Building intelligence for software defined data centers

Modeling usage patterns

Eric W D Rozier, Pin Zhou, Dwight Divine

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

As both the amount of data to be stored and the rate of data production grows, data center designers and operators face the challenge of planning and managing systems whose characteristics, workloads, and performance, availability, and reliability goals change rapidly. As we move towards software-defined data centers (SDDCs) the ability to reconfigure and adapt our solutions is increasing, but to take full advantage of that increase we must design smarter, more intelligent systems that are aware of how they are being used and able to deliver accurate predictions of their characteristics, workloads, and goals. In this paper we propose a novel algorithm for use in an intelligent, user-aware SDDC which performs run-time analysis of user storage system activity in a manner that has a minimal impact on performance and provides accurate estimations of future user activity. Our algorithm can produce both generalized models, and specific models, depending on the parameters used. Our algorithms are efficient, and have low overhead, making them ideal to use to add intelligence to SDDCs and build intelligent storage systems. We use our algorithm to analyze actual data from two real systems, monitoring user activity two and three times a day for each system respectively, over a period of roughly two years, for almost 500 distinct users.

Original languageEnglish
Title of host publicationACM International Conference Proceeding Series
DOIs
StatePublished - Aug 26 2013
Event6th Annual International Systems and Storage Conference, SYSTOR 2013 - Haifa, Israel
Duration: Jun 30 2013Jul 2 2013

Other

Other6th Annual International Systems and Storage Conference, SYSTOR 2013
CountryIsrael
CityHaifa
Period6/30/137/2/13

Fingerprint

Intelligent systems
Availability
Planning
Monitoring

Keywords

  • Software-defined data centers
  • User modeling
  • Workload modeling

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Rozier, E. W. D., Zhou, P., & Divine, D. (2013). Building intelligence for software defined data centers: Modeling usage patterns. In ACM International Conference Proceeding Series [20] https://doi.org/10.1145/2485732.2485752

Building intelligence for software defined data centers : Modeling usage patterns. / Rozier, Eric W D; Zhou, Pin; Divine, Dwight.

ACM International Conference Proceeding Series. 2013. 20.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rozier, EWD, Zhou, P & Divine, D 2013, Building intelligence for software defined data centers: Modeling usage patterns. in ACM International Conference Proceeding Series., 20, 6th Annual International Systems and Storage Conference, SYSTOR 2013, Haifa, Israel, 6/30/13. https://doi.org/10.1145/2485732.2485752
Rozier EWD, Zhou P, Divine D. Building intelligence for software defined data centers: Modeling usage patterns. In ACM International Conference Proceeding Series. 2013. 20 https://doi.org/10.1145/2485732.2485752
Rozier, Eric W D ; Zhou, Pin ; Divine, Dwight. / Building intelligence for software defined data centers : Modeling usage patterns. ACM International Conference Proceeding Series. 2013.
@inproceedings{ead189af96ac46d3b58dd3ed38dd25e2,
title = "Building intelligence for software defined data centers: Modeling usage patterns",
abstract = "As both the amount of data to be stored and the rate of data production grows, data center designers and operators face the challenge of planning and managing systems whose characteristics, workloads, and performance, availability, and reliability goals change rapidly. As we move towards software-defined data centers (SDDCs) the ability to reconfigure and adapt our solutions is increasing, but to take full advantage of that increase we must design smarter, more intelligent systems that are aware of how they are being used and able to deliver accurate predictions of their characteristics, workloads, and goals. In this paper we propose a novel algorithm for use in an intelligent, user-aware SDDC which performs run-time analysis of user storage system activity in a manner that has a minimal impact on performance and provides accurate estimations of future user activity. Our algorithm can produce both generalized models, and specific models, depending on the parameters used. Our algorithms are efficient, and have low overhead, making them ideal to use to add intelligence to SDDCs and build intelligent storage systems. We use our algorithm to analyze actual data from two real systems, monitoring user activity two and three times a day for each system respectively, over a period of roughly two years, for almost 500 distinct users.",
keywords = "Software-defined data centers, User modeling, Workload modeling",
author = "Rozier, {Eric W D} and Pin Zhou and Dwight Divine",
year = "2013",
month = "8",
day = "26",
doi = "10.1145/2485732.2485752",
language = "English",
isbn = "9781450321167",
booktitle = "ACM International Conference Proceeding Series",

}

TY - GEN

T1 - Building intelligence for software defined data centers

T2 - Modeling usage patterns

AU - Rozier, Eric W D

AU - Zhou, Pin

AU - Divine, Dwight

PY - 2013/8/26

Y1 - 2013/8/26

N2 - As both the amount of data to be stored and the rate of data production grows, data center designers and operators face the challenge of planning and managing systems whose characteristics, workloads, and performance, availability, and reliability goals change rapidly. As we move towards software-defined data centers (SDDCs) the ability to reconfigure and adapt our solutions is increasing, but to take full advantage of that increase we must design smarter, more intelligent systems that are aware of how they are being used and able to deliver accurate predictions of their characteristics, workloads, and goals. In this paper we propose a novel algorithm for use in an intelligent, user-aware SDDC which performs run-time analysis of user storage system activity in a manner that has a minimal impact on performance and provides accurate estimations of future user activity. Our algorithm can produce both generalized models, and specific models, depending on the parameters used. Our algorithms are efficient, and have low overhead, making them ideal to use to add intelligence to SDDCs and build intelligent storage systems. We use our algorithm to analyze actual data from two real systems, monitoring user activity two and three times a day for each system respectively, over a period of roughly two years, for almost 500 distinct users.

AB - As both the amount of data to be stored and the rate of data production grows, data center designers and operators face the challenge of planning and managing systems whose characteristics, workloads, and performance, availability, and reliability goals change rapidly. As we move towards software-defined data centers (SDDCs) the ability to reconfigure and adapt our solutions is increasing, but to take full advantage of that increase we must design smarter, more intelligent systems that are aware of how they are being used and able to deliver accurate predictions of their characteristics, workloads, and goals. In this paper we propose a novel algorithm for use in an intelligent, user-aware SDDC which performs run-time analysis of user storage system activity in a manner that has a minimal impact on performance and provides accurate estimations of future user activity. Our algorithm can produce both generalized models, and specific models, depending on the parameters used. Our algorithms are efficient, and have low overhead, making them ideal to use to add intelligence to SDDCs and build intelligent storage systems. We use our algorithm to analyze actual data from two real systems, monitoring user activity two and three times a day for each system respectively, over a period of roughly two years, for almost 500 distinct users.

KW - Software-defined data centers

KW - User modeling

KW - Workload modeling

UR - http://www.scopus.com/inward/record.url?scp=84882409318&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882409318&partnerID=8YFLogxK

U2 - 10.1145/2485732.2485752

DO - 10.1145/2485732.2485752

M3 - Conference contribution

SN - 9781450321167

BT - ACM International Conference Proceeding Series

ER -