skip to main content
10.1145/2371536.2371544acmconferencesArticle/Chapter ViewAbstractPublication PagesicacConference Proceedingsconference-collections
research-article

When average is not average: large response time fluctuations in n-tier systems

Published: 18 September 2012 Publication History

Abstract

Simultaneously achieving good performance and high resource utilization is an important goal for production cloud environments. Through extensive measurements of an n-tier application benchmark (RUBBoS), we show that system response time frequently presents large scale fluctuations (e.g., ranging from tens of milliseconds up to tens of seconds) during periods of high resource utilization.
Except the factor of bursty workload from clients, we found that the large scale response time fluctuations can be caused by some system environmental conditions (e.g., L2 cache miss, JVM garbage collection, inefficient scheduling policies) that commonly exist in n-tier applications. The impact of these system environmental conditions can largely amplify the end-to-end response time fluctuations because of the complex resource dependencies in the system. For instance, a 50ms response time increase in the database tier can be amplified to 500ms end-to-end response time increase. We evaluate three heuristics to stabilize response time fluctuations while still achieving high resource utilization in the system. Our results show that large scale response time fluctuations should be taken into account when designing effective autonomous self-scaling n-tier systems in cloud.

References

[1]
Rice University Bulletin Board System. "http://jmob.ow2.org/rubbos.html", 2004.
[2]
What is the average response time for displaying results. "http://support.google.com/mini/bin/answer.py?hl=en&answer=15796", 2004.
[3]
Fujitsu SysViz: System Visualization. "http://www.google.com/patents?id=0pGRAAAAEBAJ&zoom=4&pg=PA1#v=onepage&q&f=false", 2010.
[4]
Java SE 6 Performance White Paper. http://java.sun.com/performance/reference/whitepapers/6_performance.html, 2010.
[5]
T. Abdelzaher, Y. Diao, J. Hellerstein, C. Lu, and X. Zhu. Introduction to control theory and its application to computing systems. SIGMETRICS Tutorial, 2008.
[6]
P. Bodik, A. Fox, M. J. Franklin, M. I. Jordan, and D. A. Patterson. Characterizing, modeling, and generating workload spikes for stateful services. SoCC'10.
[7]
Y. Chen, S. Iyer, X. Liu, D. Milojicic, and A. Sahai. Translating service level objectives to lower level policies for multi-tier services. Cluster Computing, 2008.
[8]
S. Cho and L. Jin. Managing distributed, shared l2 caches through os-level page allocation. MICRO'06.
[9]
T. Forell, D. Milojicic, and V. Talwar. Cloud management: Challenges and opportunities. In IPDPSW'11.
[10]
M. Harchol-Balter, B. Schroeder, N. Bansal, and M. Agrawal. Size-based scheduling to improve web performance. ACM Trans. Comput. Syst., 2003.
[11]
E. C. Julie, J. Marguerite, and W. Zwaenepoel. C-JDBC: Flexible Database Clustering Middleware. 2004.
[12]
H. C. Lim, S. Babu, and J. S. Chase. Automated control for elastic storage. ICAC'10.
[13]
N. Mi, G. Casale, L. Cherkasova, and E. Smirni. Injecting realistic burstiness to a traditional client-server benchmark. ICAC'09.
[14]
K. S. Min Lee. Region scheduling: Efficiently using the cache architectures via page-level affinity. ASPLOS'12.
[15]
P. Padala, K.-Y. Hou, K. G. Shin, X. Zhu, M. Uysal, Z. Wang, S. Singhal, and A. Merchant. Automated control of multiple virtualized resources. EuroSys'09.
[16]
B. Schroeder, A. Wierman, and M. Harchol-Balter. Open versus closed: a cautionary tale. NSDI'06.
[17]
B. Snyder. Server virtualization has stalled, despite the hype. InfoWorld, 2010.
[18]
E. Thereska and G. R. Ganger. Ironmodel: robust performance models in the wild. SIGMETRICS'08.
[19]
Q. Wang, S. Malkowski, Y. Kanemasa, D. Jayasinghe, P. Xiong, M. Kawaba, L. Harada, and C. Pu. The impact of soft resource allocation on n-tier application scalability. IPDPS'11.
[20]
P. Xiong, Z. Wang, S. Malkowski, Q. Wang, D. Jayasinghe, and C. Pu. Economical and robust provisioning of n-tier cloud workloads: A multi-level control approach. ICDCS'11.
[21]
X. Zhu, M. Uysal, Z. Wang, S. Singhal, A. Merchant, P. Padala, and K. Shin. What does control theory bring to systems research? SIGOPS Oper. Syst. Rev., 2009.

Cited By

View all
  • (2024)A Study of Response Time Instability of Microservices at High Resource Utilization in the Cloud2024 IEEE 6th International Conference on Cognitive Machine Intelligence (CogMI)10.1109/CogMI62246.2024.00024(111-116)Online publication date: 28-Oct-2024
  • (2022)Predictive Hybrid Autoscaling for Containerized ApplicationsIEEE Access10.1109/ACCESS.2022.321498510(109768-109778)Online publication date: 2022
  • (2022)Impact of Resource Millibottlenecks on Large-Scale Time Fluctuations in Spark SQL2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)10.1109/ACAIT56212.2022.10137814(1-6)Online publication date: 9-Dec-2022
  • Show More Cited By

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
ICAC '12: Proceedings of the 9th international conference on Autonomic computing
September 2012
222 pages
ISBN:9781450315203
DOI:10.1145/2371536
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

In-Cooperation

  • IEEE

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 18 September 2012

Permissions

Request permissions for this article.

Check for updates

Author Tags

  1. burstiness
  2. n-tier system
  3. performance evaluation
  4. soft resources

Qualifiers

  • Research-article

Conference

ICAC '12
Sponsor:
ICAC '12: 9th International Conference on Autonomic Computing
September 18 - 20, 2012
California, San Jose, USA

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)4
  • Downloads (Last 6 weeks)0
Reflects downloads up to 17 Feb 2025

Other Metrics

Citations

Cited By

View all
  • (2024)A Study of Response Time Instability of Microservices at High Resource Utilization in the Cloud2024 IEEE 6th International Conference on Cognitive Machine Intelligence (CogMI)10.1109/CogMI62246.2024.00024(111-116)Online publication date: 28-Oct-2024
  • (2022)Predictive Hybrid Autoscaling for Containerized ApplicationsIEEE Access10.1109/ACCESS.2022.321498510(109768-109778)Online publication date: 2022
  • (2022)Impact of Resource Millibottlenecks on Large-Scale Time Fluctuations in Spark SQL2022 6th Asian Conference on Artificial Intelligence Technology (ACAIT)10.1109/ACAIT56212.2022.10137814(1-6)Online publication date: 9-Dec-2022
  • (2022)A bi-metric autoscaling approach for n-tier web applications on kubernetesFrontiers of Computer Science: Selected Publications from Chinese Universities10.1007/s11704-021-0118-116:3Online publication date: 1-Jun-2022
  • (2021)SQLR: Short-Term Memory Q-Learning for Elastic ProvisioningIEEE Transactions on Network and Service Management10.1109/TNSM.2021.307561918:2(1850-1869)Online publication date: Jun-2021
  • (2021)PerfML: Smart Management of Complex Performance Data and Analytics2021 IEEE Third International Conference on Cognitive Machine Intelligence (CogMI)10.1109/CogMI52975.2021.00027(146-155)Online publication date: Dec-2021
  • (2020)Finding Performance Patterns from Logs with High ConfidenceWeb Services – ICWS 202010.1007/978-3-030-59618-7_11(164-178)Online publication date: 19-Sep-2020
  • (2019)Quality-Elasticity: Improved Resource Utilization, Throughput, and Response Times Via Adjusting Output Quality to Current Operating Conditions2019 IEEE International Conference on Autonomic Computing (ICAC)10.1109/ICAC.2019.00017(52-62)Online publication date: Jun-2019
  • (2019)Non-linear analysis of bursty workloads using dual metrics for better cloud resource managementJournal of Ambient Intelligence and Humanized Computing10.1007/s12652-019-01183-8Online publication date: 8-Jan-2019
  • (2019)Systematic Construction, Execution, and Reproduction of Complex Performance BenchmarksCloud Computing – CLOUD 201910.1007/978-3-030-23502-4_3(26-37)Online publication date: 25-Jun-2019
  • Show More Cited By

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media