A Compressed 3D-Torus architecture for scalable router

With the explosion increasing data in Internet, high performance scalable router has become a hot spot research area both in academe and industry recently. The underlying topology is a very important fold for scalable router. In this paper, we propose a new underlying topology of the scalable router, and name it Compressed 3D-Torus (C3D-Torus). It's more like the 3D-Torus, which is a famous architecture and has been widely studied and used in high performance computer and scalable router before. With theory analysis and simulation, we find it has better performance than 3D-Torus in latency and scalability. Furthermore, we provide an internal routing algorithm which can effectively use multiple links existing simultaneously between neighbor routing nodes in C3D-Torus. We name it Multiple Link Dimension Order Routing (MLDOR) and design it's steps in this paper.

[1]  Cheng-Shang Chang,et al.  Load balanced Birkhoff-von Neumann switches, part II: multi-stage buffering , 2002, Comput. Commun..

[2]  Jianping Wu,et al.  Designing Scalable Routers with a New Switching Architecture , 2005, Joint International Conference on Autonomic and Autonomous Systems and International Conference on Networking and Services - (icas-isns'05).

[3]  V. Benes.  Optimal rearrangeable multistage connecting networks , 1964 .

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

[5]  Xiao-Ping Zhang.  Scalable Router: Scalable Router , 2008 .

[6]  Bill Lin,et al.  The Concurrent Matching Switch Architecture , 2006, IEEE/ACM Transactions on Networking.

[7]  Theodore R. Bashkow,et al.  A large scale, homogeneous, fully distributed parallel machine, I , 1977, ISCA '77.

[8]  Steven L. Scott,et al.  The Cray T3E Network: Adaptive Routing in a High Performance 3D Torus , 1996 .

[9]  Cheng-Shang Chang,et al.  Load balanced Birkhoff-von Neumann switches, part I: one-stage buffering , 2002, Computer Communications.

[10]  Zhenhua Liu,et al.  An Asymptotically Minimal Node-Degree Topology for Load-Balanced Architectures , 2008, IEEE GLOBECOM 2008 - 2008 IEEE Global Telecommunications Conference.

[11]  Nick McKeown,et al.  Scaling internet routers using optics , 2003, SIGCOMM '03.

[12]  William J. Dally,et al.  Express Cubes: Improving the Performance of k-Ary n-Cube Interconnection Networks , 1989, IEEE Trans. Computers.

[13]  William J. Dally,et al.  Performance Analysis of k-Ary n-Cube Interconnection Networks , 1987, IEEE Trans. Computers.

[14]  Jay Cheng,et al.  Using Banyan Networks for Load-Balanced Switches with Incremental Update , 2010, 2010 IEEE International Conference on Communications.

[15]  R. Srikant,et al.  Padded Frames: A Novel Algorithm for Stable Scheduling in Load-Balanced Switches , 2008, 2006 40th Annual Conference on Information Sciences and Systems.

[16]  Charles Clos,et al.  A study of non-blocking switching networks , 1953 .

[17]  Zhenhua Liu,et al.  A performance analysis framework for routing lookup in scalable routers , 2009, 2009 International Conference on Information Networking.

[18]  Mark W. Johnson,et al.  Evolving standards for IT service management , 2007, IBM Syst. J..