Environments and tools for parallel scientific computing
Environments and tools for parallel scientific computing
Network-Based Multicomputers: A Practical Supercomputer Architecture
IEEE Transactions on Parallel and Distributed Systems
File-Access Characteristics of Parallel Scientific Workloads
IEEE Transactions on Parallel and Distributed Systems
Parallel programming with MPI
Special issue on workstation clusters and network-based computing: guest editors' introduction
Journal of Parallel and Distributed Computing
In search of clusters (2nd ed.)
In search of clusters (2nd ed.)
Parallel labeling of three-dimensional clusters on networks of workstations
Journal of Parallel and Distributed Computing
On implementing MPI-IO portably and with high performance
Proceedings of the sixth workshop on I/O in parallel and distributed systems
| Hi-index | 0.00 |
This paper describes a software environment to facilitate the use of cluster computing for Numerical Weather Prediction (NWP) models. It presents an empirical study based on the NCAR/Penn State MM5 model and another mesoscale model developed at the Pacific Northwest National Laboratory (PNNL). Without appropriate tools, it is tedious and difficult to work with the legacy codes for NWP models. We have used a parallelization tool, a visualization tool, a web-based monitoring tool, a performance analysis tool, a runtime communication li-brary built on top of MPI, and a parallel I/O library. The experimental results provide a performance comparison NWP models on a PC cluster with two different types of networks, a SGI machine, and an IBM SP-2.