View in EDS

Evaluating and extending user-level fault tolerance in MPI applications.

Saved in:
Bibliographic Details
Title: Evaluating and extending user-level fault tolerance in MPI applications.
Authors: Laguna, Ignacio1 ilaguna@llnl.gov, Richards, David F.1, Gamblin, Todd1, Schulz, Martin1, de Supinski, Bronis R.1, Mohror, Kathryn1, Pritchard, Howard2
Source: International Journal of High Performance Computing Applications. Fall2016, Vol. 30 Issue 3, p305-319. 15p.
Subjects: Fault tolerance (Engineering), Message passing (Computer science), Performance evaluation, User interfaces, Molecular dynamics
Abstract: The user-level failure mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in the Message Passing Interface (MPI). Previous work presented performance evaluations of ULFM; yet questions related to its programability and applicability, especially to non-trivial, bulk synchronous applications, remain unanswered. In this article, we present our experiences on using ULFM in a case study with a large, highly scalable, bulk synchronous molecular dynamics application to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for master–worker applications, it provides few benefits for more common bulk synchronous MPI applications. To address these limitations, we introduce a new, simpler fault-tolerant interface for complex, bulk synchronous MPI programs with better applicability and support than ULFM for application-level recovery mechanisms, such as global rollback. [ABSTRACT FROM AUTHOR]
Copyright of International Journal of High Performance Computing Applications is the property of Sage Publications Inc. 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
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 117638081
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Evaluating and extending user-level fault tolerance in MPI applications.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Laguna%2C+Ignacio%22">Laguna, Ignacio</searchLink><relatesTo>1</relatesTo><i> ilaguna@llnl.gov</i><br /><searchLink fieldCode="AR" term="%22Richards%2C+David+F%2E%22">Richards, David F.</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Gamblin%2C+Todd%22">Gamblin, Todd</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Schulz%2C+Martin%22">Schulz, Martin</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22de+Supinski%2C+Bronis+R%2E%22">de Supinski, Bronis R.</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Mohror%2C+Kathryn%22">Mohror, Kathryn</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Pritchard%2C+Howard%22">Pritchard, Howard</searchLink><relatesTo>2</relatesTo>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22International+Journal+of+High+Performance+Computing+Applications%22">International Journal of High Performance Computing Applications</searchLink>. Fall2016, Vol. 30 Issue 3, p305-319. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Fault+tolerance+%28Engineering%29%22">Fault tolerance (Engineering)</searchLink><br /><searchLink fieldCode="DE" term="%22Message+passing+%28Computer+science%29%22">Message passing (Computer science)</searchLink><br /><searchLink fieldCode="DE" term="%22Performance+evaluation%22">Performance evaluation</searchLink><br /><searchLink fieldCode="DE" term="%22User+interfaces%22">User interfaces</searchLink><br /><searchLink fieldCode="DE" term="%22Molecular+dynamics%22">Molecular dynamics</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The user-level failure mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in the Message Passing Interface (MPI). Previous work presented performance evaluations of ULFM; yet questions related to its programability and applicability, especially to non-trivial, bulk synchronous applications, remain unanswered. In this article, we present our experiences on using ULFM in a case study with a large, highly scalable, bulk synchronous molecular dynamics application to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for master–worker applications, it provides few benefits for more common bulk synchronous MPI applications. To address these limitations, we introduce a new, simpler fault-tolerant interface for complex, bulk synchronous MPI programs with better applicability and support than ULFM for application-level recovery mechanisms, such as global rollback. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of International Journal of High Performance Computing Applications is the property of Sage Publications Inc. 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=117638081
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1177/1094342015623623
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 305
    Subjects:
      – SubjectFull: Fault tolerance (Engineering)
        Type: general
      – SubjectFull: Message passing (Computer science)
        Type: general
      – SubjectFull: Performance evaluation
        Type: general
      – SubjectFull: User interfaces
        Type: general
      – SubjectFull: Molecular dynamics
        Type: general
    Titles:
      – TitleFull: Evaluating and extending user-level fault tolerance in MPI applications.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Laguna, Ignacio
      – PersonEntity:
          Name:
            NameFull: Richards, David F.
      – PersonEntity:
          Name:
            NameFull: Gamblin, Todd
      – PersonEntity:
          Name:
            NameFull: Schulz, Martin
      – PersonEntity:
          Name:
            NameFull: de Supinski, Bronis R.
      – PersonEntity:
          Name:
            NameFull: Mohror, Kathryn
      – PersonEntity:
          Name:
            NameFull: Pritchard, Howard
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 09
              Text: Fall2016
              Type: published
              Y: 2016
          Identifiers:
            – Type: issn-print
              Value: 10943420
          Numbering:
            – Type: volume
              Value: 30
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: International Journal of High Performance Computing Applications
              Type: main
ResultId 1