View in EDS HTML Full Text PDF Full Text

SWIMM 2.0: Enhanced Smith-Waterman on Intel's Multicore and Manycore Architectures Based on AVX-512 Vector Extensions.

Saved in:
Bibliographic Details
Title: SWIMM 2.0: Enhanced Smith-Waterman on Intel's Multicore and Manycore Architectures Based on AVX-512 Vector Extensions.
Authors: Rucci, Enzo1, Giusti, Armando De1, Garcia Sanchez, Carlos2, Botella Juan, Guillermo2, Prieto-Matias, Manuel2, Naiouf, Marcelo3
Source: International Journal of Parallel Programming. Apr2019, Vol. 47 Issue 2, p296-316. 21p.
Subjects: Bioinformatics, Array processors, Intel Corp., SIMD (Computer architecture), Amino acid sequence
Abstract: The well-known Smith-Waterman (SW) algorithm is the most commonly used method for local sequence alignments, but its acceptance is limited by the computational requirements for large protein databases. Although the acceleration of SW has already been studied on many parallel platforms, there are hardly any studies which take advantage of the latest Intel architectures based on AVX-512 vector extensions. This SIMD set is currently supported by Intel's Knights Landing (KNL) accelerator and Intel's Skylake (SKL) general purpose processors. In this paper, we present an SW version that is optimized for both architectures: the renowned SWIMM 2.0. The novelty of this vector instruction set requires the revision of previous programming and optimization techniques. SWIMM 2.0 is based on a massive multi-threading and SIMD exploitation. It is competitive in terms of performance compared with other state-of-the-art implementations, reaching 511 GCUPS on a single KNL node and 734 GCUPS on a server equipped with a dual SKL processor. Moreover, these successful performance rates make SWIMM 2.0 the most efficient energy footprint implementation in this study achieving 2.94 GCUPS/Watts on the SKL processor. [ABSTRACT FROM AUTHOR]
Copyright of International Journal of Parallel Programming 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: 135451502
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: SWIMM 2.0: Enhanced Smith-Waterman on Intel's Multicore and Manycore Architectures Based on AVX-512 Vector Extensions.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Rucci%2C+Enzo%22">Rucci, Enzo</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Giusti%2C+Armando+De%22">Giusti, Armando De</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Garcia+Sanchez%2C+Carlos%22">Garcia Sanchez, Carlos</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Botella+Juan%2C+Guillermo%22">Botella Juan, Guillermo</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Prieto-Matias%2C+Manuel%22">Prieto-Matias, Manuel</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Naiouf%2C+Marcelo%22">Naiouf, Marcelo</searchLink><relatesTo>3</relatesTo>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22International+Journal+of+Parallel+Programming%22">International Journal of Parallel Programming</searchLink>. Apr2019, Vol. 47 Issue 2, p296-316. 21p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Bioinformatics%22">Bioinformatics</searchLink><br /><searchLink fieldCode="DE" term="%22Array+processors%22">Array processors</searchLink><br /><searchLink fieldCode="DE" term="%22Intel+Corp%2E%22">Intel Corp.</searchLink><br /><searchLink fieldCode="DE" term="%22SIMD+%28Computer+architecture%29%22">SIMD (Computer architecture)</searchLink><br /><searchLink fieldCode="DE" term="%22Amino+acid+sequence%22">Amino acid sequence</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The well-known Smith-Waterman (SW) algorithm is the most commonly used method for local sequence alignments, but its acceptance is limited by the computational requirements for large protein databases. Although the acceleration of SW has already been studied on many parallel platforms, there are hardly any studies which take advantage of the latest Intel architectures based on AVX-512 vector extensions. This SIMD set is currently supported by Intel's Knights Landing (KNL) accelerator and Intel's Skylake (SKL) general purpose processors. In this paper, we present an SW version that is optimized for both architectures: the renowned SWIMM 2.0. The novelty of this vector instruction set requires the revision of previous programming and optimization techniques. SWIMM 2.0 is based on a massive multi-threading and SIMD exploitation. It is competitive in terms of performance compared with other state-of-the-art implementations, reaching 511 GCUPS on a single KNL node and 734 GCUPS on a server equipped with a dual SKL processor. Moreover, these successful performance rates make SWIMM 2.0 the most efficient energy footprint implementation in this study achieving 2.94 GCUPS/Watts on the SKL processor. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of International Journal of Parallel Programming 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=135451502
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10766-018-0585-7
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 21
        StartPage: 296
    Subjects:
      – SubjectFull: Bioinformatics
        Type: general
      – SubjectFull: Array processors
        Type: general
      – SubjectFull: Intel Corp.
        Type: general
      – SubjectFull: SIMD (Computer architecture)
        Type: general
      – SubjectFull: Amino acid sequence
        Type: general
    Titles:
      – TitleFull: SWIMM 2.0: Enhanced Smith-Waterman on Intel's Multicore and Manycore Architectures Based on AVX-512 Vector Extensions.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Rucci, Enzo
      – PersonEntity:
          Name:
            NameFull: Giusti, Armando De
      – PersonEntity:
          Name:
            NameFull: Garcia Sanchez, Carlos
      – PersonEntity:
          Name:
            NameFull: Botella Juan, Guillermo
      – PersonEntity:
          Name:
            NameFull: Prieto-Matias, Manuel
      – PersonEntity:
          Name:
            NameFull: Naiouf, Marcelo
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 04
              Text: Apr2019
              Type: published
              Y: 2019
          Identifiers:
            – Type: issn-print
              Value: 08857458
          Numbering:
            – Type: volume
              Value: 47
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: International Journal of Parallel Programming
              Type: main
ResultId 1