View in EDS HTML Full Text PDF Full Text

Cooperative CPU, GPU, and FPGA heterogeneous execution with EngineCL.

Saved in:
Bibliographic Details
Title: Cooperative CPU, GPU, and FPGA heterogeneous execution with EngineCL.
Authors: Dávila Guzmán, María Angélica1, Gran Tejero, Rubén1, Villarroya-Gaudó, María1, Suárez Gracia, Darío1, Nozal, Raúl2, Bosque, Jose Luis2
Source: Journal of Supercomputing. Mar2019, Vol. 75 Issue 3, p1732-1746. 15p.
Subjects: Heterogeneous computing, OpenCL (Computer program language), Computer scheduling, Field programmable gate arrays, Graphics processing units, Central processing units, Load balancing (Computer networks)
Abstract: Heterogeneous systems are the core architecture of most of the high-performance computing nodes, due to their excellent performance and energy efficiency. However, a key challenge that remains is programmability, specifically, releasing the programmer from the burden of managing data and devices with different architectures. To this end, we extend EngineCL to support FPGA devices. Based on OpenCL, EngineCL is a high-level framework providing load balancing among devices. Our proposal fully integrates FPGAs into the framework, enabling effective cooperation between CPU, GPU, and FPGA. With command overlapping and judicious data management, our work improves performance by up to 96% compared with single-device execution and delivers energy-delay gains of up to 37%. In addition, adopting FPGAs does not require programmers to make big changes in their applications because the extensions do not modify the user-facing interface of EngineCL. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Supercomputing 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: 135780637
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Cooperative CPU, GPU, and FPGA heterogeneous execution with EngineCL.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Dávila+Guzmán%2C+María+Angélica%22">Dávila Guzmán, María Angélica</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Gran+Tejero%2C+Rubén%22">Gran Tejero, Rubén</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Villarroya-Gaudó%2C+María%22">Villarroya-Gaudó, María</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Suárez+Gracia%2C+Darío%22">Suárez Gracia, Darío</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Nozal%2C+Raúl%22">Nozal, Raúl</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Bosque%2C+Jose+Luis%22">Bosque, Jose Luis</searchLink><relatesTo>2</relatesTo>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Journal+of+Supercomputing%22">Journal of Supercomputing</searchLink>. Mar2019, Vol. 75 Issue 3, p1732-1746. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Heterogeneous+computing%22">Heterogeneous computing</searchLink><br /><searchLink fieldCode="DE" term="%22OpenCL+%28Computer+program+language%29%22">OpenCL (Computer program language)</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+scheduling%22">Computer scheduling</searchLink><br /><searchLink fieldCode="DE" term="%22Field+programmable+gate+arrays%22">Field programmable gate arrays</searchLink><br /><searchLink fieldCode="DE" term="%22Graphics+processing+units%22">Graphics processing units</searchLink><br /><searchLink fieldCode="DE" term="%22Central+processing+units%22">Central processing units</searchLink><br /><searchLink fieldCode="DE" term="%22Load+balancing+%28Computer+networks%29%22">Load balancing (Computer networks)</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Heterogeneous systems are the core architecture of most of the high-performance computing nodes, due to their excellent performance and energy efficiency. However, a key challenge that remains is programmability, specifically, releasing the programmer from the burden of managing data and devices with different architectures. To this end, we extend EngineCL to support FPGA devices. Based on OpenCL, EngineCL is a high-level framework providing load balancing among devices. Our proposal fully integrates FPGAs into the framework, enabling effective cooperation between CPU, GPU, and FPGA. With command overlapping and judicious data management, our work improves performance by up to 96% compared with single-device execution and delivers energy-delay gains of up to 37%. In addition, adopting FPGAs does not require programmers to make big changes in their applications because the extensions do not modify the user-facing interface of EngineCL. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Journal of Supercomputing 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=135780637
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s11227-019-02768-y
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 1732
    Subjects:
      – SubjectFull: Heterogeneous computing
        Type: general
      – SubjectFull: OpenCL (Computer program language)
        Type: general
      – SubjectFull: Computer scheduling
        Type: general
      – SubjectFull: Field programmable gate arrays
        Type: general
      – SubjectFull: Graphics processing units
        Type: general
      – SubjectFull: Central processing units
        Type: general
      – SubjectFull: Load balancing (Computer networks)
        Type: general
    Titles:
      – TitleFull: Cooperative CPU, GPU, and FPGA heterogeneous execution with EngineCL.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Dávila Guzmán, María Angélica
      – PersonEntity:
          Name:
            NameFull: Gran Tejero, Rubén
      – PersonEntity:
          Name:
            NameFull: Villarroya-Gaudó, María
      – PersonEntity:
          Name:
            NameFull: Suárez Gracia, Darío
      – PersonEntity:
          Name:
            NameFull: Nozal, Raúl
      – PersonEntity:
          Name:
            NameFull: Bosque, Jose Luis
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Text: Mar2019
              Type: published
              Y: 2019
          Identifiers:
            – Type: issn-print
              Value: 09208542
          Numbering:
            – Type: volume
              Value: 75
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Journal of Supercomputing
              Type: main
ResultId 1