Design and implementation of a high-performance interconnection network: TH-Net

We provide a high performance interconnection network that can provide high bandwidth and low latency communication on a custom gigabit crossbar network. This implementation consists of the high-speed switch TH-Switch, the VIA-supported NIC TH-NIC and protocol software TH-VIA complied with VIA. TH-Switch uses circuit-switching technique instead of packet-switching to reduce the communication latency. The design and implementation of TH-Net's communication mechanism is based on the VIA specification that the data transferring can be performed directly at the user application level, bypassing the operating system kernel, to reduce the communication overhead. The design and implementation details are discussed.

[1]  Cezary Dubnicki,et al.  VMMC-2 : Efficient Support for Reliable, Connection-Oriented Communication , 1997 .

[2]  Seth Copen Goldstein,et al.  Active messages: a mechanism for integrating communication and computation , 1998, ISCA '98.

[3]  Seth Copen Goldstein,et al.  Active Messages: A Mechanism for Integrated Communication and Computation , 1992, [1992] Proceedings the 19th Annual International Symposium on Computer Architecture.

[4]  Thorsten von Eicken,et al.  U-Net: a user-level network interface for parallel and distributed computing , 1995, SOSP.

[5]  Dhabaleswar K. Panda,et al.  Efficient virtual interface architecture (VIA) support for the IBM SP switch-connected NT clusters , 2000, Proceedings 14th International Parallel and Distributed Processing Symposium. IPDPS 2000.

[6]  David E. Culler,et al.  An Implementation and Analysis of the Virtual Interface Architecture , 1998, Proceedings of the IEEE/ACM SC98 Conference.

[7]  Joseph Pasquale,et al.  Profiling and reducing processing overheads in TCP/IP , 1996, TNET.

[8]  Kai Li,et al.  Virtual-Memory-Mapped Network Interfaces , 1995, IEEE Micro.

[9]  Bernard Tourancheau,et al.  BIP: A New Protocol Designed for High Performance Networking on Myrinet , 1998, IPPS/SPDP Workshops.

[10]  Scott Pakin,et al.  High Performance Messaging on Workstations: Illinois Fast Messages (FM) for Myrinet , 1995, Proceedings of the IEEE/ACM SC95 Conference.

[11]  Robert J. Harrison,et al.  Performance and experience with LAPI-a new high-performance communication library for the IBM RS/6000 SP , 1998, Proceedings of the First Merged International Parallel Processing Symposium and Symposium on Parallel and Distributed Processing.