An RDMA QP communication mechanism of next-generation intelligent computing center.

Saved in:
Bibliographic Details
Title: An RDMA QP communication mechanism of next-generation intelligent computing center.
Authors: WANG, Junliang1 wangjl23@chinatelecom.cn, LIN, Baohong1 linbaoh@chinatelecom.cn, ZHANG, Jiao2 jiaozhang@bupt.edu.cn, SUN, Mengyu3 sunmy@chinatelecom.cn, PAN, Yongchen2 ychpan@bupt.edu.cn
Source: Computer Engineering & Science / Jisuanji Gongcheng yu Kexue. Feb2026, Vol. 48 Issue 2, p228-237. 10p.
Subjects: Network interface devices, Computer network protocols, Digital communications, Computer network reliability, Data transmission systems, Machine-to-machine communications
Abstract: Currently, intelligent computing centers primarily employ RDMA (remote direct memory access) protocol to achieve ultra-high-performance communication within clusters, where each pair of processes needs to establish a queue pair (QP) based on the reliable connection (RC) type. In the context of AI large model scenarios in next-generation large-scale intelligent computing centers, distributed collective communication operations such as All-to-All and All Reduce will trigger fully connected communication between processes. The number of QPs that need to be maintained under the RC-based mechanism will exceed one million, posing significant challenges to the limited memory and performance of RDMA network interface cards (NICs). To address this issue, an RDMA QP communication mechanism named ERD (efficient reliable datagram) is proposed. On one hand, it replaces traditional RC with RD (reliable datagram) to enhance the scalability of QPs on NICs; on the other hand, it designs an RD-based reliable reception mechanism that incorporates packet loss handling and rapid ordered processing in the network stack, ensuring network reliability while improving transmission performance. Through experiments and NS3 simulation tests, ERD can reduce the number of QPs by 99.96% and enhance transmission performance by over 15% during network congestion [ABSTRACT FROM AUTHOR]
Copyright of Computer Engineering & Science / Jisuanji Gongcheng yu Kexue is the property of Computer Engineering & Science 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 Links:
  – Type: pdflink
Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 192370947
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: An RDMA QP communication mechanism of next-generation intelligent computing center.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22WANG%2C+Junliang%22">WANG, Junliang</searchLink><relatesTo>1</relatesTo><i> wangjl23@chinatelecom.cn</i><br /><searchLink fieldCode="AR" term="%22LIN%2C+Baohong%22">LIN, Baohong</searchLink><relatesTo>1</relatesTo><i> linbaoh@chinatelecom.cn</i><br /><searchLink fieldCode="AR" term="%22ZHANG%2C+Jiao%22">ZHANG, Jiao</searchLink><relatesTo>2</relatesTo><i> jiaozhang@bupt.edu.cn</i><br /><searchLink fieldCode="AR" term="%22SUN%2C+Mengyu%22">SUN, Mengyu</searchLink><relatesTo>3</relatesTo><i> sunmy@chinatelecom.cn</i><br /><searchLink fieldCode="AR" term="%22PAN%2C+Yongchen%22">PAN, Yongchen</searchLink><relatesTo>2</relatesTo><i> ychpan@bupt.edu.cn</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Computer+Engineering+%26+Science+%2F+Jisuanji+Gongcheng+yu+Kexue%22">Computer Engineering & Science / Jisuanji Gongcheng yu Kexue</searchLink>. Feb2026, Vol. 48 Issue 2, p228-237. 10p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Network+interface+devices%22">Network interface devices</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+network+protocols%22">Computer network protocols</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+communications%22">Digital communications</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+network+reliability%22">Computer network reliability</searchLink><br /><searchLink fieldCode="DE" term="%22Data+transmission+systems%22">Data transmission systems</searchLink><br /><searchLink fieldCode="DE" term="%22Machine-to-machine+communications%22">Machine-to-machine communications</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Currently, intelligent computing centers primarily employ RDMA (remote direct memory access) protocol to achieve ultra-high-performance communication within clusters, where each pair of processes needs to establish a queue pair (QP) based on the reliable connection (RC) type. In the context of AI large model scenarios in next-generation large-scale intelligent computing centers, distributed collective communication operations such as All-to-All and All Reduce will trigger fully connected communication between processes. The number of QPs that need to be maintained under the RC-based mechanism will exceed one million, posing significant challenges to the limited memory and performance of RDMA network interface cards (NICs). To address this issue, an RDMA QP communication mechanism named ERD (efficient reliable datagram) is proposed. On one hand, it replaces traditional RC with RD (reliable datagram) to enhance the scalability of QPs on NICs; on the other hand, it designs an RD-based reliable reception mechanism that incorporates packet loss handling and rapid ordered processing in the network stack, ensuring network reliability while improving transmission performance. Through experiments and NS3 simulation tests, ERD can reduce the number of QPs by 99.96% and enhance transmission performance by over 15% during network congestion [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Computer Engineering & Science / Jisuanji Gongcheng yu Kexue is the property of Computer Engineering & Science 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=192370947
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.3969/j.issn.1007-130X.2026.02.004
    Languages:
      – Code: chi
        Text: Chinese
    PhysicalDescription:
      Pagination:
        PageCount: 10
        StartPage: 228
    Subjects:
      – SubjectFull: Network interface devices
        Type: general
      – SubjectFull: Computer network protocols
        Type: general
      – SubjectFull: Digital communications
        Type: general
      – SubjectFull: Computer network reliability
        Type: general
      – SubjectFull: Data transmission systems
        Type: general
      – SubjectFull: Machine-to-machine communications
        Type: general
    Titles:
      – TitleFull: An RDMA QP communication mechanism of next-generation intelligent computing center.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: WANG, Junliang
      – PersonEntity:
          Name:
            NameFull: LIN, Baohong
      – PersonEntity:
          Name:
            NameFull: ZHANG, Jiao
      – PersonEntity:
          Name:
            NameFull: SUN, Mengyu
      – PersonEntity:
          Name:
            NameFull: PAN, Yongchen
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 02
              Text: Feb2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 1007130X
          Numbering:
            – Type: volume
              Value: 48
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Computer Engineering & Science / Jisuanji Gongcheng yu Kexue
              Type: main
ResultId 1