Software architecture and system validation of an open, unified model for accelerated multicore computing.

Saved in:
Bibliographic Details
Title: Software architecture and system validation of an open, unified model for accelerated multicore computing.
Authors: Crawford, C. H.1 catcraw@us.ibm.com, Burdick, D. J.2 deanburdick@us.ibm.com, Dale, J. N.1 jndale@us.ibm.com, Ford, E. F.3 egan@us.ibm.com, Mikosh, R. A.1 mikosh@us.ibm.com, Nobles, A.4 aenobles@us.ibm.com, To, V.1 vto@us.ibm.com
Source: IBM Journal of Research & Development. Sep2009, Vol. 53 Issue 5, p6:1-6:12. 12p.
Subjects: IBM Systems Application Architecture, IBM software, International Business Machines Corp., Multicore processors, Hybrid computers (Computer architecture), Software validation, Workload of computer networks, Microprocessor programming
Abstract: For systems that use hardware accelerators to combine multicore and multiprocess technology with libraries and computational kernels, the drawbacks are the complexity of the programming model and the corresponding verification of the software and validation of the system performance capabilities. In this paper, we describe a software approach to utilizing the compute power of the Cell Broadband Enginet processor and a cluster composed of x86-64 and IBM PowerXCelle 8i processors integrated within a single hybrid compute node. We review past approaches to provide motivation for our development of the Data Communication and Synchronization (DaCS) library and Accelerated Library Framework (ALF), which enable developers to create new applications and adapt existing applications to exploit hybrid computing platforms. We follow with examples of porting existing x86-64 processor-based applications to the hybrid cluster platform in order to demonstrate the capabilities of ALF and DaCS and discuss how one application was extended to become a stress and performance test for various system components. Finally, we present the applicability of this programming model, accelerator design, and test architecture to other system architectures, applications, and workload segments. [ABSTRACT FROM AUTHOR]
Copyright of IBM Journal of Research & Development is the property of IBM Corporation/IBM Journals 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: 47838430
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Software architecture and system validation of an open, unified model for accelerated multicore computing.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Crawford%2C+C%2E+H%2E%22">Crawford, C. H.</searchLink><relatesTo>1</relatesTo><i> catcraw@us.ibm.com</i><br /><searchLink fieldCode="AR" term="%22Burdick%2C+D%2E+J%2E%22">Burdick, D. J.</searchLink><relatesTo>2</relatesTo><i> deanburdick@us.ibm.com</i><br /><searchLink fieldCode="AR" term="%22Dale%2C+J%2E+N%2E%22">Dale, J. N.</searchLink><relatesTo>1</relatesTo><i> jndale@us.ibm.com</i><br /><searchLink fieldCode="AR" term="%22Ford%2C+E%2E+F%2E%22">Ford, E. F.</searchLink><relatesTo>3</relatesTo><i> egan@us.ibm.com</i><br /><searchLink fieldCode="AR" term="%22Mikosh%2C+R%2E+A%2E%22">Mikosh, R. A.</searchLink><relatesTo>1</relatesTo><i> mikosh@us.ibm.com</i><br /><searchLink fieldCode="AR" term="%22Nobles%2C+A%2E%22">Nobles, A.</searchLink><relatesTo>4</relatesTo><i> aenobles@us.ibm.com</i><br /><searchLink fieldCode="AR" term="%22To%2C+V%2E%22">To, V.</searchLink><relatesTo>1</relatesTo><i> vto@us.ibm.com</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22IBM+Journal+of+Research+%26+Development%22">IBM Journal of Research & Development</searchLink>. Sep2009, Vol. 53 Issue 5, p6:1-6:12. 12p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22IBM+Systems+Application+Architecture%22">IBM Systems Application Architecture</searchLink><br /><searchLink fieldCode="DE" term="%22IBM+software%22">IBM software</searchLink><br /><searchLink fieldCode="DE" term="%22International+Business+Machines+Corp%2E%22">International Business Machines Corp.</searchLink><br /><searchLink fieldCode="DE" term="%22Multicore+processors%22">Multicore processors</searchLink><br /><searchLink fieldCode="DE" term="%22Hybrid+computers+%28Computer+architecture%29%22">Hybrid computers (Computer architecture)</searchLink><br /><searchLink fieldCode="DE" term="%22Software+validation%22">Software validation</searchLink><br /><searchLink fieldCode="DE" term="%22Workload+of+computer+networks%22">Workload of computer networks</searchLink><br /><searchLink fieldCode="DE" term="%22Microprocessor+programming%22">Microprocessor programming</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: For systems that use hardware accelerators to combine multicore and multiprocess technology with libraries and computational kernels, the drawbacks are the complexity of the programming model and the corresponding verification of the software and validation of the system performance capabilities. In this paper, we describe a software approach to utilizing the compute power of the Cell Broadband Enginet processor and a cluster composed of x86-64 and IBM PowerXCelle 8i processors integrated within a single hybrid compute node. We review past approaches to provide motivation for our development of the Data Communication and Synchronization (DaCS) library and Accelerated Library Framework (ALF), which enable developers to create new applications and adapt existing applications to exploit hybrid computing platforms. We follow with examples of porting existing x86-64 processor-based applications to the hybrid cluster platform in order to demonstrate the capabilities of ALF and DaCS and discuss how one application was extended to become a stress and performance test for various system components. Finally, we present the applicability of this programming model, accelerator design, and test architecture to other system architectures, applications, and workload segments. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of IBM Journal of Research & Development is the property of IBM Corporation/IBM Journals 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=47838430
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1147/JRD.2009.5429072
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 12
        StartPage: 6:1
    Subjects:
      – SubjectFull: IBM Systems Application Architecture
        Type: general
      – SubjectFull: IBM software
        Type: general
      – SubjectFull: International Business Machines Corp.
        Type: general
      – SubjectFull: Multicore processors
        Type: general
      – SubjectFull: Hybrid computers (Computer architecture)
        Type: general
      – SubjectFull: Software validation
        Type: general
      – SubjectFull: Workload of computer networks
        Type: general
      – SubjectFull: Microprocessor programming
        Type: general
    Titles:
      – TitleFull: Software architecture and system validation of an open, unified model for accelerated multicore computing.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Crawford, C. H.
      – PersonEntity:
          Name:
            NameFull: Burdick, D. J.
      – PersonEntity:
          Name:
            NameFull: Dale, J. N.
      – PersonEntity:
          Name:
            NameFull: Ford, E. F.
      – PersonEntity:
          Name:
            NameFull: Mikosh, R. A.
      – PersonEntity:
          Name:
            NameFull: Nobles, A.
      – PersonEntity:
          Name:
            NameFull: To, V.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 09
              Text: Sep2009
              Type: published
              Y: 2009
          Identifiers:
            – Type: issn-print
              Value: 00188646
          Numbering:
            – Type: volume
              Value: 53
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: IBM Journal of Research & Development
              Type: main
ResultId 1