Advertisement

ATOLL, a new switched, high speed Interconnect in Comparison to Myrinet and SCI

  • Markus Fischer
  • Ulrich Brüning
  • Jörg Kluge
  • Lars Rzymianowicz
  • Patrick Schulz
  • Mathias Waack
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1800)

Abstract

While standard processors achieve supercomputer performance, a performance gap exists between the interconnect of MPP’s and COTS. Standard solutions like Ethernet can not keep up with the demand for high speed communication of todays powerful CPU’s. Hence, high speed interconnects have an important impact on a cluster’s performance. While standard solutions for processing nodes exist, communication hardware is currently only available as a special, expensive non portable solution. ATOLL presents a switched, high speed interconnect, which fulfills the current needs for user level communication and concurrency in computation and communication. A TOLL is a single chip solution, additional switching hardware is not required.

Keywords

Network Interface Remote Memory Host Memory Workshop Cluster Bisection Bandwidth 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Warschko, Blum and Tichy. On the Design and Semantics of User-Space Communication Subsystems, PDPTA 99, Las Vegas, Nevada.Google Scholar
  2. [2]
    Santos, Bianchini and Amorim. A Survey of Messaging Software Issues and System on Myrinet Based Clusters Google Scholar
  3. [3]
    IEEE 345, 47th Street New York. IEEE Standard for Scalable Coherent Interface (SCI), 1993Google Scholar
  4. [4]
    O’Caroll, Tezuka, Hori and Ishikawa. The Design and Implementation of Zero Copy MPI using..., In International Conference on Supercomputing’ 98, pages 243–250, July 1998.Google Scholar
  5. [5]
    Rzymianowicz, Bruening, Kluge, Schulz and Waack. Atoll, A Network on a Chip, PDPTA 99, Las Vegas, NevadaGoogle Scholar
  6. [6]
  7. [7]
    Kay and Pasquale. Profiling and Tuning Processing overheads in TCP/IP. IEEE/ACM Transactions on Networking, Dec. 1996Google Scholar
  8. [8]
    Warschko. Efficient Communication in Parallel Computers. PhD thesis, University of Karlsruhe, 1998Google Scholar
  9. [9]
    Tezuka, O’Caroll, Hori, and Ishikawa. Pin-down Cache: A Virtual Memory Management Technique for Zero-copy Communication, IPPS98, pages 308–314, 1998Google Scholar
  10. [10]
    Prylli and Tourancheau. BIP: A new protocol designed for high performance networking on Myrinet, PCNOWWorkshop, IPPS/SPDP98, 1998Google Scholar
  11. [11]
    O’Carroll, Tezukua, Hori, and Ishikawa. MPICH-PM: Design and Implementation of Zero Copy MPI for PM, Technical Report TR-97011, RWC, March 1998.Google Scholar
  12. [12]
  13. [13]
    Scholtyssik and Dormanns. Simplifying the use of SCI shared memory by using software SVM techniques, 2nd Workshop Cluster Computing, Karlsruhe, 1999Google Scholar
  14. [14]
    Mukherjee and Hill. The Impact of Data Transfer and Buffering Alternatives on Network Interface Design, HPCA98, Feb. 1998Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Markus Fischer
    • 1
  • Ulrich Brüning
    • 1
  • Jörg Kluge
    • 1
  • Lars Rzymianowicz
    • 1
  • Patrick Schulz
    • 1
  • Mathias Waack
    • 1
  1. 1.University of MannheimGermany

Personalised recommendations