Congestion control for high bandwidth-delay product networks
Proceedings of the 2002 conference on Applications, technologies, architectures, and protocols for computer communications
Scalable TCP: improving performance in highspeed wide area networks
ACM SIGCOMM Computer Communication Review
High Performance TCP/IP Networking
High Performance TCP/IP Networking
CHEETAH: circuit-switched high-speed end-to-end transport architecture testbed
IEEE Communications Magazine
Ultrascience net: network testbed for large-scale science applications
IEEE Communications Magazine
High Performance Data Transfer in Grid Environment Using GridFTP over InfiniBand
CCGRID '10 Proceedings of the 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing
A long-distance InfiniBand interconnection between two clusters in production use
State of the Practice Reports
Operating two InfiniBand grid clusters over 28 km distance
International Journal of Grid and Utility Computing
Performance analysis and prediction for distributed homogeneous clusters
Computer Science - Research and Development
Hi-index | 0.00 |
High-performance file systems that span across supercomputers and storage systems are becoming increasingly important, as these resources are being dispersed both within the enterprise and across the wide-area. Infiniband transport has been proposed as a candidate to provide high-performance access over interconnects primarily within an enterprise. In this paper, we present an experimental study of Infiniband transport over ORNL infrastructure of Cray XT3 supercomputer and cluster storage system for local-area and DOE UltraScience Network (USN) for wide-area. Our results indicate that Infiniband throughput of 7.6Gbps (4x) scales to 8,000 miles with 5% degradation with no customization over 10Gbps wide-area connections. We conducted preliminary performance tests of Lustre file system over these connections, which identified potential parameter and modules optimizations.