Replacing global wires with an on-chip network: a power analysis

ABSTRACT

This paper explores the power implications of replacing global chip wires with an on-chip network. We optimize network links by varying repeater spacing, link pipelining, and voltage scaling, to significantly reduce the energy to send a bit across chip. We develop an analytic model of large chip designs with an on-chip two-dimensional mesh network and estimate the power savings possible in a 70 nm process for two different design points: a circuit-switched ASIC or FPGA design, and a dynamic packet-switched tiled architecture. For circuit-switched networks, achievable power savings are 35--50% for a mesh with 1 mm links. The packet switched designs use multiplexing and signal encoding to reduce the number of link wires required, but the router overhead limits peak wire power savings to around 20% with optimal tile sizes of around 2 mm

REFERENCES

• 1: H. B. Bakoglu. Circuits, Interconnections, and Packaging for VLSI. Addision-Wesley, 1990.
• 2: K. Banerjee and A. Mehrotra. Power dissipation issues in interconnect performance optimization for sub-180 nm designs. In Symposium on VLSI circuits, pages 12--15, June 2002.
• 3: A. Chandrakasan et al. Low-power CMOS digital design. IEEE JSSC, 27(4):473--484, Apr. 1992.
• 4: Xuning Chen , Li-Shiuan Peh, Leakage power modeling and optimization in interconnection networks, Proceedings of the 2003 international symposium on Low power electronics and design, August 25-27, 2003, Seoul, Korea [doi>10.1145/871506.871531]
• 5: Philip Christie , Dirk Stroobandt, The interpretation and application of Rent's rule, IEEE Transactions on Very Large Scale Integration (VLSI) Systems, v.8 n.6, p.639-648, Dec. 2000 [doi>10.1109/92.902258]
• 6: Pasquale Cocchini, Concurrent flip-flop and repeater insertion for high performance integrated circuits, Proceedings of the 2002 IEEE/ACM international conference on Computer-aided design, p.268-273, November 10-14, 2002, San Jose, California [doi>10.1145/774572.774612]
• 7: William J. Dally , Brian Towles, Route packets, net wires: on-chip inteconnectoin networks, Proceedings of the 38th conference on Design automation, p.684-689, June 2001, Las Vegas, Nevada, United States [doi>10.1145/378239.379048]
• 8: Device Group at UC Berkeley. Predictive technology model. Technical report, UC Berkeley, 2001. http://www-device.eecs.berkely.edu/ptm/.
• 9: Noel Eisley , Li-Shiuan Peh, High-level power analysis for on-chip networks, Proceedings of the 2004 international conference on Compilers, architecture, and synthesis for embedded systems, September 22-25, 2004, Washington DC, USA [doi>10.1145/1023833.1023849]
• 10: International Technology Roadmap for Semiconductors. 2004 update. Technical report, ITRS, 2004.
• 11: Pallav Gupta , Lin Zhong , Niraj K. Jha, A High-level Interconnect Power Model for Design Space Exploration, Proceedings of the 2003 IEEE/ACM international conference on Computer-aided design, p.551, November 09-13, 2003 [doi>10.1109/ICCAD.2003.10]
• 12: Seongmoo Heo , Krste AsanoviC, Power-optimal pipelining in deep submicron technology, Proceedings of the 2004 international symposium on Low power electronics and design, August 09-11, 2004, Newport Beach, California, USA [doi>10.1145/1013235.1013291]
• 13: R. Ho et al. The future of wires. Proceedings of the IEEE, 89(4):490--504, Apr. 2001.
• 14: Pawan Kapur , Gaurav Chandra , Krishna C. Saraswat, Power estimation in global interconnects and its reduction using a novel repeater optimization methodology, Proceedings of the 39th conference on Design automation, June 10-14, 2002, New Orleans, Louisiana, USA [doi>10.1145/513918.514035]
• 15: Weiping Liao , Lei He, Full-Chip Interconnect Power Estimation and Simulation Considering Concurrent Repeater and Flip-Flop Insertion, Proceedings of the 2003 IEEE/ACM international conference on Computer-aided design, p.574, November 09-13, 2003 [doi>10.1109/ICCAD.2003.75]
• 16: Robert Mullins , Andrew West , Simon Moore, Low-Latency Virtual-Channel Routers for On-Chip Networks, Proceedings of the 31st annual international symposium on Computer architecture, p.188, June 19-23, 2004, München, Germany
• 17: M. Sgroi , M. Sheets , A. Mihal , K. Keutzer , S. Malik , J. Rabaey , A. Sangiovanni-Vencentelli, Addressing the system-on-a-chip interconnect woes through communication-based design, Proceedings of the 38th conference on Design automation, p.667-672, June 2001, Las Vegas, Nevada, United States [doi>10.1145/378239.379045]
• 18: Hang-Sheng Wang , Xinping Zhu , Li-Shiuan Peh , Sharad Malik, Orion: a power-performance simulator for interconnection networks, Proceedings of the 35th annual ACM/IEEE international symposium on Microarchitecture, November 18-22, 2002, Istanbul, Turkey
• 19: Hang-Sheng Wang , Li-Shiuan Peh , Sharad Malik, A Power Model for Routers: Modeling Alpha 21364 and InfiniBand Routers, IEEE Micro, v.23 n.1, p.26-35, January 2003 [doi>10.1109/MM.2003.1179895]
• 20: S. Yang et al. Scaling and integration of high performance interconnects. In MRS Symposium on Advanced Interconnect, Apr. 1998.
• 21: M. Yazdani et al. Microprocessor pin predicting. IEEE Circuits and Devices Magazine, 13(2):28--31, Mar. 1997.
• 22: Terry Tao Ye , Giovanni De Micheli , Luca Benini, Analysis of power consumption on switch fabrics in network routers, Proceedings of the 39th conference on Design automation, June 10-14, 2002, New Orleans, Louisiana, USA [doi>10.1145/513918.514051]

INDEX TERMS

Primary Classification:
  B. Hardware
  B.7 INTEGRATED CIRCUITS
      B.7.1 Types and Design Styles
          Subjects: Advanced technologies

Additional Classification:
  B. Hardware
  B.7 INTEGRATED CIRCUITS
      B.7.1 Types and Design Styles
          Subjects: VLSI (very large scale integration); Microprocessors and microcomputers

General Terms:
Design, Performance, Theory

Keywords:
on-chip network power model, pipelining, router, tile size, tiled architecture, wire power model