Path-Diverse Inorder Routing

We present Path-Diverse In-Order Routing (PDIOR), an oblivious routing method which guarantees network-level inorder delivery for multi-path routing. Based on Exclusive Dynamic Virtual Channel Allocation (EDVCA), which allows single-path efficient inorder delivery with dynamic virtual channel allocation, PDIOR extends the same guarantees to routing schemes where each flow may be routed via more than one path. As with EDVCA, PDIOR avoids the overheads inherent in reordering packets at the destination core, and requires only minor, inexpensive changes to traditional oblivious router architectures: for example, an implementation of PDIOR on 8× 8 mesh network with 4 VCs per port requires 492 bytes of memory per node, while inorder packet delivery in a comparable conventional network may requires tens to hundreds of kilobytes of reorder buffer memory at each node.

[1]  Ge-Ming Chiu,et al.  The Odd-Even Turn Model for Adaptive Routing , 2000, IEEE Trans. Parallel Distributed Syst..

[2]  Idit Keidar,et al.  NoC-Based FPGA: Architecture and Routing , 2007, First International Symposium on Networks-on-Chip (NOCS'07).

[3]  G. Edward Suh,et al.  Static virtual channel allocation in oblivious routing , 2009, 2009 3rd ACM/IEEE International Symposium on Networks-on-Chip.

[4]  David A. Patterson,et al.  Computer Architecture: A Quantitative Approach , 1969 .

[5]  S. Lennart Johnsson,et al.  ROMM Routing: A Class of Efficient Minimal Routing Algorithms , 1994, PCRCW.

[6]  Lionel M. Ni,et al.  The Turn Model for Adaptive Routing , 1992, [1992] Proceedings the 19th Annual International Symposium on Computer Architecture.

[7]  William Thies,et al.  StreamIt: A Language for Streaming Applications , 2002, CC.

[8]  Leslie G. Valiant,et al.  Universal schemes for parallel communication , 1981, STOC '81.

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

[10]  G. Edward Suh,et al.  Application-aware deadlock-free oblivious routing , 2009, ISCA '09.

[11]  Luca Benini,et al.  A multi-path routing strategy with guaranteed in-order packet delivery and fault-tolerance for networks on chip , 2006, 2006 43rd ACM/IEEE Design Automation Conference.

[12]  Arnab Banerjee,et al.  Flow-aware allocation for on-chip networks , 2009, 2009 3rd ACM/IEEE International Symposium on Networks-on-Chip.

[13]  Akif Ali,et al.  Near-optimal worst-case throughput routing for two-dimensional mesh networks , 2005, 32nd International Symposium on Computer Architecture (ISCA'05).

[14]  Natalie D. Enright Jerger,et al.  Virtual tree coherence: Leveraging regions and in-network multicast trees for scalable cache coherence , 2008, 2008 41st IEEE/ACM International Symposium on Microarchitecture.

[15]  Srinivas Devadas,et al.  Guaranteed in-order packet delivery using Exclusive Dynamic Virtual Channel Allocation , 2009 .

[16]  S. Lennart Johnsson,et al.  ROMM routing on mesh and torus networks , 1995, SPAA '95.