umu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Autonomous cloud resource provisioning: accounting, allocation, and performance control
Umeå University, Faculty of Science and Technology, Department of Computing Science.
2015 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

The emergence of large-scale Internet services coupled with the evolution of computing technologies such as distributed systems, parallel computing, utility computing, grid, and virtualization has fueled a movement toward a new resource provisioning paradigm called cloud computing. The main appeal of cloud computing lies in its ability to provide a shared pool of infinitely scalable computing resources for cloud services, which can be quickly provisioned and released on-demand with minimal effort. The rapidly growing interest in cloud computing from both the public and industry together with the rapid expansion in scale and complexity of cloud computing resources and the services hosted on them have made monitoring, controlling, and provisioning cloud computing resources at runtime into a very challenging and complex task. This thesis investigates algorithms, models and techniques for autonomously monitoring, controlling, and provisioning the various resources required to meet services’ performance requirements and account for their resource usage.

Quota management mechanisms are essential for controlling distributed shared resources so that services do not exceed their allocated or paid-for budget. Appropriate cloud-wide monitoring and controlling of quotas must be exercised to avoid over- or under-provisioning of resources. To this end, this thesis presents new distributed algorithms that efficiently manage quotas for services running across distributed nodes.

Determining the optimal amount of resources to meet services’ performance requirements is a key task in cloud computing. However, this task is extremely challenging due to multi-faceted issues such as the dynamic nature of cloud environments, the need for supporting heterogeneous services with different performance requirements, the unpredictable nature of services’ workloads, the non-triviality of mapping performance measurements into resources, and resource shortages. Models and techniques that can predict the optimal amount of resources needed to meet service performance requirements at runtime irrespective of variations in workloads are proposed. Moreover, different service differentiation schemes are proposed for managing temporary resource shortages due to, e.g., flash crowds or hardware failures.

In addition, the resources used by services must be accounted for in order to properly bill customers. Thus, monitoring data for running services should be collected and aggregated to maintain a single global state of the system that can be used to generate a single bill for each customer. However, collecting and aggregating such data across geographical distributed locations is challenging because the management task itself may consume significant computing and network resources unless done with care. A consistency and synchronization mechanism that can alleviate this task is proposed.

Place, publisher, year, edition, pages
Umeå: Umeå University , 2015. , 39 p.
Series
Report / UMINF, ISSN 0348-0542 ; 15.10
Keyword [en]
cloud computing, distributed infrastructure, monitoring, accounting, performance modeling, service differentiation
National Category
Computer Systems
Research subject
Computing Science
Identifiers
URN: urn:nbn:se:umu:diva-107955ISBN: 978-91-7601-334-2 (print)OAI: oai:DiVA.org:umu-107955DiVA: diva2:849876
Public defence
2015-09-28, MA121 (MIT building), Umeå University, Umeå, 10:15 (English)
Opponent
Supervisors
Available from: 2015-09-07 Created: 2015-08-31 Last updated: 2017-01-17Bibliographically approved
List of papers
1. Management of distributed resource allocations in multi-cluster environments
Open this publication in new window or tab >>Management of distributed resource allocations in multi-cluster environments
2012 (English)In: Performance Computing and Communications Conference (IPCCC) 2012, 31st International, IEEE, New York, USA: IEEE , 2012, 275-284 p.Conference paper, Published paper (Other academic)
Abstract [en]

We present a fully distributed solution for managing resource allocation for services running across multiple clusters in a large-scale cloud computing environment. Our solution allows individual services running across clusters to compete dynamically for allocations based on their rate of consumption while maintaining the global cloud level allocation limits. The solution monitors resource consumption by services that are spread over a number of clusters. Global polls are triggered only when the allocated balance in a cluster decreases below a threshold and allocations are reassigned in a manner that avoids further immediate global polls. Our solution achieves scalability by minimizing global message exchanges, increases performance by distributing requests, and improves availability by avoiding a single point of failure. We perform a range of simulations to verify the accuracy of our approach, to validate our theoretical results, and to evaluate against competing approaches.

Place, publisher, year, edition, pages
New York, USA: IEEE, 2012
Series
2012 IEEE 31ST International performance computing and communications conference (IPCCC), ISSN 1097-2641
Keyword
Distributed monitoring, distributed resource allocations management, distributed quota management
National Category
Computer Science
Identifiers
urn:nbn:se:umu:diva-66430 (URN)10.1109/PCCC.2012.6407768 (DOI)000313524400040 ()978-1-4673-4881-2 (ISBN)
Conference
IEEE 31st International Performance Computing and Communications Conference (IPCCC)2012-12-01--03, Austin, Texas, USA
Available from: 2013-02-22 Created: 2013-02-19 Last updated: 2015-09-01Bibliographically approved
2. A Tree-based Protocol for Enforcing Quotas in Clouds
Open this publication in new window or tab >>A Tree-based Protocol for Enforcing Quotas in Clouds
Show others...
2014 (English)In: the IEEE 10th 2014 World Congress on Services (SERVICES 2014), IEEE Computer Society, 2014Conference paper, Published paper (Refereed)
Abstract [en]

Services are more and more hosted on cloud nodes for enhancing their performance and increasing their availability. The virtually unlimited availability of resources enables service owners to consume resources without quantitative restrictions, paying only for what they consume. To avoid cost overrun, resource consumption must be controlled and capped when necessary.We present a distributed tree-based protocol to manage quotas in clouds that minimizes communication overhead and reduces the time required to inspect if a quota has been exhausted. Experimental evaluation shows that our protocol provides 42% more communication savings and is up to 15 times faster compared to a distributed baseline solution.

Place, publisher, year, edition, pages
IEEE Computer Society, 2014
Keyword
Distributed Quota Monitoring, Distributed Quota Enforcement and Management, Distributed Credit Management, Clouds
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-93382 (URN)
Conference
the IEEE 10th 2014 World Congress on Services (SERVICES 2014)
Available from: 2014-09-18 Created: 2014-09-18 Last updated: 2015-09-01Bibliographically approved
3. A synchronization mechanism for cloud accounting systems
Open this publication in new window or tab >>A synchronization mechanism for cloud accounting systems
Show others...
2014 (English)In: 2014 International Conference on Cloud and Autonomic Computing (ICCAC 2014), 2014, 111-120 p.Conference paper, Published paper (Refereed)
Abstract [en]

In current cloud systems, services run across multiple geographically distributed clusters and continuously generate resource usage data due to constant resource consumption. In the context of accounting, resource usage data generated from each cluster during service runtime must be collected and aggregated into a single cloud-wide record so that a single bill can be created. This paper presents a mechanism to synchronize accounting records among distributed accounting system peers. Run time resource usage generated from different clusters is synchronized to maintain a single cloud-wide view of the data so that a single bill can be created. We provide a set of accounting system requirements and an evaluation which verifies that the solution fulfills these requirements. Experimental results show that our solution produces less overhead in terms of data exchange and scales near-linearly with the size of clusters with no single point of failure.

Keyword
Distributed Accounting, Postpaid, Cloud, Synchronization
National Category
Computer and Information Science
Identifiers
urn:nbn:se:umu:diva-87753 (URN)10.1109/ICCAC.2014.11 (DOI)000370731000016 ()978-1-4799-5841-2 (ISBN)
Conference
2014 IEEE International Conference on Cloud and Autonomic Computing (ICCAC), Imperial College, London, ENGLAND, SEP 08-12, 2014
Available from: 2014-04-08 Created: 2014-04-08 Last updated: 2017-01-17Bibliographically approved
4. Towards faster response time models for vertical elasticity
Open this publication in new window or tab >>Towards faster response time models for vertical elasticity
2014 (English)In: 2014 IEEE/ACM 7TH INTERNATIONAL CONFERENCE ON UTILITY AND CLOUD COMPUTING (UCC), 2014, 560-565 p.Conference paper, Published paper (Refereed)
Abstract [en]

Resource provisioning in cloud computing is typ- ically coarse-grained. For example, entire CPU cores may be allocated for periods of up to an hour. The Resource-as-a- Service cloud concept has been introduced to improve the efficiency of resource utilization in clouds. In this concept, resources are allocated in terms of CPU core fractions, with granularities of seconds. Such infrastructures could be created using existing technologies such as lightweight virtualization using LXC or by exploiting the Xen hypervisor’s capacity for vertical elasticity. However, performance models for de- termining how much capacity to allocate to each application are currently lacking. To address this deficit, we evaluate two performance models for predicting mean response times: the previously proposed queue length model and the novel inverse model. The models are evaluated using 3 applications under both open and closed system models. The inverse model reacted rapidly and remained stable even with targets as low as 0.5 seconds. 

National Category
Computer Systems
Research subject
Computing Science
Identifiers
urn:nbn:se:umu:diva-93798 (URN)10.1109/UCC.2014.86 (DOI)000380558700079 ()978-1-4799-7881-6 (ISBN)
Conference
IEEE/ACM 7th International Conference on Utility and Cloud Computing (UCC), DEC 08-11, 2014, London, UNITED KINGDOM
Available from: 2014-10-01 Created: 2014-10-01 Last updated: 2017-01-16Bibliographically approved
5. Tail Response Time Modeling and Control for Interactive Cloud Services
Open this publication in new window or tab >>Tail Response Time Modeling and Control for Interactive Cloud Services
(English)Manuscript (preprint) (Other academic)
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-108028 (URN)
Available from: 2015-09-01 Created: 2015-09-01 Last updated: 2015-09-01Bibliographically approved
6. Coordinating CPU and Memory Elasticity Controllers to Meet Service Response Time Constraints
Open this publication in new window or tab >>Coordinating CPU and Memory Elasticity Controllers to Meet Service Response Time Constraints
Show others...
2015 (English)In: 2015 INTERNATIONAL CONFERENCE ON CLOUD AND AUTONOMIC COMPUTING (ICCAC), 2015, 69-80 p.Conference paper, Published paper (Refereed)
Abstract [en]

Vertical elasticity is recognized as a key enabler for efficient resource utilization of cloud infrastructure through fine-grained resource provisioning, e.g., allowing CPU cycles to be leased for as short as a few seconds. However, little research has been done to support vertical elasticity where the focus is mostly on a single resource, either CPU or memory, while an application may need arbitrary combinations of these resources at different stages of its execution. Nonetheless, the existing techniques cannot be readily used as-is without proper orchestration since they may lead to either under-or over-provisioning of resources and consequently result in undesirable behaviors such as performance disparity. The contribution of this paper is the design of an autonomic resource controller using a fuzzy control approach as a coordination technique. The novel controller dynamically adjusts the right amount of CPU and memory required to meet the performance objective of an application, namely its response time. We perform a thorough experimental evaluation using three different interactive benchmark applications, RUBiS, RUBBoS, and Olio, under workload traces generated based on open and closed system models. The results show that the coordination of memory and CPU elasticity controllers using the proposed fuzzy control provisions the right amount of resources to meet the response time target without over-committing any of the resource types. In contrast, with no coordinating between controllers, the behaviour of the system is unpredictable e.g., the application performance may be met but at the expense of over-provisioning of one of the resources, or application crashing due to severe resource shortage as a result of conflicting decisions.

Keyword
cloud-based application, fuzzy control, cloud computing, vertical elasticity, performance, cpu utilization, memory utilization
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-108032 (URN)10.1109/ICCAC.2015.20 (DOI)000380476500007 ()978-1-4673-9566-3 (ISBN)
Conference
2015 IEEE International Conference on Cloud and Autonomic Computing(ICCAC), Cambridge, MA, USA, September 21-24, 2015
Note

Originally included in thesis in accepted form.

Available from: 2015-09-01 Created: 2015-09-01 Last updated: 2016-09-23Bibliographically approved
7. Performance-Based Service Differentiation in Clouds
Open this publication in new window or tab >>Performance-Based Service Differentiation in Clouds
2015 (English)In: 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid), IEEE conference proceedings, 2015, 505-514 p.Conference paper, Published paper (Refereed)
Abstract [en]

Due to fierce competition, cloud providers need to run their data-centers efficiently. One of the issues is to increase data-center utilization while maintaining applications' performance targets. Achieving high data-center utilization while meeting applications' performance is difficult, as data-center overload may lead to poor performance of hosted services. Service differentiation has been proposed to control which services get degraded. However, current approaches are capacity-based, which are oblivious to the observed performance of each service and cannot divide the available capacity among hosted services so as to minimize overall performance degradation. In this paper we propose performance-based service differentiation. In case enough capacity is available, each service is automatically allocated the right amount of capacity that meets its target performance, expressed either as response time or throughput. In case of overload, we propose two service differentiation schemes that dynamically decide which services to degrade and to what extent. We carried out an extensive set of experiments using different services -- interactive as well as non-interactive -- by varying the workload mixes of each service over time. The results demonstrate that our solution precisely provides guaranteed performance or service differentiation depending on available capacity.

Place, publisher, year, edition, pages
IEEE conference proceedings, 2015
Keyword
cloud computing, computer centres, interactive systems, cloud providers, data centers, performance-based service differentiation, Computational modeling, Degradation, Noise, Resource management, Throughput, Time factors, Time measurement, cloud, elasticity, guaranteed service, performance models, service differentiation
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-108026 (URN)10.1109/CCGrid.2015.145 (DOI)000380493100051 ()978-1-4799-8006-2 (ISBN)
External cooperation:
Conference
2015 15th IEEE ACM International Symposium on Cluster Cloud and Grid Computing (CCGrid 2015),Shenzhen, PEOPLES R CHINA,MAY 04-07, 2015
Available from: 2015-09-01 Created: 2015-09-01 Last updated: 2016-09-13Bibliographically approved
8. Towards Optimized Self-Management of Distributed Object Storage Systems
Open this publication in new window or tab >>Towards Optimized Self-Management of Distributed Object Storage Systems
Show others...
2015 (English)Report (Other academic)
Abstract [en]

Cloud storage is increasingly adopted by users due to simplified storage systems compared to on-premise storage. These systems are mostly presented as Object Storage Systems (OSSs), hiding issues, such as redundancy, from users. As new industries are considering adopting clouds for storage, OSSs have to evolve to support new needs. Among the most challenging is assuring guaranteed performance.

In this paper, we present Controllable Trade-offs (CTO), an OSS-agnostic solution to add performance guarantees. CTO presents itself as a thin layer that mediates requests between the user and the OSS. For generic support, performance is controlled by tuning the rejection probability, and implemented as a user-side queue. Results show that CTO may reduce penalties 3.23 times on average and up to 68 times when the load is high.

Publisher
15 p.
Series
Report / UMINF, ISSN 0348-0542 ; 15.11
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-108033 (URN)
Available from: 2015-09-01 Created: 2015-09-01 Last updated: 2015-09-01Bibliographically approved

Open Access in DiVA

fulltext(757 kB)583 downloads
File information
File name FULLTEXT03.pdfFile size 757 kBChecksum SHA-512
58526d3d9599b03e453234a539b0bc9c6f927d6535bc202a59851b66f989b97d23c4ba1476c24186420e405d5e6e2040ada9c6ad2f4f931e20c46b054a559a0b
Type fulltextMimetype application/pdf
spikblad(119 kB)28 downloads
File information
File name FULLTEXT02.pdfFile size 119 kBChecksum SHA-512
df3b2eda535d7d750806a222a8b3b59dca4436113612f4707dc1285c04f721a65d9a712db8184674c3e563ac25bde9a48cb15e46e6cfc02a45d181d3eb56ddf3
Type spikbladMimetype application/pdf

Authority records BETA

Lakew, Ewnetu Bayuh

Search in DiVA

By author/editor
Lakew, Ewnetu Bayuh
By organisation
Department of Computing Science
Computer Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 611 downloads
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

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 2217 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf