have to keep the average response time of a cloud application below a certain threshold by acting on the probability of serving requests with optional computations disabled, where the pressure exerted by each request on the infrastructure is diminished, at the expense of user experience.

Series
IFAC PAPERSONLINE, ISSN 2405-8963
Keywords
Computer systems, Feedback loops, Model-based control, Multiprocessor systems, Probabilitstic models, Queuing theory
National Category
Computer Systems Control Engineering
Research subject
Automatic Control; Computing Science
Identifiers
urn:nbn:se:umu:diva-84211 (URN)000391107100112 ()
Conference
19th World Congress of the International-Federation-of-Automatic-Control (IFAC), AUG 24-29, 2014, Cape Town, SOUTH AFRICA
Funder
ELLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationseSSENCE - An eScience CollaborationEU, FP7, Seventh Framework Programme, 257019Swedish Research Council, C0590801Linnaeus research environment CADICS
Note

Issue: 3

Available from: 2013-12-18 Created: 2013-12-18 Last updated: 2018-06-08Bibliographically approved
Klein, C., Papadopoulos, A. V., Dellkrantz, M., Dürango, J., Maggio, M., Årzén, K.-E., . . . Elmroth, E. (2014). Improving Cloud Service Resilience using Brownout-Aware Load-Balancing. In: 2014 IEEE 33RD INTERNATIONAL SYMPOSIUM ON RELIABLE DISTRIBUTED SYSTEMS (SRDS): . Paper presented at 2014 IEEE 33RD INTERNATIONAL SYMPOSIUM ON RELIABLE DISTRIBUTED SYSTEMS (SRDS), OCT 06-09, 2014, Nara, JAPAN (pp. 31-40). IEEE Computer Society
Open this publication in new window or tab >>Improving Cloud Service Resilience using Brownout-Aware Load-Balancing
Show others...
2014 (English)In: 2014 IEEE 33RD INTERNATIONAL SYMPOSIUM ON RELIABLE DISTRIBUTED SYSTEMS (SRDS), IEEE Computer Society, 2014, p. 31-40Conference paper, Published 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
Series
Symposium on Reliable Distributed Systems Proceedings, ISSN 1060-9857
Keywords
cloud, load-balancing, self-adaptation, control theory, statistical evaluation
National Category
Computer Systems Control Engineering
Research subject
Computer Science
Identifiers
urn:nbn:se:umu:diva-91327 (URN)000380439400004 ()978-1-4799-5584-8 (ISBN)
Conference
2014 IEEE 33RD INTERNATIONAL SYMPOSIUM ON RELIABLE DISTRIBUTED SYSTEMS (SRDS), OCT 06-09, 2014, Nara, JAPAN
Projects
Cloud Control
Funder
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: 2018-06-07Bibliographically approved
Klein, C., Maggio, M., Årzén, K.-E. & Hernández-Rodriguez, F. (2013). Introducing Service-level Awareness in the Cloud. In: Proceedings of the 4th annual Symposium on Cloud Computing: . Paper presented at 2013 ACM Symposium on Cloud Computing, October 1-3, 2013, Santa Clara, CA. Association for Computing Machinery (ACM)
Open this publication in new window or tab >>Introducing Service-level Awareness in the Cloud
2013 (English)In: Proceedings of the 4th annual Symposium on Cloud Computing, Association for Computing Machinery (ACM), 2013Conference paper, Poster (with or without abstract) (Refereed)
Place, publisher, year, edition, pages
Association for Computing Machinery (ACM), 2013
Keywords
service-level, cloud, resource management, controller, performance modelling
National Category
Control Engineering Computer Systems
Research subject
Computer Science; Automatic Control
Identifiers
urn:nbn:se:umu:diva-84207 (URN)10.1145/2523616.2525936 (DOI)978-1-4503-2428-1 (ISBN)
Conference
2013 ACM Symposium on Cloud Computing, October 1-3, 2013, Santa Clara, CA
Projects
Cloud Control
Funder
EU, FP7, Seventh Framework Programme, 257019Swedish Research Council, C0590801eLLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationsLinnaeus research environment CADICSeSSENCE - An eScience Collaboration
Available from: 2013-12-18 Created: 2013-12-18 Last updated: 2018-06-08Bibliographically approved
Klein, C., Maggio, M., Årzén, K.-E. & Hernández-Rodriguez, F. (2013). Introducing Service-level Awareness in the Cloud. Lund University Open Access
Open this publication in new window or tab >>Introducing Service-level Awareness in the Cloud
2013 (English)Report (Other academic)
Abstract [en]

Resource allocation in clouds is mostly done assuming hard requirements, applications either receive the requested resources or fail. Given the dynamic nature of workloads, guaranteeing on-demand allocations requires large spare capacity. Hence, one cannot have a system that is both reliable and efficient.

To solve this issue, we introduce Service-Level (SL) awareness in clouds, assuming applications contain some optional code that can be dynamically deactivated as needed. First, we design a model for such applications and synthesize a controller to decide when to execute the optional code and when to skip it. Then, we propose a Resource Manager (RM) that allocates resources to multiple SL-aware applications in a fair manner. We theoretically prove properties of the overall system using control and game theory.

To show the practical applicability, we implemented SL-aware versions of RUBiS and RUBBoS with less than 170 lines of code. Experiments show that SL-awareness may enable a factor 8 improvement in withstanding flash-crowds or failures. SL-awareness opens up more flexibility in cloud resource management, which is why we encourage further research by publishing all source code.

Place, publisher, year, edition, pages
Lund University Open Access, 2013
Keywords
resource management, service level, control theory, game theory, cloud
National Category
Computer Systems Control Engineering
Research subject
Automatic Control; Computer Science
Identifiers
urn:nbn:se:umu:diva-84213 (URN)
Projects
Cloud Control
Funder
eLLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationseSSENCE - An eScience CollaborationEU, FP7, Seventh Framework Programme, 257019Swedish Research Council, C0590801Linnaeus research environment CADICS
Available from: 2013-12-18 Created: 2013-12-18 Last updated: 2018-06-08Bibliographically approved
Klein, C., Maggio, M., Årzén, K.-E. & Hernández-Rodriguez, F. (2013). Resource management for service level aware cloud applications. In: : . Paper presented at International Workshop on Real-time and Distributed The 2nd International Workshop on Real-Time and Distributed Computing in Emerging Applications (Co-located with 34th IEEE Real-time Systems Symposium). IEEE Computer Society
Open this publication in new window or tab >>Resource management for service level aware cloud applications
2013 (English)Conference paper, Published paper (Refereed)
Abstract [en]

Resource allocation in clouds is mostly done assuming hard requirements, time-sensitive applications either receive the requested resources or fail. Given the dynamic nature of workloads, guaranteeing on-demand allocations requires large spare capacity. Hence, one cannot have a system that is both reliable and efficient.

To mitigate this issue, we introduce service-level awareness in clouds, assuming applications contain some optional code that can be dynamically deactivated as needed. We propose a resource manager that allocates resources to multiple service-level-aware applications in a fair manner. To show the practical applicability, we implemented service-level-aware versions of RUBiS and RUBBoS, two popular cloud benchmarks, together with our resource manager. Experiments show that service-level awareness helps in withstanding flash-crowds or failures, opening up more flexibility in cloud resource management.

Place, publisher, year, edition, pages
IEEE Computer Society, 2013
Keywords
resource management; service level; control theory
National Category
Computer Systems Control Engineering
Research subject
Computer Science; Automatic Control
Identifiers
urn:nbn:se:umu:diva-84209 (URN)
Conference
International Workshop on Real-time and Distributed The 2nd International Workshop on Real-Time and Distributed Computing in Emerging Applications (Co-located with 34th IEEE Real-time Systems Symposium)
Funder
EU, FP7, Seventh Framework Programme, 257019eSSENCE - An eScience CollaborationSwedish Research Council, C0590801eLLIIT - The Linköping‐Lund Initiative on IT and Mobile CommunicationsLinnaeus research environment CADICS
Available from: 2013-12-18 Created: 2013-12-18 Last updated: 2018-06-08Bibliographically approved
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-1143-1127

Search in DiVA

Show all publications