Evaluation of Crossbar Architectures for Deadlock Recovery Routers

The performance of interconnection networks is significantly affected by router speed and routing adaptivity, which can be competing factors. To achieve a high-speed, true-fully-adaptive router design, this paper explores the exploitation of dynamic routing behavior identified as routing locality. When routing locality is exploited, it enables the internal crossbar of a router to be partitioned into smaller and faster units without sacrificing true-fully-adaptive routing capabilities. Extensive evaluation of partitioned crossbar designs which exploit routing locality shows that the increased adaptivity offered by deadlock recovery-based routing algorithms can be implemented in routers without sacrificing router speed. The partitioned crossbar designs reduce average message latency by up to 65% and increase maximum network throughput by up to 51%.

[1]  Andrew A. Chien,et al.  Compressionless routing: a framework for adaptive and fault-tolerant routing , 1994, ISCA '94.

[2]  Kyusun Choi,et al.  VLSI implementation of a 256*256 crossbar interconnection network , 1992, Proceedings Sixth International Parallel Processing Symposium.

[3]  Timothy Mark Pinkston,et al.  Characterization of deadlocks in interconnection networks , 1997, Proceedings 11th International Parallel Processing Symposium.

[4]  Carmen Carrión,et al.  A flow control mechanism to avoid message deadlock in k-ary n-cube networks , 1997, Proceedings Fourth International Conference on High-Performance Computing.

[5]  José Duato.  A Necessary and Sufficient Condition for Deadlock-Free Adaptive Routing in Wormhole Networks , 1995, IEEE Trans. Parallel Distributed Syst..

[6]  Charles M. Flaig.  VLSI Mesh Routing Systems , 1987 .

[7]  Timothy Mark Pinkston,et al.  Crossbar analysis for optimal deadlock recovery router architecture , 1997, Proceedings 11th International Parallel Processing Symposium.

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

[9]  Andrew A. Chien,et al.  Planar-adaptive routing: low-cost adaptive networks for multiprocessors , 1992, ISCA '92.

[10]  A. A. Chein,et al.  A cost and speed model for k-ary n-cube wormhole routers , 1998 .

[11]  T.M. Pinkston,et al.  On Deadlocks In Interconnection Networks , 1997, Conference Proceedings. The 24th Annual International Symposium on Computer Architecture.

[12]  Mike Galles.  Spider: a high-speed network interconnect , 1997, IEEE Micro.

[13]  Lionel M. Ni,et al.  The turn model for adaptive routing , 1992, ISCA '92.

[14]  José Duato,et al.  A New Theory of Deadlock-Free Adaptive Routing in Wormhole Networks , 1993, IEEE Trans. Parallel Distributed Syst..

[15]  Steven L. Scott,et al.  Optimized Routing in the Cray T3D , 1994, PCRCW.

[16]  Timothy Mark Pinkston,et al.  DISHA: a deadlock recovery scheme for fully adaptive routing , 1995, Proceedings of 9th International Parallel Processing Symposium.

[17]  William J. Dally,et al.  Deadlock-Free Message Routing in Multiprocessor Interconnection Networks , 1987, IEEE Transactions on Computers.