A look at application performance sensitivity to the bandwidth and latency of InfiniBand networks

This work explores the expected performance of three applications on a High Performance Computing cluster interconnected using Infiniband. In particular, the expected performance across a range of configurations is analyzed notably Infiniband 4x, 8x and 12x representing link-speeds of 10Gb/s, 20Gb/s, and 30Gb/s respectively as well as near-neighbor MPI message latencies of 4 s and

• Type: Conference
• Year: 2006
• Where: IPPS
• Authors: Darren J. Kerbyson
• Keywords: Distributed And Parallel Computing, Expected Performance, IPPS 2006, MPI Message Latencies, Performance Computing Cluster