Advertisement

Towards modeling and integrated design automation of supercomputing clusters (MIDAS)

  • Venkateswaran NagarajanEmail author
  • Aravind Vasudevan
  • Balaji Subramaniam
  • Ravindhiran Mukundrajan
  • T. P. Ramnath Sai Sagar
  • Madhavan Manivannan
  • Sriram Murali
  • Vinoth Krishnan Elangovan
Special Issue Paper

Abstract

The choice of supercomputers, by the user community, should not merely be based on the benchmarks. With the advent of supercomputers delivering petaops performance, computationally intensive applications such as brain modeling and energy requirement prediction have become predominant. Conventional benchmarks do not reflect the functional characteristics of these applications. Thus, the gap between the performance projected and the actual performance delivered when the application is ported onto the cluster is wide. The primary cause for this disparity in performance is due to the user community’s lack of knowledge about the system and improper design choices. With the supercomputer market being seller biased this problem is worsened. Design and development of very large systems like the clusters involve massive efforts and financial investment. Creating a design automation environment for the clusters is bound to bring down the man years and hence the production cost. Currently there are adhoc approaches which lack an integrated design space to cater to the wider needs of the user community. Thus, an integrated design framework that takes all the components of the supercomputing cluster and the relationship across the design spaces is essential. In this paper we propose a novel methodology called the Modeling and Integrated Design Automation of Supercomputing Cluster (MIDAS). MIDAS is the birth of a new design philosophy which requires lot of research and tuning, providing scope for evolution and optimization.

Keywords

Design automation  Supercomputer model 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Venkateswaran N et al (2008) On the concept of simultaneous execution of multiple applications on hierarchically based cluster and the silicon operating system, presented at LSPP Workshop Held in Conjunction with International Parallel and Distributed Processing Symposium (IPDPS’08)Google Scholar
  2. 2.
    Venkateswaran N et al (2008) Towards Node Architecture Designs for Realizing High Productivity Supercomputers, presented at on the Pre-Conference of International Supercomputing Conference (ISC’08)Google Scholar
  3. 3.
    Subramaniam B (2009) Towards Modelling and Design Automation of Supercomputing Clusters: On Chip Networks, A Thesis Submitted to WARFT, http://www.warftindia.org/isc09submission/midas/balajisubramaniam-thesis-09.pdf. Accessed 26 May 2009
  4. 4.
    Vasudevan A (2009) Towards Modelling and Design Automation of Supercomputing Clusters: Hierarchical Compiler On Silicon, A Thesis Submitted to WARFT, http://www.warftindia.org/isc09submission/midas/aravindvasudevan-thesis-09.pdf. Accessed 26 May 2009
  5. 5.
    Mukhundarajan R (2008) Towards Modelling and Design Automation of Supercomputing Clusters: SCOC IP Cores, A Thesis Submitted to WARFT, http://www.warftindia.org/isc09submission/midas/ravindhiran-thesis-09.pdf. Accessed 26 May 2009
  6. 6.
    Petrini F, Feng W-C (2000) Buffered Coscheduling: A New Methodology for Multitasking Parallel Jobs on Distributed Systems, Presented in the proceedings of International Parallel and Distributed Processing Symposium (IPDPS’00)Google Scholar
  7. 7.
    Report on the SEQUOIA Supercomputer (2009) http://www.internetnews.com/hardware/article.php/3800296. Accessed 26 May 2009
  8. 8.
    Tiwari V et al (1998) Reducing Power in High-performance Microprocessors, Proceedings of Design Automation ConferenceGoogle Scholar
  9. 9.
    Venketeswaran N et al (2005) Digital Model for InterConnect Analysis, Proceedings of NORCHIP ConferenceGoogle Scholar
  10. 10.
    Venkateswaran N et al (2008) Power Estimation of Higher Level Functional Units in High Performance Heterogeneous Multi-Core Processors, a WARFT White PaperGoogle Scholar
  11. 11.
    Murali S (2008) On Node Network Optimisation for Large Scale Heterogenous Multi-Core Node Architecture, A Thesis submitted to WARFT, http://www.warftindia.org/isc09submission/midas/srirammurali-thesis-08.pdf. Accessed 26 May 2009
  12. 12.
    Manivannan M (2007) A Generalized Methodology for On-Node Network Design in Heterogeneous Multi-core Memory in Processor Architectures, A Thesis Submitted to WARFT, http://www.warftindia.org/isc09submission/midas/madhavanmanivannan-thesis-07.pdf. Accessed 26 May 2009
  13. 13.
    Venkateswaran N (2008) Towards Node Architecture Designs for Realizing High Productivity Supercomputers, presented at on the Pre-Conference of International Supercomputing Conference (ISC’08), under review for the Concurrency and Computation, Wiley Interscience Journal. http://www.warftindia.org/isc09submission/midas/revisedisc.pdf. Accessed 26 May 2009

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • Venkateswaran Nagarajan
    • 1
    Email author
  • Aravind Vasudevan
    • 1
  • Balaji Subramaniam
    • 1
  • Ravindhiran Mukundrajan
    • 1
  • T. P. Ramnath Sai Sagar
    • 2
  • Madhavan Manivannan
    • 4
  • Sriram Murali
    • 3
  • Vinoth Krishnan Elangovan
    • 4
  1. 1.Waran Research Foundation (WARFT)ChennaiIndia
  2. 2.Barcelona Supercomputing CenterBarcelonaSpain
  3. 3.University of British ColumbiaVancouverCanada
  4. 4.TU DelftDelftThe Netherlands

Personalised recommendations