Intelligent NOC Hotspot Prediction

  • Elena Kakoulli
  • Vassos Soteriou
  • Theocharis Theocharides
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 105)


Hotspots are Network on-Chip (NoC) routers or modules which occasionally receive packetized traffic at a higher rate that they can process. This phenomenon reduces the performance of an NoC, especially in the case wormhole flow-control. Such situations may also lead to deadlocks, raising the need of a hotspot prevention mechanism. Such mechanism can potentially enable the system to adjust its behavior and prevent hotspot formation, subsequently sustaining performance and efficiency. This Chapter presents an Artificial Neural Network-based (ANN) hotspot prediction mechanism, potentially triggering a hotspot avoidance mechanism before the hotspot is formed. The ANN monitors buffer utilization and reactively predicts the location of an about to-be-formed hotspot, allowing enough time for the system to react to these potential hotspots. The neural network is trained using synthetic traffic models, and evaluated using both synthetic and real application traces. Results indicate that a relatively small neural network can predict hotspot formation with accuracy ranges between 76 and 92%.


Network on-Chip Hotspots Artificial Neural Networks VLSI Systems 


  1. 1.
    Baydal E et al (2005) A Family of mechanisms for congestion control in wormhole networks. In IEEE TPDS 16(9):772–784 Sept 2005Google Scholar
  2. 2.
    Bell S et al (2008) TILE64 Processor: A 64-Core SoC with mesh interconnect. In: ISSCC, pp 88–598 Feb 2008Google Scholar
  3. 3.
    Bertozzi D, Benini L (2004) Xpipes: A Network-on-Chip architecture for gigascale Systems-on-Chip. In: IEEE Circ Syst 4(2):18–31, Second QuarterGoogle Scholar
  4. 4.
    Bjerregaard T, Mahadevan S (2006) A survey of research and practices of Network-on-Chip. In ACM CSUR 38(1):1–51 March 2006CrossRefGoogle Scholar
  5. 5.
    Bolotin E et al (2004) QNoC: QoS architecture and design process for Network on Chip. In Elsevier JSA 50(2–3):105–128 Feb 2004Google Scholar
  6. 6.
    Dally WJ (1992) Virtual-channel flow control. In IEEE TPDS 3(2):94–205 March 1992Google Scholar
  7. 7.
    Dally WJ, Towles B (2001) Route packets, not wires: on-Chip interconnection networks. In: DAC, pp 684–689 June 2001Google Scholar
  8. 8.
    Dally WJ, Towles B (2004) Principles and practices of interconnection networks. Morgan kaufmann publishers Inc. ISBN 9780122007514Google Scholar
  9. 9.
    Daneshtalab M et al (2006) NoC hot spot minimization using antNet dynamic routing algorithm. In: ASAP, pp 33–38 Dec 2006Google Scholar
  10. 10.
    Duato J et al (2005) A new scalable and cost-effective congestion management strategy for lossless multistage interconnection networks. In: HPCA, pp 108–119 Feb 2005Google Scholar
  11. 11.
    Goossens K et al (2005) AEtherealn Network on chip: concepts, architectures, and implementations. In: IEEE DTC, pp 414–421 Sept-Oct 2005Google Scholar
  12. 12.
    Hashem S et al (1999) A novel approach for training neural networks for long-term prediction. In IJCNN 3:1594–1599 July 1999Google Scholar
  13. 13.
    Hashemi KS et al (1991) On the number of training points needed for adequate training of feedforward neural networks. In: IFNNPS, pp 232–236 July 1991Google Scholar
  14. 14.
    Ho WS, Eager DL (1989) A novel strategy for controlling hot-spot congestion. In: IEEE ICPP, pp 14–18Google Scholar
  15. 15.
    Gaughan PT, Yalamanchili S (1993) Adaptive routing protocols for hypercube interconnection networks. In IEEE Computer 26(5):12–23 May 1993Google Scholar
  16. 16.
    Jain AK et al (1996) Artificial neural networks: a tutorial. In IEEE Computer 29(29):31–44 March 1996Google Scholar
  17. 17.
    Maqsood I et al (2004) An ensemble of neural networks for weather forecasting. In Neural Computing & Applications 13(2):112–122 June 2004Google Scholar
  18. 18.
    McCoy A et al (2007) Multistep-Ahead Neural-Network Predictors for Network Traffic Reduction in Distributed Interactive Applications. In: ACM TOMACS 17(4):1–30MathSciNetGoogle Scholar
  19. 19.
    Nilsson E et al (2003) Load Distribution with the Proximity Congestion Awareness in a Network on Chip. In: DATE, pp 11126–11127 March 2003Google Scholar
  20. 20.
    Peh L-S, Dally WJ (2000) Flit-Reservation Flow Control. In: HPCA, pp 73–84 Jan 2000Google Scholar
  21. 21.
    Pande PP et al (2005) Performance evaluation and design trade-offs for Network-on-Chip interconnect architectures. In IEEE TPDS 54(8):1025–1040 Aug 2005Google Scholar
  22. 22.
    Sarbazi-Azad H et al (2001) An analytical model of fully-adaptive wormhole-routed k-ary n-cubes in the presence of hot spot traffic. In IEEE TOC 50(7):623–634 July 2001MathSciNetGoogle Scholar
  23. 23.
    Steven G et al (2001) Dynamic branch prediction using neural networks. In: DSD, pp 178–185 Sept 2001Google Scholar
  24. 24.
    Taylor MB et al (2004) Evaluation of the raw microprocessor: an exposed-wire-delay architecture for ILP and streams. In: ISCA, pp 2–13Google Scholar
  25. 25.
    Teixeira A et al (2000) A multi-objective optimization approach for training artificial neural networks. In: IEEE SBRN, pp 168–172 Jan 2000Google Scholar
  26. 26.
    Vangal S et al (2007) An 80-tile 1.28TFLOPS Network-on-Chip in 65 nm CMOS. In: ISSCC, pp 98–99 Feb 2007Google Scholar
  27. 27.
    Walter I et al (2007) Access regulation to hot-modules in wormhole NoCs. In: NoCs, pp 137–148 May 2007Google Scholar

Copyright information

© Springer Science+Business Media B.V. 2011

Authors and Affiliations

  • Elena Kakoulli
    • 1
  • Vassos Soteriou
    • 1
  • Theocharis Theocharides
    • 2
  1. 1.Department of Electrical Engineering and Information TechnologyCyprus University of TechnologyLemesosCyprus
  2. 2.Department of Electrical and Computer Engineering KIOS Research Center for Intelligent Systems and NetworksUniversity of CyprusNicosiaCyprus

Personalised recommendations