A hybrid SoC interconnect with dynamic TDMA-based transaction-less buses and on-chip networks

The two dominant architectural choices for implementing efficient communication fabrics for SoC's have been transaction-based buses and packet-based networks-on-chip (NoC). Both implementations have some inherent disadvantages - the former resulting from poor scalability and the transactional character of their operation, and the latter from inconsistent access times and deterioration of performance at high injection rates. In this paper, we propose a transaction-less, time-division-based bus architecture, which dynamically allocates timeslots on-the-fly - the dTDMA bus. This architecture addresses the contention issues of current bus architectures, while avoiding the multi-hop overhead of NoC's. It is compared to traditional bus architectures and NoC's and shown to outperform both for configurations with fewer than 10 PE's. In order to exploit the advantages of the dTDMA bus for smaller configurations, and the scalability of NoC's, we propose a new hybrid SoC interconnect combining the two, showing significant improvement in both latency and power consumption.

[1]  Henry Hoffmann,et al.  The Raw Microprocessor: A Computational Fabric for Software Circuits and General-Purpose Programs , 2002, IEEE Micro.

[2]  J. Komo,et al.  Modified Kasami sequences for CDMA , 1990, [1990] Proceedings. The Twenty-Second Southeastern Symposium on System Theory.

[3]  W. Dally,et al.  Route packets, not wires: on-chip interconnection networks , 2001, Proceedings of the 38th Design Automation Conference (IEEE Cat. No.01CH37232).

[4]  Cheng-Kok Koh,et al.  SAMBA-Bus: A High Performance Bus Architecture for System-on-Chips , 2003, ICCAD.

[5]  A. Raghunathan,et al.  LOTTERYBUS: a new high-performance communication architecture for system-on-chip designs , 2001, Proceedings of the 38th Design Automation Conference (IEEE Cat. No.01CH37232).

[6]  Chita R. Das,et al.  A low latency router supporting adaptivity for on-chip interconnects , 2005, Proceedings. 42nd Design Automation Conference, 2005..

[7]  Ulrich Rückert,et al.  A scalable parallel SoC architecture for network processors , 2005, IEEE Computer Society Annual Symposium on VLSI: New Frontiers in VLSI Design (ISVLSI'05).

[8]  Narayanan Vijaykrishnan,et al.  Thermal-aware IP virtualization and placement for networks-on-chip architecture , 2004, IEEE International Conference on Computer Design: VLSI in Computers and Processors, 2004. ICCD 2004. Proceedings..

[9]  Luca Benini,et al.  Networks on Chips : A New SoC Paradigm , 2022 .

[10]  T. F. Chen,et al.  Segmented bus design for low-power systems , 1999, IEEE Trans. Very Large Scale Integr. Syst..

[11]  I. Verbauwhede,et al.  CT-bus: a heterogeneous CDMA/TDMA bus for future SOC , 2004, Conference Record of the Thirty-Eighth Asilomar Conference on Signals, Systems and Computers, 2004..

[12]  Cheng-Kok Koh,et al.  SAMBA-Bus: A High Performance Bus Architecture for System-on-Chips , 2007, IEEE Trans. Very Large Scale Integr. Syst..

[13]  Chang Yong Kang,et al.  CDMA as a multiprocessor interconnect strategy , 2001, Conference Record of Thirty-Fifth Asilomar Conference on Signals, Systems and Computers (Cat.No.01CH37256).

[14]  William J. Dally,et al.  Principles and Practices of Interconnection Networks , 2004 .

[15]  T. Higgins Book reviewSystems engineering handbook: edited by R. E. Machol, W. P. Tanner, Jr., and S. N. Alexander. 1054 pages, diagrams, illustr., New York, McGraw-Hill Book Co., 1965. Price, $29.50 , 1966 .