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
Energy-efficient cloud computing: autonomic resource provisioning for datacenters
Umeå University, Faculty of Science and Technology, Department of Computing Science. (Cloud computing)
2018 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Energy efficiency has become an increasingly important concern in data centers because of issues associated with energy consumption, such as capital costs, operating expenses, and environmental impact. While energy loss due to suboptimal use of facilities and non-IT equipment has largely been reduced through the use of best-practice technologies, addressing energy wastage in IT equipment still requires the design and implementation of energy-aware resource management systems. This thesis focuses on the development of resource allocation methods to improve energy efficiency in data centers. The thesis employs three approaches to improve efficiency for optimized power and performance: scaling virtual machine (VM) and server processing capabilities to reduce energy consumption; improving resource usage through workload consolidation; and exploiting resource heterogeneity.

To achieve these goals, the first part of the thesis proposes models, algorithms, and techniques that reduce energy usage through the use of VM scaling, VM sizing for CPU and memory, CPU frequency adaptation, as well as hardware power capping for server-level resource allocation. The proposed online performance and power models capture system behavior while adapting to changes in the underlying infrastructure. Based on these models, the thesis proposes controllers that dynamically determine power-efficient resource allocations while minimizing performance penalty.

These methods are then extended to support resource overbooking and workload consolidation to improve resource utilization and energy efficiency across the cluster or data center. In order to cater for different performance requirements among collocated applications, such as latency-sensitive services and batch jobs, the controllers apply service differentiation among prioritized VMs and performance isolation techniques, including CPU pinning, quota enforcement, and online resource tuning.

This thesis also considers resource heterogeneity and proposes heterogeneousaware scheduling techniques to improve energy efficiency by integrating hardware accelerators (in this case FPGAs) and exploiting differences in energy footprint of different servers. In addition, the thesis provides a comprehensive study of the overheads associated with a number of virtualization platforms in order to understand the trade-offs provided by the latest technological advances and to make the best resource allocation decisions accordingly. The proposed methods in this thesis are evaluated by implementing prototypes on real testbeds and conducting experiments using real workload data taken from production systems and synthetic workload data that we generated. Our evaluation results demonstrate that the proposed approaches provide improved energy management of resources in virtualized data centers.

Place, publisher, year, edition, pages
Umeå: Umeå University , 2018. , p. 63
Series
Report / UMINF, ISSN 0348-0542 ; 18.05
Keywords [en]
Cloud computing, datacenter, energy-efficiency, performance management, virtualization
National Category
Computer Systems
Identifiers
URN: urn:nbn:se:umu:diva-145926ISBN: 978-91-7601-862-0 (print)OAI: oai:DiVA.org:umu-145926DiVA, id: diva2:1192294
Public defence
2018-04-16, MA121, MIT-building, Umeå, 10:15 (English)
Opponent
Supervisors
Available from: 2018-03-26 Created: 2018-03-22 Last updated: 2018-06-09Bibliographically approved
List of papers
1. A combined frequency scaling and application elasticity approach for energy-efficient cloud computing
Open this publication in new window or tab >>A combined frequency scaling and application elasticity approach for energy-efficient cloud computing
2014 (English)In: Sustainable Computing: Informatics and Systems, ISSN 2210-5379, E-ISSN 2210-5387, Vol. 4, no 4, p. 205-214Article in journal (Refereed) Published
Abstract [en]

Energy management has become increasingly necessary in large-scale cloud data centers to address high operational costs and carbon footprints to the environment. In this work, we combine three management techniques that can be used to control cloud data centers in an energy-efficient manner: changing the number of virtual machines, the number of cores, and scaling the CPU frequencies. We present a feedback controller that determines an optimal configuration to minimize energy consumption while meeting performance objectives. The controller can be configured to accomplish these goals in a stable manner, without causing large oscillations in the resource allocations. To meet the needs of individual applications under different workload conditions, the controller parameters are automatically adjusted at runtime based on a system model that is learned online. The potential of the proposed approach is evaluated in a video encoding scenario. The results show that our combined approach achieves up to 34% energy savings compared to the constituent approaches—core change, virtual machine change, and CPU frequency change policies, while meeting the performance target.

Place, publisher, year, edition, pages
Amsterdam: Elsevier, 2014
Keywords
Cloud computing, Energy-efficiency, Quality-of-service, Virtualization, Frequency scaling, Application elasticity
National Category
Computer Sciences
Research subject
Computer Science
Identifiers
urn:nbn:se:umu:diva-100656 (URN)10.1016/j.suscom.2014.08.007 (DOI)000209576700002 ()
Funder
Swedish Research Council, 2012-5908 f
Available from: 2015-03-05 Created: 2015-03-05 Last updated: 2018-06-07Bibliographically approved
2. Autonomic resource management for optimized power and performance in multi-tenant clouds
Open this publication in new window or tab >>Autonomic resource management for optimized power and performance in multi-tenant clouds
2016 (English)In: 2016 IEEE INTERNATIONAL CONFERENCE ON AUTONOMIC COMPUTING (ICAC) / [ed] Samuel Kounev, Holger Giese, Jie Liu, LOS ALAMITOS: IEEE Computer Society, 2016, p. 85-94Conference paper, Published paper (Refereed)
Abstract [en]

We present an autonomic resource management framework that takes advantage of both virtual machine resizing (CPU and memory) and physical CPU frequency scaling to reduce the power consumption of servers while meeting performance requirements of colocated applications. We design online performance and power model estimators that capture the complex relationships between applications' performance and server power (respectively), and resource utilization. Based on these models, we devise two optimization strategies to determine the most power efficient configuration. We also show that an operator can tune the tradeoff between power and performance. Our evaluation using a set of cloud benchmarks compares the proposed solution in power savings against the Linux ondemand and performance CPU governors. The results show that our solution achieves power savings between 12% to 20% compared to the baseline performance governor, while still meeting applications' performance goals.

Place, publisher, year, edition, pages
LOS ALAMITOS: IEEE Computer Society, 2016
Series
Proceedings of the International Conference on Autonomic Computing, ISSN 2474-0756
National Category
Computer Sciences
Identifiers
urn:nbn:se:umu:diva-121089 (URN)10.1109/ICAC.2016.32 (DOI)000390681200013 ()978-1-5090-1654-9 (ISBN)
Conference
13th IEEE International Conference on Autonomic Computing (ICAC), JUL 17-22, 2016, Würzburg, Germany
Available from: 2016-05-26 Created: 2016-05-26 Last updated: 2018-06-07Bibliographically approved
3. OptiBook: Optimal Resource Booking for Energy-efficient Datacenters
Open this publication in new window or tab >>OptiBook: Optimal Resource Booking for Energy-efficient Datacenters
2017 (English)In: 2017 IEEE/ACM 25th International Symposium on Quality of Service (IWQoS), IEEE Communications Society, 2017Conference paper, Published paper (Refereed)
Abstract [en]

A lack of energy proportionality, low resource utilization, and interference in virtualized infrastructure make the cloud a challenging target environment for improving energy efficiency. In this paper we present OptiBook, a system that improves energy proportionality and/or resource utilization to optimize performance and energy efficiency. OptiBook shares servers between latency-sensitive services and batch jobs, over- books the system in a controllable manner, uses vertical (CPU and DVFS) scaling for prioritized virtual machines, and applies performance isolation techniques such as CPU pinning and quota enforcement as well as online resource tuning to effectively improve energy efficiency. Our evaluations show that on average, OptiBook improves performance per watt by 20% and reduces energy consumption by 9% while minimizing SLO violations. 

Place, publisher, year, edition, pages
IEEE Communications Society, 2017
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-145492 (URN)10.1109/IWQoS.2017.7969135 (DOI)000428199300029 ()978-1-5386-2704-4 (ISBN)978-1-5386-2705-1 (ISBN)
Conference
2017 IEEE/ACM 25th International Symposium on Quality of Service (IWQoS), Vilanova i la Geltrú, Spain, June 14-16, 2017
Available from: 2018-03-07 Created: 2018-03-07 Last updated: 2018-06-09Bibliographically approved
4. Power and Performance Optimization in FPGA-accelerated Clouds
Open this publication in new window or tab >>Power and Performance Optimization in FPGA-accelerated Clouds
Show others...
2018 (English)In: Concurrency and Computation, ISSN 1532-0626, E-ISSN 1532-0634, Vol. 30, no 18, article id e4526Article in journal (Other academic) Published
Abstract [en]

Energy management has become increasingly necessary in data centers to address all energy-related costs, including capital costs, operating expenses, and environmental impacts. Heterogeneous systems with mixed hardware architectures provide both throughput and processing efficiency for different specialized application types and thus have a potential for significant energy savings. However, the presence of multiple and different processing elements increases the complexity of resource assignment. In this paper, we propose a system for efficient resource management in heterogeneous clouds. The proposed approach maps applications' requirement to different resources reducing power usage with minimum impact on performance. A technique that combines the scheduling of custom hardware accelerators, in our case, Field-Programmable Gate Arrays (FPGAs) and optimized resource allocation technique for commodity servers, is proposed. We consider an energy-aware scheduling technique that uses both the applications' performance and their deadlines to control the assignment of FPGAs to applications that would consume the most energy. Once the scheduler has performed the mapping between a VM and an FPGA, an optimizer handles the remaining VMs in the server, using vertical scaling and CPU frequency adaptation to reduce energy consumption while maintaining the required performance. Our evaluation using interactive and data-intensive applications compare the effectiveness of the proposed solution in energy savings as well as maintaining applications performance, obtaining up to a 32% improvement in the performance-energy ratio on a mix of multimedia and e-commerce applications.

Place, publisher, year, edition, pages
John Wiley & Sons, 2018
Keywords
cloud computing, energy efficiency, FPGA-aware
National Category
Computer Sciences
Identifiers
urn:nbn:se:umu:diva-121092 (URN)10.1002/cpe.4526 (DOI)000442575600010 ()
Funder
Swedish Research Council
Available from: 2016-05-26 Created: 2016-05-26 Last updated: 2019-01-15Bibliographically approved
5. PerfGreen: Performance and Energy Aware Resource Provisioning for Heterogeneous Clouds
Open this publication in new window or tab >>PerfGreen: Performance and Energy Aware Resource Provisioning for Heterogeneous Clouds
2018 (English)In: 2018 IEEE International Conference on Autonomic Computing (ICAC), 2018, p. 81-90Conference paper, Published paper (Refereed)
Abstract [en]

Improving energy efficiency in a cloud environment is challenging because of poor energy proportionality, low resource utilization, interference as well as workload, application, and hardware dynamism. In this paper we present PerfGreen, a dynamic auto-tuning resource management system for improving energy efficiency with minimal performance impact in heterogeneous clouds. PerfGreen achieves this through a combination of admission control, scheduling, and online resource allocation methods with performance isolation and application priority techniques. Scheduling in PerfGreen is energy aware and power management capabilities such as CPU frequency adaptation and hard CPU power limiting are exploited. CPU scaling is combined with performance isolation techniques, including CPU pinning and quota enforcement, for prioritized virtual machines to improve energy efficiency. An evaluation based on our prototype implementation shows that PerfGreen with its energy-aware scheduler and resource allocator on average reduces energy usage by 53%, improves performance per watt by 64%, and server density by 25% while keeping performance deviations to a minimum.

Series
Proceedings of the International Conference on Autonomic Computing, ISSN 2474-0756
National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-145925 (URN)10.1109/ICAC.2018.00018 (DOI)000450120900009 ()
Conference
15TH IEEE INTERNATIONAL CONFERENCE ON AUTONOMIC COMPUTING (ICAC 2018), Trento, ITALY, SEP 03-07, 2018
Available from: 2018-03-22 Created: 2018-03-22 Last updated: 2019-01-07Bibliographically approved
6. Virtualization Techniques Compared: Performance, Resource, and Power Usage Overheads in Clouds
Open this publication in new window or tab >>Virtualization Techniques Compared: Performance, Resource, and Power Usage Overheads in Clouds
2018 (English)Manuscript (preprint) (Other academic)
Abstract [en]

Virtualization solutions based on hypervisors or containers are enabling technologies

for scalable, flexible, and cost-effective resource sharing. As the fundamental

limitations of each technology are yet to be understood, they need to be regularly

reevaluated to better understand the trade-off provided by latest technological advances.

This paper presents an in-depth quantitative analysis of virtualization

overheads in these two groups of systems and their gaps relative to native environments

based on a diverse set of workloads that stress CPU, memory, storage,

and networking resources. KVM and XEN are used to represent hypervisor-based

virtualization, and LXC and Docker for container-based platforms. The systems

were evaluated with respect to several cloud resource management dimensions including

performance, isolation, resource usage, energy efficiency, start-up time,

and density. Our study is useful both to practitioners to understand the current

state of the technology in order to make the right decision in the selection, operation

and/or design of platforms and to scholars to illustrate how these technologies

evolved over time.

National Category
Computer Systems
Identifiers
urn:nbn:se:umu:diva-145924 (URN)
Conference
ACM/SPEC Internation Conference on Performance Engineering (ICPE)
Available from: 2018-03-22 Created: 2018-03-22 Last updated: 2018-06-09

Open Access in DiVA

fulltext(795 kB)365 downloads
File information
File name FULLTEXT02.pdfFile size 795 kBChecksum SHA-512
f68bdbd6b927c45efb9b1c59571b4d708f5406bc0af63134bd2ba5a29794d2751ea4b4d2837d1c23c363342ad7ef799057f3bab053d657e733f38b14cbfeb0bb
Type fulltextMimetype application/pdf
spikblad(127 kB)42 downloads
File information
File name SPIKBLAD01.pdfFile size 127 kBChecksum SHA-512
b10bf0bd5966c036211bd6d3118ec3efb6379477d66f853597af459237f13ef21a0d9f024b7b9f8d387b44a542c77b4dd52b5fb73236d4b9f70f17017054fb61
Type spikbladMimetype application/pdf

Authority records BETA

Tesfatsion, Selome Kostentinos

Search in DiVA

By author/editor
Tesfatsion, Selome Kostentinos
By organisation
Department of Computing Science
Computer Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 365 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: 1561 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