DodOrg—A Self-adaptive Organic Many-core Architecture

  • Thomas Ebi
  • David Kramer
  • Christian Schuck
  • Alexander von Renteln
  • Jürgen Becker
  • Uwe Brinkschulte
  • Jörg Henkel
  • Wolfgang Karl
Part of the Autonomic Systems book series (ASYS, volume 1)

Abstract

The Digital on-Demand Computing Organism (DodOrg) project addresses a new computer architecture principle. The complexity of today’s computing systems and the dynamics which arise from the interaction of various subsystems with each other as well as with the environment, restrain an efficient optimisation and maintenance of such systems.

DodOrg is a novel, biologically inspired, heterogeneous, and adaptive computer architecture, that features self-x properties in order to ease management and optimisation. Multiple interleaved control loops, that span all system layers, are employed for realisation of these self-x properties, such as self-optimisation. A dedicated monitoring infrastructure provides the basis for these control loops and realises the basic property of self-awareness. The modular architecture of Organic Processing Cells (OPC) provide a flexible hardware infrastructure with reconfiguration capabilities, that are essential for realisation of adaptive systems. Distributed task mapping is achieved through an Artificial Hormone System (AHS) within the organic middleware. The organic thermal management increases the system’s mean time to failure by balancing the thermal distribution between OPCs.

Keywords

Self-adaptive computing Self-x Distributed middleware Monitoring Thermal management Reconfigurable computing 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Al Faruque, M., Jahn, J., Ebi, T., Henkel, J.: Runtime thermal management using software agents for multi- and many-core architectures. IEEE Des. Test Comput. 27(6), 58–68 (2010) CrossRefGoogle Scholar
  2. 2.
    Borkar, S.: Thousand core chips: a technology perspective. In: Proceedings of the 44th Annual Design Automation Conference (DAC), pp. 746–749 (2007) CrossRefGoogle Scholar
  3. 3.
    Brinkschulte, U., Pacher, M., von Renteln, A.: Towards an artificial hormone system for self-organizing real-time task allocation. In: 5th IFIP Workshop on Software Technologies for Future Embedded & Ubiquitous Systems (SEUS) (2007) Google Scholar
  4. 4.
    Brinkschulte, U., Pacher, M., von Renteln, A.: An artificial hormone system for self-organizing real-time task allocation. In: Organic Computing. Springer, Berlin (2008) Google Scholar
  5. 5.
    Brinkschulte, U., von Renteln, A.: Analyzing the behaviour of an artificial hormone system for task allocation. In: ATC, Brisbane, QLD, Australia (2009) Google Scholar
  6. 6.
    Brinkschulte, U., von Renteln, A., Pacher, M.: Measuring the quality of an artificial hormone system based task mapping. In: ACM Autonomics, Turin, Italy (2008) Google Scholar
  7. 7.
    Buchty, R., Kramer, D., Karl, W.: An Organic Computing Approach to Sustained Real-time Monitoring. In: Proceedings of WCC2008/BICC. IFIP, vol. 268, pp. 151–162. Springer, Berlin (2008) Google Scholar
  8. 8.
    Coskun, A.K., Rosing, T.S., Gross, K.C.: Temperature management in multiprocessor socs using online learning. In: Proceedings of the 45th Annual Design Automation Conference (DAC), pp. 890–893 (2008) CrossRefGoogle Scholar
  9. 9.
    Ebi, T., Al Faruque, M.A., Henkel, J.: TAPE: thermal-aware agent-based power economy for multi/many-core architectures. In: International Conference on Computer-Aided Design (ICCAD), pp. 302–309 (2009). Best Paper Award Google Scholar
  10. 10.
    Ebi, T., Al Faruque, M.A., Henkel, J.: NeuroNoC: neural network inspired runtime adaptation for an on-chip communication architecture. In: Proceedings of the Eighth IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis. CODES/ ISSS’10, pp. 223–230. ACM, New York (2010) CrossRefGoogle Scholar
  11. 11.
    Schuck, C., Haetzer, B., Becker, J.: An interface for a decentralized 2D reconfiguration on Xilinx Virtex-FPGAs for organic computing. Int. J. Reconfigurable Comput. 2009, 1–12 (2009) CrossRefGoogle Scholar
  12. 12.
    Schuck, C., Haetzer, B., Becker, J.: Dynamic online reconfiguration of digital clock managers on Xilinx Virtex-II/Virtex II-pro FPGAs: A case study of distributed power management. In: ReCoSoC2010 (2010) Google Scholar
  13. 13.
    Schuck, C., Haetzer, B., Becker, J.: Reconfiguration techniques for self-x power and performance management on Xilinx Virtex-II/Virtex-II-Pro FPGAs. Int. J. Reconfigurable Comput. 2010, 1–17 (2010) Google Scholar
  14. 14.
    Schuck, C., Lamparth, S., Becker, J.: artNoC—A novel multi-functional router architecture for organic computing. In: FPL, pp. 371–376 (2007) Google Scholar
  15. 15.
    Sprunt, B.: The basics of performance-monitoring hardware. In: IEEE Micro, pp. 64–71 (2002) Google Scholar
  16. 16.
    Tesfatsion, L.: Agent-based computational economics: Growing economies from the bottom up. Artif. Life 8(1), 55–82 (2002) MathSciNetCrossRefGoogle Scholar
  17. 17.
    von Renteln, A., Brinkschulte, U.: Reliability of an artificial hormone system with self-x properties. In: Parallel and Distributed Computing and Systems, Cambridge, Massachusetts, USA, 2007 Google Scholar
  18. 18.
    von Renteln, A., Brinkschulte, U.: Implementing and evaluating the AHS organic middleware—a first approach. In: ISORC, Carmona, Spain, 2010 Google Scholar
  19. 19.
    Yeo, I., Liu, C.C., Kim, E.J.: Predictive dynamic thermal management for multicore systems. In: Proceedings of the 45th Annual Design Automation Conference (DAC), pp. 734–739 (2008) CrossRefGoogle Scholar

Copyright information

© Springer Basel AG 2011

Authors and Affiliations

  • Thomas Ebi
    • 2
  • David Kramer
    • 2
  • Christian Schuck
    • 3
  • Alexander von Renteln
    • 1
  • Jürgen Becker
    • 3
  • Uwe Brinkschulte
    • 1
  • Jörg Henkel
    • 2
  • Wolfgang Karl
    • 2
  1. 1.Institute of Computer ScienceUniversity of FrankfurtFrankfurtGermany
  2. 2.Institute for Computer Science and EngineeringKarlsruhe Institute of TechnologyKarlsruheGermany
  3. 3.Institute for Information Processing TechnologiesKarlsruhe Institute of TechnologyKarlsruheGermany

Personalised recommendations