article

Translating Service Level Objectives to lower level policies for multi-tier services

Authors:

Yuan Chen,

Subu Iyer,

Xue Liu,

Dejan Milojicic,

Akhil SahaiAuthors Info & Claims

Cluster Computing, Volume 11, Issue 3

Pages 299 - 311

https://doi.org/10.1007/s10586-008-0059-6

Published: 01 September 2008 Publication History

Abstract

Service providers and their customers agree on certain quality of service guarantees through Service Level Agreements (SLA). An SLA contains one or more Service Level Objectives (SLO)s that describe the agreed-upon quality requirements at the service level. Translating these SLOs into lower-level policies that can then be used for design and monitoring purposes is a difficult problem. Usually domain experts are involved in this translation that often necessitates application of domain knowledge to this problem. In this article, we propose an approach that combines performance modeling with regression analysis to solve this problem. We demonstrate that our approach is practical and that it can be applied to different n -tier services. Our experiments show that for a typical 3-tier e-commerce application in a virtualized environment, the SLA can be met while improving CPU utilization by up to 3 times.

References

[1]

Barham, P., et al.: Xen and the art of virtualization. In: Proc. of the Nineteenth ACM SOSP, 2003.

Crossref

Google Scholar

[2]

Cecchet, E., Chanda, A., Elnikety, S., Marguerite, J., Zwaenepoel, W.: A comparison of software architectures for E-business applications. In: Proc. of 4th Middleware Conference, Rio de Janeiro, Brazil, June 2003.

Google Scholar

[3]

Chandra, A., Gong, W., Shenoy, P.: Dynamic resource allocation for shared data centers using online measurements. In: Proc. of International Workshop on Quality of Service, June 2003.

Crossref

Google Scholar

[4]

Cockcroft, A., Walker, B.: Capacity Planning for Internet Services. Sun Press, Cleveland (2001).

Crossref

Google Scholar

[5]

Council, T.P.C.: TPC-W. http://www.tpc.org/tpcw

Google Scholar

[6]

Doyle, R., Chase, J., Asad, O., Jin, W., Vahdat, A.: Model-based resource provisioning in a Web service utility. In: Proc. of the 4th USENIX USITS, Mar. 2003.

Crossref

Google Scholar

[7]

Edward, D., Lazowska, J., Zahorjan, G., Graham, S., Sevcik, K.C.: Quantitative System Performance: Computer System Analysis Using Queueing Network Models. Prentice-Hall, Englewood Cliffs (1984).

Crossref

Google Scholar

[8]

Gennaro, C., King, P.J.B.: Parallelising the mean value analysis algorithm. Simulation 72 (3), 148 (1999). 003754979907200304.

Crossref

Google Scholar

[9]

Graupner, S., Kotov, V., Trinks, H.: Resource-sharing and service deployment in virtual data centers. In: Proc. of the 22nd ICDCS, pp. 666-674, July 2002.

Crossref

Google Scholar

[10]

Kelley, T.: Detecting performance anomalies in global applications. In: Proc. of Second USENIX Workshop on Real, Large Distributed Systems (WORLDS 2005), 2005.

Crossref

Google Scholar

[11]

Levy, R., Nagarajarao, J., Pacifici, G., Spreitzer, M., Tantawi, A., Yousse, A.: Performance management for cluster based Web services. In: Proc. of IFIP/IEEE 8th IM, 2003.

Google Scholar

[12]

Liu, X., Heo, J., Sha, L.: Modeling 3-tiered Web applications. In: Proc. of 13th IEEE MASCOTS, Atlanta, Georgia, 2005.

Crossref

Google Scholar

[13]

Menasce, D., Almeida, V.: Capacity Planning for Web Services: Metrics, Models, and Methods. Prentice-Hall PTR, Englewood Cliffs (2001).

Crossref

Google Scholar

[14]

Reiser, M., Lavenberg, S.S.: Mean-value analysis of closed multichain queueing networks. J. ACM 27 , 313-322 (1980).

Digital Library

Google Scholar

[15]

RUBiS: Rice University Bidding System. http://www.cs.rice.edu/ CS/Systems/DynaServer/rubis

Google Scholar

[16]

Seidmann, A., Schweitzer, P.J., Shalev-Oren, S.: Computerized closed queueing network models of flexible manufacturing systems. Large Scale Syst. 12 , 91-107 (1987).

Google Scholar

[17]

Slothouber, L.: A model of Web server performance. In: Proc. of Int'l World Wide Web Conference, 1996.

Google Scholar

[18]

Stewart, C., Shen, K.: Performance modeling and system management for multi-component online services. In: Proc. of USENIX NSDI, 2005.

Digital Library

Google Scholar

[19]

Stewart, C., Kelly, T., Zhang, A.: Exploiting nonstationarity for performance prediction. In: Proc. of EuroSys, 2007.

Crossref

Google Scholar

[20]

Urgaonkar, B., Shenoy, P.: Cataclysm. Handling extreme overloads in Internet services. In: Proc. of ACM SIGACT-SIGOPS PODC, July 2004.

Crossref

Google Scholar

[21]

Urgaonkar, B., Pacifici, G., Shenoy, P., Spreitzer, M., Tantawi, A.: An analytical model for multi-tier Internet services and its applications. In: Proc. of ACM SIGMETRICS, June 2005.

Crossref

Google Scholar

[22]

Urgaonkar, B., Shenoy, P., Chandra, A., Goyal, O.P.: Dynamic provisioning of multi-tier Internet applications. In: Proc. of IEEE ICAC, June 2005.

Crossref

Google Scholar

[23]

VMware, Inc., VMware ESX Server User's Manual Version 1.5, Palo Alto, CA, April 2002.

Google Scholar

[24]

Yaikhom, G., Cole, M., Gilmore, S.: Combining measurement and stochastic modelling to enhance scheduling decisions for a parallel mean value analysis algorithm. In: Proc. of International Conference on Computational Science (ICCS 2006), LNCS. Springer, Berlin (2006).

Crossref

Google Scholar

[25]

Zhang, A., Santos, P., Beyer, D., Tang, H.: Optimal server resource allocation using an open queueing network model of response time. HP Labs Technical Report, HPL-2002-301.

Google Scholar

Cited By

View all

Rampérez VSoriano JLizcano DAljawarneh SLara J(2022)From SLA to vendor‐neutral metricsInternational Journal of Intelligent Systems10.1002/int.2263837:12(10533-10575)Online publication date: 29-Dec-2022
https://dl.acm.org/doi/10.1002/int.22638
Zhang LDown DDe Rango FMittal SRisco Martín JCetinkaya D(2015)Approximate mean value analysis for multi-core systemsProceedings of the International Symposium on Performance Evaluation of Computer and Telecommunication Systems10.5555/2874988.2874989(1-8)Online publication date: 26-Jul-2015
https://dl.acm.org/doi/10.5555/2874988.2874989
Mann Z(2015)Allocation of Virtual Machines in Cloud Data Centers—A Survey of Problem Models and Optimization AlgorithmsACM Computing Surveys10.1145/279721148:1(1-34)Online publication date: 10-Aug-2015
https://dl.acm.org/doi/10.1145/2797211
Show More Cited By

Index Terms

Translating Service Level Objectives to lower level policies for multi-tier services

Recommendations

Performance modeling and analysis of virtualized multi-tier applications under dynamic workloads

Virtual machine technology facilitates implementation of modern Internet services, especially multi-tier applications. Server virtualization aims to reduce the cost of service provisioning and improve fault tolerance, portability and security of ...
Service Level Aware - Contract Management
SCC '15: Proceedings of the 2015 IEEE International Conference on Services Computing

The success achieved by Cloud Computing Services (CCSs) in modern IT scenarios is nowadays a matter of fact and service offerings between providers and customers continuously grow up and widen their scope. Similarly, service offerings are more and more ...
Does service composition suffice to define business contracts for IT services in networked organizations
MEDES '13: Proceedings of the Fifth International Conference on Management of Emergent Digital EcoSystems

Currently several enterprises depend on services provided to them by other parties for the realization of their own service offer. The interconnections in terms of services offered and required by companies shape complex systems, called service ...

Reviews

Reviewer: Guenter Haring

Enterprise data centers are designed with on-demand computing and resource sharing in mind. In such centers, virtualization technologies enable applications to share computing resources. A crucial and challenging task in such environments is the translation of high-level service level objectives (SLOs) to low-level system thresholds. This paper discusses this problem for multi-tier services that are made up of a large number of components that interact with one another in a complex manner. In the first section, the authors describe the general problem and their approach, which is built on profiles characterizing per-component performance metrics and uses analytical models to provide relationships between high-level performance goals and low-level component goals. The authors restrict their approach to using the central processing unit (CPU) as the only shared resource. After a brief second section presents a motivating scenario, the SLO decomposition, which performs the translation, is explained in Section 3, in very general terms. Both the component profiling and the performance modeling are described, in order to give the reader an idea of the general concept of the approach. In Section 4, the modeling of multi-tier Web applications is extensively discussed, covering both single- and multi-class cases. The presented material is commonly known in the field. In Section 5, on profiling and SLO decomposition, the reader misses, on the one hand, an exact and detailed procedure and discussion of potential problems, and on the other hand, a more elaborated approach for the decomposition. Section 6 presents an experimental evaluation, where the authors show that their approach works. The results are presented without a thorough discussion, leaving the reader to ask what she or he has learned from this section. Before the short concluding section, the authors describe some related work. I'm not sure if the authors are fully aware of related work. For example, there are no references to the Workshops on Software and Performance (WOSP), which is an important event dealing with such issues. There is also no reference to layered queueing network models. Obviously, the paper was not done very carefully. Furthermore, there are many typographical errors and some of the references are incorrect. Though the topic is very important and interesting, this contribution is rather disappointing. Online Computing Reviews Service

Access critical reviews of Computing literature here

Become a reviewer for Computing Reviews.

Comments

Information & Contributors

Information

Published In

Cluster Computing Volume 11, Issue 3

September 2008

109 pages

ISSN:1386-7857

Issue’s Table of Contents

Publisher

Kluwer Academic Publishers

United States

Publication History

Published: 01 September 2008

Author Tags

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

7
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 25 Dec 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Rampérez VSoriano JLizcano DAljawarneh SLara J(2022)From SLA to vendor‐neutral metricsInternational Journal of Intelligent Systems10.1002/int.2263837:12(10533-10575)Online publication date: 29-Dec-2022
https://dl.acm.org/doi/10.1002/int.22638
Zhang LDown DDe Rango FMittal SRisco Martín JCetinkaya D(2015)Approximate mean value analysis for multi-core systemsProceedings of the International Symposium on Performance Evaluation of Computer and Telecommunication Systems10.5555/2874988.2874989(1-8)Online publication date: 26-Jul-2015
https://dl.acm.org/doi/10.5555/2874988.2874989
Mann Z(2015)Allocation of Virtual Machines in Cloud Data Centers—A Survey of Problem Models and Optimization AlgorithmsACM Computing Surveys10.1145/279721148:1(1-34)Online publication date: 10-Aug-2015
https://dl.acm.org/doi/10.1145/2797211
Wang QKanemasa YKawaba MPu CMilojicic DXu DTalwar V(2012)When average is not averageProceedings of the 9th international conference on Autonomic computing10.1145/2371536.2371544(33-42)Online publication date: 18-Sep-2012
https://dl.acm.org/doi/10.1145/2371536.2371544
Li HCasale GEllahi TAdamson ABondi AJuiz CSquillante M(2010)SLA-driven planning and optimization of enterprise applicationsProceedings of the first joint WOSP/SIPEW international conference on Performance engineering10.1145/1712605.1712625(117-128)Online publication date: 28-Jan-2010
https://dl.acm.org/doi/10.1145/1712605.1712625
Almeida JAlmeida VArdagna DCunha íFrancalanci CTrubian M(2010)Joint admission control and resource allocation in virtualized serversJournal of Parallel and Distributed Computing10.1016/j.jpdc.2009.08.00970:4(344-362)Online publication date: 1-Apr-2010
https://dl.acm.org/doi/10.1016/j.jpdc.2009.08.009
Rolia JCasale GKrishnamurthy DDawson SKraft SStea GMairesse JMendes J(2009)Predictive modelling of SAP ERP applicationsProceedings of the Fourth International ICST Conference on Performance Evaluation Methodologies and Tools10.4108/ICST.VALUETOOLS2009.7988(1-9)Online publication date: 20-Oct-2009
https://dl.acm.org/doi/10.4108/ICST.VALUETOOLS2009.7988

Abstract

References

Cited By

Index Terms

Recommendations

Performance modeling and analysis of virtualized multi-tier applications under dynamic workloads

Service Level Aware - Contract Management

Does service composition suffice to define business contracts for IT services in networked organizations

Reviews

Access critical reviews of Computing literature here

Comments

Information

Published In

Publisher

Publication History

Author Tags

Qualifiers

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

View options

Figures

Other

Share

Share this Publication link

Share on social media

Affiliations