Verifying large-scale system performance during installation using modelling
High performance scientific and engineering computing
PARSE: A Tool for Parallel Application Run Time Sensitivity Evaluation
ICPADS '06 Proceedings of the 12th International Conference on Parallel and Distributed Systems - Volume 1
Towards 100 gbit/s ethernet: multicore-based parallel communication protocol design
Proceedings of the 23rd international conference on Supercomputing
Employing transport layer multi-railing in cluster networks
Journal of Parallel and Distributed Computing
Efficient RDMA-based multi-port collectives on multi-rail QsNetII clusters
IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
A network performance sensitivity metric for parallel applications
International Journal of High Performance Computing and Networking
HiPC'05 Proceedings of the 12th international conference on High Performance Computing
A network performance sensitivity metric for parallel applications
ISPA'07 Proceedings of the 5th international conference on Parallel and Distributed Processing and Applications
| Hi-index | 0.00 |
Using multiple independent networks (also known as rails) is an emerging technique to overcome bandwidth limitations and enhance fault tolerance of current high-performance clusters. We present an extensive experimental comparison of the behavior of various allocation schemes in terms of bandwidth and latency. We show that striping messages over multiple rails can substantially reduce network latency, depending on average message size, network load, and allocation scheme. The compared methods include a basic round-robin rail allocation, a local-dynamic allocation based on local knowledge, and a dynamic rail allocation that reserves both end-points of a message before sending it. The last method is shown to perform better than the others at higher loads: up to 49% better than local-knowledge allocation and 37% better than the round-robin allocation. This allocation scheme also shows lower latency and it saturates on higher loads (for messages large enough). Most importantly, this proposed allocation scheme scales well with the number of rails and message sizes. In addition we propose a hybrid algorithm that combines the benefits of the local-dynamic for short messages with those of the dynamic algorithm for large messages.