A preliminary analysis of the InfiniPath and XD1 network interfaces
Authors:
- 1 January 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15302075,p. 8 pp.-275
Abstract
Two recently delivered systems have begun a new trend in cluster interconnects. Both the InfiniPath network from PathScale, Inc., and the rapidarray fabric in the XDI system from Cray, Inc., leverage commodity network fabrics while customizing the network interface in an attempt to add value specifically for the high performance computing (HPC) cluster market. Both network interfaces are compatible with standard InfiniBand (IB) switches, but neither use the traditional programming interfaces to support MPI. Another fundamental difference between these networks and other modern network adapters is that much of the processing needed for the network protocol stack is performed on the host processor(s) rather than by the network interface itself. This approach stands in stark contrast to the current direction of most high-performance networking activities, which is to offload as much protocol processing as possible to the network interface. In this paper, we provide an initial performance comparison of the two partially custom networks (PathScale's InfiniPath and Cray's XDI) with a more commodity network (standard IB) and a more custom network (Quadrics Elan4). Our evaluation includes several micro-benchmark results as well as some initial application performance dataKeywords
This publication has 14 references indexed in Scilit:
- PathScale InfiniPath: A First Look2006
- Efficient Barrier and Allreduce on Infiniband clusters using multicast and adaptive algorithms2005
- A Comparison of Three MPI Implementations for Red StormLecture Notes in Computer Science, 2005
- An analysis of the impact of MPI overlap and independent progress2004
- Efficient collective operations using remote memory operations on VIA-based clusters2004
- Microbenchmark performance comparison of high-speed cluster interconnectsIEEE Micro, 2004
- Efficient and scalable all-to-all personalized exchange for InfiniBand-based clusters2004
- Performance Comparison of MPI Implementations over InfiniBand, Myrinet and Quadrics2003
- The Quadrics network: high-performance clustering technologyIEEE Micro, 2002
- Effects of communication latency, overhead, and bandwidth in a cluster architectureACM SIGARCH Computer Architecture News, 1997