View in EDS HTML Full Text PDF Full Text

Flow-Aware Workload Migration in Data Centers.

Saved in:
Bibliographic Details
Title: Flow-Aware Workload Migration in Data Centers.
Authors: Desmouceaux, Yoann1,2 yoann.desmouceaux@polytechnique.edu, Toubaline, Sonia3 sonia.toubaline@dauphine.fr, Clausen, Thomas1 thomas.clausen@polytechnique.edu
Source: Journal of Network & Systems Management. Oct2018, Vol. 26 Issue 4, p1034-1057. 24p.
Subjects: Data libraries, Workload of computer networks, Linear programming, Heuristic, Approximation theory, Performance evaluation
Abstract: In data centers, subject to workloads with heterogeneous (and sometimes short) lifetimes, workload migration is a way of attaining a more efficient utilization of the underlying physical machines. To not introduce performance degradation, such workload migration must take into account not only machine resources, and per-task resource requirements, but also application dependencies in terms of network communication. This paper presents a workload migration model capturing all of these constraints. A linear programming framework is developed allowing accurate representation of per-task resources requirements and inter-task network demands. Using this, a multi-objective problem is formulated to compute a re-allocation of tasks that (1) maximizes the total inter-task throughput, while (2) minimizing the cost incurred by migration and (3) allocating the maximum number of new tasks. A baseline algorithm, solving this multi-objective problem using the ε-constraint method is proposed, in order to generate the set of Pareto-optimal solutions. As this algorithm is compute-intensive for large topologies, a heuristic, which computes an approximation of the Pareto front, is then developed, and evaluated on different topologies and with different machine load factors. These evaluations show that the heuristic can provide close-to-optimal solutions, while reducing the solving time by one to two order of magnitudes. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Network & Systems Management is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 131436376
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Flow-Aware Workload Migration in Data Centers.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Desmouceaux%2C+Yoann%22">Desmouceaux, Yoann</searchLink><relatesTo>1,2</relatesTo><i> yoann.desmouceaux@polytechnique.edu</i><br /><searchLink fieldCode="AR" term="%22Toubaline%2C+Sonia%22">Toubaline, Sonia</searchLink><relatesTo>3</relatesTo><i> sonia.toubaline@dauphine.fr</i><br /><searchLink fieldCode="AR" term="%22Clausen%2C+Thomas%22">Clausen, Thomas</searchLink><relatesTo>1</relatesTo><i> thomas.clausen@polytechnique.edu</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Journal+of+Network+%26+Systems+Management%22">Journal of Network & Systems Management</searchLink>. Oct2018, Vol. 26 Issue 4, p1034-1057. 24p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Data+libraries%22">Data libraries</searchLink><br /><searchLink fieldCode="DE" term="%22Workload+of+computer+networks%22">Workload of computer networks</searchLink><br /><searchLink fieldCode="DE" term="%22Linear+programming%22">Linear programming</searchLink><br /><searchLink fieldCode="DE" term="%22Heuristic%22">Heuristic</searchLink><br /><searchLink fieldCode="DE" term="%22Approximation+theory%22">Approximation theory</searchLink><br /><searchLink fieldCode="DE" term="%22Performance+evaluation%22">Performance evaluation</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: In data centers, subject to workloads with heterogeneous (and sometimes short) lifetimes, workload migration is a way of attaining a more efficient utilization of the underlying physical machines. To not introduce performance degradation, such workload migration must take into account not only machine resources, and per-task resource requirements, but also application dependencies in terms of network communication. This paper presents a workload migration model capturing all of these constraints. A linear programming framework is developed allowing accurate representation of per-task resources requirements and inter-task network demands. Using this, a multi-objective problem is formulated to compute a re-allocation of tasks that (1) maximizes the total inter-task throughput, while (2) minimizing the cost incurred by migration and (3) allocating the maximum number of new tasks. A baseline algorithm, solving this multi-objective problem using the ε<inline-graphic></inline-graphic>-constraint method is proposed, in order to generate the set of Pareto-optimal solutions. As this algorithm is compute-intensive for large topologies, a heuristic, which computes an approximation of the Pareto front, is then developed, and evaluated on different topologies and with different machine load factors. These evaluations show that the heuristic can provide close-to-optimal solutions, while reducing the solving time by one to two order of magnitudes. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Journal of Network & Systems Management is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=131436376
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10922-018-9452-5
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 24
        StartPage: 1034
    Subjects:
      – SubjectFull: Data libraries
        Type: general
      – SubjectFull: Workload of computer networks
        Type: general
      – SubjectFull: Linear programming
        Type: general
      – SubjectFull: Heuristic
        Type: general
      – SubjectFull: Approximation theory
        Type: general
      – SubjectFull: Performance evaluation
        Type: general
    Titles:
      – TitleFull: Flow-Aware Workload Migration in Data Centers.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Desmouceaux, Yoann
      – PersonEntity:
          Name:
            NameFull: Toubaline, Sonia
      – PersonEntity:
          Name:
            NameFull: Clausen, Thomas
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 10
              Text: Oct2018
              Type: published
              Y: 2018
          Identifiers:
            – Type: issn-print
              Value: 10647570
          Numbering:
            – Type: volume
              Value: 26
            – Type: issue
              Value: 4
          Titles:
            – TitleFull: Journal of Network & Systems Management
              Type: main
ResultId 1