Kang Ji-chang

MPICH for FC-based Token-Routing Switch-Network

In order to replace the SCI network by our high cost-effective FC-based token-routing switch-network in the high performance computing fields, the most commonly used and freely available MPICH is extended. FC-VIAI, a low-level communication library based on our user-level communication protocol FC-VIA, is appended into the MPICH. So a new extended MPICH, named FC-VIA-MPI, comes into being. The experiment results show the FC-VIA-MPI on FC-based token-routing switch-network has achieved better performance with maximum bandwidth of 989 Mb/s and minimum latency of 1.46 mus.

On the communication structures of hyper-ring and hypercube multicomputers

The well-known torus and its variants, which we call hyper-rings, as well as hypercube architectures are further studied and evaluated as interconnection networks for multicomputers. Comparisons are made among hyper-rings and between hyper-ring and hypercube networks under different communication patterns. It is concluded that although it is believed that a hypercube is generally superior to hyper-rings in performance, this is not always the case, particularly for locally constrained applications, where communications occur mostly among neighboring nodes.The well-known torus and its variants, which we call hyper-rings, as well as hypercube architectures are further studied and evaluated as interconnection networks for multicomputers. Comparisons are made among hyper-rings and between hyper-ring and hypercube networks under different communication patterns. It is concluded that although it is believed that a hypercube is generally superior to hyper-rings in performance, this is not always the case, particularly for locally constrained applications, where communications occur mostly among neighboring nodes.